[Objective] The dynamic prediction of carbon emission from cold chain distribution is an important basis for the accurate assessment of carbon emission and its green credit grade. Facing the fact that the carbon emission of vehicles is affected by multiple factors, such as road condition information, driving characteristics, refrigeration parameters, etc., a dynamic prediction model of carbon emission was proposed from refrigerated vehicles that integrates multi-source information. [Methods] The backbone feature extraction network, neck feature fusion network and loss function of YOLOv8s was firstly improved. The full-dimensional dynamic convolution was introduced into the backbone feature extraction network, and the multidimensional attention mechanism was introduced to capture the contextual key information to improve the model feature extraction capability. A progressive feature pyramid network was introduced into the feature extraction network, which reduced the loss of key information by fusing features layer by layer and improved the feature fusion efficiency. The road condition information recognition model based on improved YOLOv8s was constructed to characterize the road condition information in terms of the number of road vehicles and the percentage of pixel area. Pearson's correlation coefficient was used to compare and analyze the correlation between carbon emissions of refrigerated vehicles and different influencing factors, and to verify the necessity and criticality of the selection of input parameters of the carbon emission prediction model. Then the iTransformer temporal prediction model was improved, and the external attention mechanism was introduced to enhance the feature extraction ability of iTransformer model and reduce the computational complexity. The dynamic prediction model of carbon emission of refrigerated vehicles based on the improved iTransformer was constructed by taking the road condition information, driving characteristics (speed, acceleration), cargo weight, and refrigeration parameters (temperature, power) as inputs. Finally, the model was compared and analyzed with other models to verify the robustness of the road condition information and the prediction accuracy of the vehicle carbon emission dynamic prediction model, respectively. [Results and Discussions] The results of correlation analysis showed that the vehicle driving parameters were the main factor affecting the intensity of vehicle carbon emissions, with a correlation of 0.841. The second factor was cargo weight, with a correlation of 0.807, which had a strong positive correlation. Compared with the vehicle refrigeration parameters, the road condition information had a stronger correlation between vehicle carbon emissions, the correlation between refrigeration parameters and the vehicle carbon emissions impact factor were above 0.67. In order to further ensure the accuracy of the vehicle carbon emissions prediction model, The paper was selected as the input parameters for the carbon emissions prediction model. The improved YOLOv8s road information recognition model achieved 98.1%, 95.5%, and 98.4% in precision, recall, and average recognition accuracy, which were 1.2%, 3.7%, and 0.2% higher than YOLOv8s, respectively, with the number of parameters and the amount of computation being reduced by 12.5% and 31.4%, and the speed of detection being increased by 5.4%. This was due to the cross-dimensional feature learning through full-dimensional dynamic convolution, which fully captured the key information and improved the feature extraction capability of the model, and through the progressive feature pyramid network after fusing the information between different classes through gradual step-by-step fusion, which fully retained the important feature information and improved the recognition accuracy of the model. The predictive performance of the improved iTransformer carbon emission prediction model was better than other time series prediction models, and its prediction curve was closest to the real carbon emission curve with the best fitting effect. The introduction of the external attention mechanism significantly improved the prediction accuracy, and its MSE, MAE, RMSE and R2 were 0.026 1 %VOL, 0.079 1 %VOL, 0.161 5 %VOL and 0.940 0, respectively, which were 0.4%, 15.3%, 8.7% and 1.3% lower, respectively, when compared with iTransformer. As the degree of road congestion increased, the prediction accuracy of the constructed carbon emission prediction model increased. [Conclusions] The carbon emission prediction model for cold chain distribution under multi-source information fusion proposed in this study can realize accurate prediction of carbon emission from refrigerated vehicles, provide theoretical basis for rationally formulating carbon emission reduction strategies and promoting the development of low-carbon cold chain distribution.
[Objective] Rice plays a crucial role in daily diet. The rice industry involves numerous links, from paddy planting to the consumer's table, and the integrity of the quality control data chain directly affects the credibility of rice quality control and traceability information. The process of rice traceability also faces security issues, such as the leakage of privacy information, which need immediate solutions. Additionally, the previous practice of uploading all information onto the blockchain leads to high storage costs and low system efficiency. To address these problems, this study proposed a differential privacy-enhanced blockchain-based quality control model for rice, providing new ideas and solutions to optimize the traditional quality regulation and traceability system. [Methods] By exploring technologies of blockchain, interplanetary file system (IPFS), and incorporating differential privacy techniques, a blockchain-based quality control model for rice with differential privacy enhancement was constructed. Firstly, the data transmission process was designed to cover the whole industry chain of rice, including cultivation, acquisition, processing, warehousing, and sales. Each module stored the relevant data and a unique number from the previous link, forming a reliable information chain and ensuring the continuity of the data chain for quality control. Secondly, to address the issue of large data volume and low efficiency of blockchain storage, the key quality control data of each link in the rice industry chain was stored in the IPFS. Subsequently, the hash value of the stored data was returned and recorded on the blockchain. Lastly, to enhance the traceability of the quality control model information, the sensitive information in the key quality control data related to the cultivation process was presented to users after undergoing differential privacy processing. Individual data was obfuscated to increase the credibility of the quality control information while also protecting the privacy of farmers' cultivation practices. Based on this model, a differential privacy-enhanced blockchain-based quality control system for rice was designed. [Results and Discussions] The architecture of the differential privacy-enhanced blockchain-based quality control system for rice consisted of the physical layer, transport layer, storage layer, service layer, and application layer. The physical layer included sensor devices and network infrastructure, ensuring data collection from all links of the industry chain. The transport layer handled data transmission and communication, securely uploading collected data to the cloud. The storage layer utilized a combination of traditional databases, IPFS, and blockchain to efficiently store and manage key data on and off the blockchain. The traditional database was used for the management and querying of structured data. IPFS stored the key quality control data in the whole industry chain, while blockchain was employed to store the hash values returned by IPFS. This integrated storage method improved system efficiency, ensured the continuity, reliability, and traceability of quality control data, and provided consumers with reliable information. The service layer was primarily responsible for handling business logic and providing functional services. The implementation of functions in the application layer relied heavily on the design of a series of interfaces within the service layer. Positioned at the top of the system architecture, the application layer was responsible for providing user-centric functionality and interfaces. This encompassed a range of applications such as web applications and mobile applications, aiming to present data and facilitate interactive features to fulfill the requirements of both consumers and businesses. Based on the conducted tests, the average time required for storing data in a single link of the whole industry chain within the system was 1.125 s. The average time consumed for information traceability query was recorded as 0.691 s. Compared to conventional rice quality regulation and traceability systems, the proposed system demonstrated a reduction of 6.64% in the storage time of single-link data and a decrease of 16.44% in the time required to perform information traceability query. [Conclusions] This study proposes a differential privacy-enhanced blockchain-based quality control model for rice. The model ensures the continuity of the quality control data chain by integrating the various links of the whole industry chain of rice. By combining blockchain with IPFS storage, the model addresses the challenges of large data volume and low efficiency of blockchain storage in traditional systems. Furthermore, the model incorporates differential privacy protection to enhance traceability while safeguarding the privacy of individual farmers. This study can provide reference for the design and improvement of rice quality regulation and traceability systems.
[Objective] In the retail market for agricultural products, consumers are increasingly concerned about the safety and health aspects of those products. Traceability of blockchain has emerged as a crucial solution to address these concerns. Essentially, a blockchain functions as a dynamic, distributed, and shared database. When implemented in the agricultural supply chain, it not only improves product transparency to attract more consumers but also raises concerns about consumer privacy disclosure. The level of consumer apprehension regarding privacy will directly influence their choice to purchase agricultural products traced through blockchain-traced. Moreover, retailers' choices to sell blockchain-traced produce are influenced by consumer privacy concerns. By analyzing the impact of blockchain technology on the competitive strategies, pricing, and decision-making among agricultural retailers, they can develop market competition strategies that suit their market conditions to bolster their competitiveness and optimize the agricultural supply chain to maximize overall benefits. [Methods] Based on Nash equilibrium and Stackelberg game theory, a market competition model was developed to analyze the interactions between existing and new agricultural product retailers. The competitive strategies adopted by agricultural product retailers were analyzed under four different options of whether two agricultural retailers sell blockchain agricultural products. It delved into product utility, optimal pricing, demand, and profitability for each retailer under these different scenarios. How consumer privacy concerns impact pricing and profits of two agricultural product retailers and the optimal response strategy choice of another retailer when the competitor made the decision choice first were also analyzed. This analysis aimed to guide agricultural product retailers in making strategic choices that would safeguard their profits and market positions. To address the cooperative game problem of agricultural product retailers in market competition, ensure that retailers could better cooperate in the game, blockchain smart contract technology was used. By encoding the process and outcomes of the Stackelberg game into smart contracts, retailers could input their specific variables and receive tailored strategy recommendations. Uploading game results onto the blockchain network ensured transparency and encouraged cooperative behavior among retailers. By using the characteristics of blockchain, the game results were uploaded to the blockchain network to regulate the cooperative behavior, to ensure the maximization of the overall interests of the supply chain. [Results and Discussions] The research highlighted the significant improvement in agricultural product quality transparency through blockchain traceability technology. However, concerns regarding consumer privacy arising from this traceability could directly impact the pricing, profitability and retailers' decisions to provide blockchain-traceable items. Furthermore, an analysis of the strategic balance between two agricultural product retailers revealed that in situations of low and high product information transparency, both retailers were inclined to simultaneously offer sell traceable products. In such a scenario, blockchain traceability technology enhanced the utility and profitability of retail agricultural products, leading consumers to prefer purchase these traceable products from retailers. In cases where privacy concerns and agricultural product information transparency were both moderate, the initial retailer was more likely to opt for blockchain-based traceable products. This was because consumers had higher trust in the initial retailer, enabling them to bear a higher cost associated with privacy concerns. Conversely, new retailers failed to gain a competitive advantage and eventually exit the market. When consumer privacy concerns exceeded a certain threshold, both competing agricultural retailers discovered that offering blockchain-based traceable products led to a decline in their profits. [Conclusions] When it comes to agricultural product quality and safety, incorporating blockchain technology in traceability significantly improves the transparency of quality-related information for agricultural products. However, it is important to recognize that the application of blockchain for agricultural product traceability is not universally suitable for all agricultural retailers. Retailers must evaluate their unique circumstances and make the most suitable decisions to enhance the effectiveness of agricultural products, drive sales demand, and increase profits. Within the competitive landscape of the agricultural product retail market, nurturing a positive collaborative relationship is essential to maximize mutual benefits and optimize the overall profitability of the agricultural product supply chain.
[Objective] Camellia oleifera is one of the four major woody oil plants in the world. Diseases is a significant factor leading to the decline in quality of Camellia oleifera and the financial loss of farmers. Among these diseases, anthracnose is a common and severe disease in Camellia oleifera forests, directly impacting yields and production rates. Accurate disease assessment can improve the prevention and control efficiency and safeguarding the farmers' profit. In this study, an improved You Only Look at CoefficienTs (YOLACT) based method was proposed to realize automatic and efficient grading of the severity of Camellia oleifera leaf anthracnose. [Methods] High-resolution images of Camellia oleifera anthracnose leaves were collected using a smartphone at the National Camellia oleifera Seed Base of Jiangxi Academy of Forestry, and finally 975 valid images were retained after a rigorous screening process. Five data enhancement means were applied, and a data set of 5 850 images was constructed finally, which was divided into training, validation, and test sets in a ratio of 7:2:1. For model selection, the Camellia-YOLACT model was proposed based on the YOLACT instance segmentation model, and by introducing improvements such as Swin-Transformer, weighted bi-directional feature pyramid network, and HardSwish activation function. The Swin Transformer was utilized for feature extraction in the backbone network part of YOLACT, leveraging the global receptive field and shift window properties of the self-attention mechanism in the Transformer architecture to enhance feature extraction capabilities. Additionally, a weighted bidirectional feature pyramid network was introduced to fuse feature information from different scales to improve the detection ability of the model for objects at different scales, thereby improving the detection accuracy. Furthermore, to increase the the model's robustness against the noise in the input data, the HardSwish activation function with stronger nonlinear capability was adopted to replace the ReLu activation function of the original model. Since images in natural environments usually have complex background and foreground information, the robustness of HardSwish helped the model better handling these situations and further improving the detection accuracy. With the above improvements, the Camellia-YOLACT model was constructed and experimentally validated by testing the Camellia oleifera anthracnose leaf image dataset. [Results and Discussions] A transfer learning approach was used for experimental validation on the Camellia oleifera anthracnose severity grading dataset, and the results of the ablation experiments showed that the mAP75 of Camellia-YOLACT proposed in this study was 86.8%, mAPall was 78.3%, mAR was 91.6% which were 5.7%, 2.5% and 7.9% higher than YOLACT model. In the comparison experiments, Camellia-YOLACT performed better than Segmenting Objects by Locations (SOLO) in terms of both accuracy and speed, and its detection speed was doubled compared to Mask R-CNN algorithm. Therefore, the Camellia-YOLACT algorithm was suitable in Camellia oleifera gardens for anthracnose real-time segmentation. In order to verify the outdoors detection performance of Camellia-YOLACT model, 36 groups of Camellia oleifera anthracnose grading experiments were conducted. Experimental results showed that the grading correctness of Camellia oleifera anthracnose injection severity reached 94.4%, and the average absolute error of K-value was 1.09%. Therefore, the Camellia-YOLACT model proposed in this study has a better performance on the grading of the severity of Camellia oleifera anthracnose. [Conclusions] The Camellia-YOLACT model proposed got high accuracy in leaf and anthracnose segmentation of Camellia oleifera, on the basis of which it can realize automatic grading of the severity of Camellia oleifera anthracnose. This research could provide technical support for the precise control of Camellia oleifera diseases.
[Objective] As a critical component of agricultural product supply chain management, cold chain logistics demand prediction encounters challenges such as inadequate feature extraction, high nonlinearity of data, and the propensity for algorithms to become trapped in local optima during the digital transformation process. To address these issues and enhance the accuracy of demand prediction, achieve intelligent management of the agricultural product supply chain, a combined forecasting model that integrates grey relational analysis (GRA), the wild horse optimizer (WHO), and temporal convolutional networks (TCN) is proposed in this research. [Methods] Firstly, a cold chain logistics indicator system was established for the data of Zhejiang province, China, spanning the years 2000 to 2020. This system covered four key aspects: the economic scale of agricultural products, logistics transportation, digital technology, and agricultural product supply. Then, the GRA was applied to identify relevant indicators of cold chain logistics for agricultural products in Zhejiang province, with 17 indicators selected that had a correlation degree higher than 0.75. Sliding window technology, a problem-solving approach for data structures and algorithms, suitable for reducing the time complexity of data to a better level and improving the execution efficiency of algorithms, was used to partition the selected indicators. Secondly, the TCN model was employed to extract features of different scales by stacking multiple convolutional layers. Each layer utilized different-sized convolutional kernels to capture features within different time ranges. By utilizing the dilated convolutional module of TCN, temporal and spatial relationships within economic data were effectively mined, considering the temporal characteristics of socio-economic data and logistics information in the agricultural supply chain, and exploring the temporal and spatial features of economic data. Simultaneously, the WHO algorithm was applied to optimize five hyperparameters of the TCN model, including the number of TCN layers, the number of filters, residual blocks, Dense layers, and neurons within the Dense layer. Finally, the optimized GRA-WHO-TCN model was used to extract and analyze features from highly nonlinear multidimensional economic data, ultimately facilitating the prediction of cold chain logistics demand. [Results and Discussions] For comparative analysis of the superiority of the GRA-WHO-TCN model, the 17 selected indicators were input into long short-term memory (LSTM), TCN, WHO-LSTM, and WHO-TCN models. The parameters optimized by the WHO algorithm for the TCN model were set respectively: 2 TCN layer was, 2 residual blocks, 1 dense layer, 60 filters, and 16 neurons in the dense layer. The optimized GRA-WHO-TCN temporal model can effectively extract the temporal and spatial features of multidimensional data, fully explore the implicit relationships among indicator factors, and demonstrating good fitting effects. Compared to GRA-LSTM and GRA-TCN models, the GRA-TCN model exhibited superior performance, with a lower root mean square error of 37.34 and a higher correlation coefficient of 0.91, indicating the advantage of the TCN temporal model in handling complex nonlinear data. Furthermore, the GRA-WHO-LSTM and GRA-WHO-TCN models optimized by the WHO algorithm had improved prediction accuracy and stability compared to GRA-LSTM and GRA-TCN models, illustrating that the WHO algorithm effectively optimized model parameters to enhance the effectiveness of model fitting. When compared to the GRA-WHO-LSTM model, the GRA-WHO-TCN model displayed a lower root mean square error of 11.3 and an effective correlation coefficient of 0.95, predicting cold chain logistics demand quantities in Zhejiang province for the years 2016-2020 as 29.8, 30.46, 24.87, 26.45, and 27.99 million tons, with relative errors within 0.6%, achieving a high level of prediction accuracy. This achievement showcases a high level of prediction accuracy and underscores the utility of the GRA-WHO-TCN model in forecasting complex data scenarios. [Conclusions] The proposed GRA-WHO-TCN model demonstrated superior parameter optimization capabilities and predictive accuracy compared to the GRA-LSTM and GRA-TCN models. The predicted results align well with the development of cold chain logistics of agricultural products in Zhejiang province. This provides a scientific prediction foundation and practical reference value for the development of material flow and information flow in the agricultural supply chain under the digital economy context.
Objective Knowledge services in agricultural scenarios have the characteristics of long periodicity and prolonged activity time. Traditional recommendation models cannot effectively mine hidden information in agricultural scenarios, in order to improve the quality of agricultural knowledge recommendation services, agricultural contextual information based on agricultural time should be fully considered. To address these issues, a Time-aware and filter-enhanced sequential recommendation model for agricultural knowledge (TiFSA) was proposed, integrating temporal perception and enhanced filtering. Methods First, based on the temporal positional embedding, combining the temporal information of farmers' interactions with positional embedding based on time perception, it helped to learn project relevance based on agricultural season in agricultural contexts. A multi-head self-attention network recommendation algorithm based on time-awareness was proposed for the agricultural knowledge recommendation task, which extracted different interaction time information in the user interaction sequence and introduced it into the multi-head self-attention network to calculate the attention weight, which encoded the user's periodic interaction information based on the agricultural time, and also effectively captured the user's dynamic preference information over time. Then, through the temporal positional embedding, a filter filtering algorithm was introduced to adaptively attenuate the noise in farmers' situational data adaptively. The filtering algorithm was introduced to enhance the filtering module to effectively filter the noisy information in the agricultural dataset and alleviate the overfitting problem due to the poorly normalized and sparse agricultural dataset. By endowing the model with lower time complexity and adaptive noise attenuation capability. The applicability of this method in agricultural scenarios was improved. Next, a multi-head self attention network with temporal information was constructed to achieve unified modeling of time, projects, and features, and represent farmers' preferences of farmers over time in context, thereby providing reliable recommendation results for users. Finally, the AdamW optimizer was used to update and compute the model parameters. AdamW added L2 regularization and an appropriate penalty mechanism for larger weights, which could update all weights more smoothly and alleviate the problem of falling into local minima. Applied in the field of agricultural recommendation, it could further improve the training effect of the model. The experimental data came from user likes, comments, and corresponding time information in the "National Agricultural Knowledge Intelligent Service Cloud Platform", and the dataset ml-1m in the movie recommendation scenario was selected as an auxiliary validation of the performance of this model. Results and Discussions According to the user interaction sequence datasets in the "National Agricultural Knowledge Intelligent Service Cloud Platform", from the experimental results, it could be learned that TiFSA outperforms the other models on two different datasets, in which the enhancement was more obvious on the Agriculture dataset, where HR and NDCG were improved by 14.02% and 16.19%, respectively, compared to the suboptimal model, TiSASRec; while on the ml-1m dataset compared to the suboptimal model, SASRec, HR and NDCG were improved by 1.90% and 2.30%, respectively. In summary, the TiFSA model proposed in this paper has a large improvement compared with other models, which verifies verified the effectiveness of the TiFSA model and showed that the time interval information of farmer interaction and the filtering algorithm play an important role in the improvement of the model performance in the agricultural context. From the results of the ablation experiments, it could be seen that when the time-aware and enhanced filtering modules were removed, the values of the two metrics HR@10 and NDCG@10 were 0.293 6 and 0.203 9, respectively, and the recommended performance was poor. When only the time-aware module and only the augmentation filtering module were removed, the experimental results had different degrees of improvement compared to TiFSA-tf, and the TiFSA model proposed in this paper achieved the optimal performance in the two evaluation metrics. When only the multi-head self-attention network was utilized for recommendation, both recommendation metrics of the model were lower, indicating that the traditional sequence recommendation method that only considered the item number was not applicable to agricultural scenarios. When the augmented filtering module was introduced without the time-aware module, the model performance was improved, but still failed to achieve the ideal recommendation effect. When only the time-aware module was introduced without the augmented filtering module, there was a significant improvement in the model effect, which proved that the time-aware module was more applicable to agricultural scenarios and can effectively improve the model performance of the sequence recommendation task. When both time-aware and augmented filtering modules were introduced, the model performance was further improved, which on the one hand illustrated the dependence of the augmented filtering module on the time-aware module, and on the other hand verified the necessity of adopting the augmented filtering to the time-aware self-attention network model. Conclusions This research proposes an agricultural knowledge recommendation model that integrates time-awareness and augmented filtering, which introduces the user's interaction time interval into the embedded information, so that the model effectively learns the information of agricultural time in the agricultural scene, and the prediction of the user's interaction time and the object is more closely related to the actual scene; augmented filtering algorithms are used to attenuate the noise in the agricultural data. At the same time, the enhanced filtering algorithm is used to attenuate the noise in the agricultural data, and can be effectively integrated into the model for use, further improving the recommendation performance of the model. The experimental results show the effectiveness of the proposed TiFSA model on the agricultural dataset. The ablation experiments confirm the positive effect of time-awareness and enhanced filtering modules on the improvement of recommendation performance.
Objective Flax, characterized by its short growth cycle and strong adaptability, is one of the major cash crops in northern China. Due to its versatile uses and unique quality, it holds a significant position in China's oil and fiber crops. The quality of flax seeds directly affects the yield of the flax plant. Seed evaluation is a crucial step in the breeding process of flax. Common parameters used in the seed evaluation process of flax include circumference, area, length axis, and 1 000-seed weight. To ensure the high-quality production of flax crops, it is of great significance to understand the phenotypic characteristics of flax seeds, select different resources as parents based on breeding objectives, and adopt other methods for the breeding, cultivation, and evaluation of seed quality and traits of flax. Methods In response to the high error rates and low efficiency issues observed during the automated seed testing of flax seeds, the measurement methods were explored of flax seed contours based on machine vision research. The flax seed images were preprocessed, and the collected color images were converted to grayscale. A filtering and smoothing process was applied to obtain binary images. To address the issues of flax seed overlap and adhesion, a contour fitting image segmentation method based on fused corner features was proposed. This method incorporated adaptive threshold selection during edge detection of the image contour. Only multi-seed target areas that met certain criteria were subjected to image segmentation processing, while single-seed areas bypassed this step and were directly summarized for seed testing data. After obtaining the multi-seed adhesion target areas, the flax seeds underwent contour approximation, corner extraction, and contour fitting. Based on the provided image contour information, the image contour shape was approximated to another contour shape with fewer vertices, and the original contour curve was simplified to a more regular and compact line segment or polygon, minimizing computational complexity. All line shape characteristics in the image were marked as much as possible. Since the pixel intensity variations in different directions of image corners were significant, the second derivative matrix based on pixel grayscale values was used to detect image corners. Based on the contour approximation algorithm, contour corner detection was performed to obtain the coordinates of each corner. The resulting contour points and corners were used as outputs to further improve the accuracy and precision of subsequent contour fitting methods, resulting in a two-dimensional discrete point dataset of the image contour. Using the contour point dataset as an input, the geometric moments of the image contour were calculated, and the optimal solution for the ellipse parameters was obtained through numerical optimization based on the least squares method and the geometric features of the ellipse shape. Ultimately, the optimal contour was fitted to the given image, achieving the segmentation and counting of flax seed images. Meanwhile, each pixel in the digital image was a uniform small square in size and shape, so the circumference, area, and major and minor axes of the flax seeds could be represented by the total number of pixels occupied by the seeds in the image. The weight of a single seed could be calculated by dividing the total weight of the seeds by the total number of seeds detected by the contour, thereby obtaining the weight of the individual seed and converting it accordingly. Through the pixelization of the 1 yuan and 1 jiao coins from the fifth iteration of the 2019 Renminbi, a summary of the circumference, area, major axis, minor axis, and 1 000-seed weight of the flax seeds was achieved. Additionally, based on the aforementioned method, this study designed an automated real-time analysis system for flax seed testing data, realizing the automation of flax seed testing research. Experiments were conducted on images of flax seeds captured by an industrial camera. Results and Discussions The proposed automated seed identification method achieved an accuracy rate of 97.28% for statistically distinguishing different varieties of flax seeds. The average processing time for 100 seeds was 69.58 ms. Compared to the extreme erosion algorithm and the watershed algorithm based on distance transformation, the proposed method improved the average calculation accuracy by 19.6% over the extreme erosion algorithm and required a shorter average computation time than the direct use of the watershed algorithm. Considering the practical needs of automated seed identification, this method did not employ methods such as dilation or erosion for image morphology processing, thereby preserving the original features of the image to the greatest extent possible. Additionally, the flax seed automated seed identification data real-time analysis system could process image information in batches. By executing data summarization functions, it automatically generated corresponding data table folders, storing the corresponding image data summary tables. Conclusions The proposed method exhibits superior computational accuracy and processing speed, with shorter operation time and robustness. It is highly adaptable and able to accurately acquire the morphological feature parameters of flax seeds in bulk, ensuring measurement errors remain within 10%, which could provide technical support for future flax seed evaluation and related industrial development.
Objective To effectively tackle the unique attributes of wheat leaf pests and diseases in their native environment, a high-caliber and efficient pest detection model named YOLOv8-SS (You Only Look Once Version 8-SS) was proposed. This innovative model is engineered to accurately identify pests, thereby providing a solid scientific foundation for their prevention and management strategies. Methods A total of 3 639 raw datasets of images of wheat leaf pests and diseases were collected from 6 different wheat pests and diseases in various farmlands in the Yuchong County area of Gansu Province, at different periods of time, using mobile phones. This collection demonstrated the team's proficiency and commitment to advancing agricultural research. The dataset was meticulously constructed using the LabelImg software to accurately label the images with targeted pest species. To guarantee the model's superior generalization capabilities, the dataset was strategically divided into a training set and a test set in an 8:2 ratio. The dataset includes thorough observations and recordings of the wheat leaf blade's appearance, texture, color, as well as other variables that could influence these characteristics. The compiled dataset proved to be an invaluable asset for both training and validation activities. Leveraging the YOLOv8 algorithm, an enhanced lightweight convolutional neural network, ShuffleNetv2, was selected as the basis network for feature extraction from images. This was accomplished by integrating a 3×3 Depthwise Convolution (DWConv) kernel, the h-swish activation function, and a Squeeze-and-Excitation Network (SENet) attention mechanism. These enhancements streamlined the model by diminishing the parameter count and computational demands, all while sustaining high detection precision. The deployment of these sophisticated methodologies exemplified the researchers' commitment and passion for innovation. The YOLOv8 model employs the SEnet attention mechanism module within both its Backbone and Neck components, significantly reducing computational load while bolstering accuracy. This method exemplifies the model's exceptional performance, distinguishing it from other models in the domain. By integrating a dedicated small target detection layer, the model's capabilities have been augmented, enabling more efficient and precise pest and disease detection. The introduction of a new detection feature map, sized 160×160 pixels, enables the network to concentrate on identifying small-targeted pests and diseases, thereby enhancing the accuracy of pest and disease recognition. Results and Discussion The YOLOv8-SS wheat leaf pests and diseases detection model has been significantly improved to accurately detect wheat leaf pests and diseases in their natural environment. By employing the refined ShuffleNet V2 within the DarkNet-53 framework, as opposed to the conventional YOLOv8, under identical experimental settings, the model exhibited a 4.53% increase in recognition accuracy and a 4.91% improvement in F1-Score, compared to the initial model. Furthermore, the incorporation of a dedicated small target detection layer led to a subsequent rise in accuracy and F1-Scores of 2.31% and 2.16%, respectively, despite a minimal upsurge in the number of parameters and computational requirements. The integration of the SEnet attention mechanism module into the YOLOv8 model resulted in a detection accuracy rate increase of 1.85% and an F1-Score enhancement of 2.72%. Furthermore, by swapping the original neural network architecture with an enhanced ShuffleNet V2 and appending a compact object detection sublayer (namely YOLOv8-SS), the resulting model exhibited a heightened recognition accuracy of 89.41% and an F1-Score of 88.12%. The YOLOv8-SS variant substantially outperformed the standard YOLOv8, showing a remarkable enhancement of 10.11% and 9.92% in accuracy, respectively. This outcome strikingly illustrates the YOLOv8-SS's prowess in balancing speed with precision. Moreover, it achieves convergence at a more rapid pace, requiring approximately 40 training epochs, to surpass other renowned models such as Faster R-CNN, MobileNetV2, SSD, YOLOv5, YOLOX, and the original YOLOv8 in accuracy. Specifically, the YOLOv8-SS boasted an average accuracy 23.01%, 15.13%, 11%, 25.21%, 27.52%, and 10.11% greater than that of the competing models, respectively. In a head-to-head trial involving a public dataset (LWDCD 2020) and a custom-built dataset, the LWDCD 2020 dataset yielded a striking accuracy of 91.30%, outperforming the custom-built dataset by a margin of 1.89% when utilizing the same network architecture, YOLOv8-SS. The AI Challenger 2018-6 and Plant-Village-5 datasets did not perform as robustly, achieving accuracy rates of 86.90% and 86.78% respectively. The YOLOv8-SS model has shown substantial improvements in both feature extraction and learning capabilities over the original YOLOv8, particularly excelling in natural environments with intricate, unstructured backdrops. Conclusion The YOLOv8-SS model is meticulously designed to deliver unmatched recognition accuracy while consuming a minimal amount of storage space. In contrast to conventional detection models, this groundbreaking model exhibits superior detection accuracy and speed, rendering it exceedingly valuable across various applications. This breakthrough serves as an invaluable resource for cutting-edge research on crop pest and disease detection within natural environments featuring complex, unstructured backgrounds. Our method is versatile and yields significantly enhanced detection performance, all while maintaining a lean model architecture. This renders it highly appropriate for real-world scenarios involving large-scale crop pest and disease detection.
To deal with the issues of low accuracy of apple leaf disease images recognition under weak supervision with only image category labeling, an improved CBAM-ResNet-based algorithm was proposed in this research. Using ResNet18 as the base model, the multilayer perceptron (MLP) in the lightweight convolutional block attention module (CBAM) attention mechanism channel was improved by up-dimensioning to amplify the details of apple leaf disease features. The improved CBAM attention module was incorporated into the residual module to enhance the key details of AlphaDropout with SeLU (Scaled Exponential Linearunits) to prevent overfitting of its network and accelerate the convergence effect of the model. Finally, the learning rate was adjusted using a single-cycle cosine annealing algorithm to obtain the disease recognition model. The training test was performed under weak supervision with only image-level annotation of all sample images, which greatly reduced the annotation cost. Through ablation experiments, the best dimensional improvement of MLP in CBAM was explored as 2. Compared with the original CBAM, the accuracy rate was increased by 0.32%, and the training time of each round was reduced by 8 s when the number of parameters increased by 17.59%. Tests were conducted on a dataset of 6185 images containing five diseases, including apple spotted leaf drop, brown spot, mosaic, gray spot, and rust, and the results showed that the model achieved an average recognition accuracy of 98.44% for the five apple diseases under weakly supervised learning. The improved CBAM-ResNet18 had increased by 1.47% compared with the pre-improved ResNet18, and was higher than VGG16, DesNet121, ResNet50, ResNeXt50, EfficientNet-B0 and Xception control model. In terms of learning efficiency, the improved CBAM-ResNet18 compared to ResNet18 reduced the training time of each round by 6 s under the condition that the number of parameters increased by 24.9%, and completed model training at the fastest speed of 137 s per round in VGG16, DesNet121, ResNet50, ResNeXt50, Efficient Net-B0 and Xception control models. Through the results of the confusion matrix, the average precision, average recall rate, and average F1 score of the model were calculated to reach 98.43%, 98.46%, and 0.9845, respectively. The results showed that the proposed improved CBAM-ResNet18 model could perform apple leaf disease identification and had good identification results, and could provide technical support for intelligent apple leaf disease identification providing.
Pre-training refers to the process of training deep neural network parameters on a large corpus before a specific task model performs a particular task. This approach enables downstream tasks to fine-tune the pre-trained model parameters based on a small amount of labeled data, eliminating the need to train a new model from scratch. Currently, research on named entity recognition (NER) using pre-trained language model (PLM) only uses the last layer of the PLM to express output when facing challenges such as complex entity naming methods and fuzzy entity boundaries in the agricultural field. This approach ignores the rich information contained in the internal layers of the model themselves. To address these issues, a named entity recognition method based on progressive convolutional networks has been proposed. This method stores natural sentences and outputs representations of each layer obtained through PLM. The intermediate outputs of the pre-trained model are sequentially convolved to extract shallow feature information that may have been overlooked previously. Using the progressive convolutional network module proposed in this research, the adjacent two-layer representations are convolved from the first layer, and the fusion result continues to be convolved with the next layer, resulting in enhanced sentence embedding that includes the entire information dimension of the model layer. The method does not require the introduction of external information, which makes the sentence representation contain richer information. Research has shown that the sentence embedding output of the model layer near the input contains more fine-grained information, such as phrases and phrases, which can assist with NER problems in the agricultural field. Fully utilizing the computational power already used, the results obtained can enhance the representation embedding of sentences. Finally, the conditional random field (CRF) model was used to generate the global optimal sequence. On a constructed agricultural dataset containing four types of agricultural entities, the proposed method's comprehensive indicator F1 value increased by 3.61% points compared to the basic BERT (Bidirectional Encoder Representation from Transformers) model. On the open dataset MSRA, the F1 value also increased to 94.96%, indicating that the progressive convolutional network can enhance the model's ability to represent natural language and has advantages in NER tasks.
Crops have diverse structures and complex growth environments. RGB image data can reflect the texture and color features of plants accurately, while 3D data contains information about crop volume. The combination of RGB image and 3D point cloud data can achieve the extraction of two-dimensional and three-dimensional phenotypic parameters of crops, which is of great significance for the research of phenomics methods. In this study, potatoe plants were chosen as the research subject, and RGB cameras and laser scanners were used to collect 50 potato RGB images and 3D laser point cloud data. The segmentation accuracy of four deep learning semantic segmentation methods, OCRNet, UpNet, PaNet, and DeepLab v3+, were compared and analyzed for the RGB images. OCRNet, which demonstrated higher accuracy, was used to perform semantic segmentation on top-view RGB images of potatoes. Mean shift clustering algorithm was optimized for laser point cloud data processing, and single-plant segmentation of laser point cloud data was completed. Stem and leaf segmentation of single-plant potato point cloud data were accurately performed using Euclidean clustering and K-Means clustering algorithms. In addition, a strategy was proposed to establish a one-to-one correspondence between RGB images and point clouds of single-plant potatoes using pot numbering. 8 2D phenotypic parameters and 10 3D phenotypic parameters, including maximum width, perimeter, area, plant height, volume, leaf length, and leaf width, etc., were extracted from RGB images and laser point clouds, respectively. Finally, the accuracy of three representative and easily measurable phenotypic parameters, leaf number, plant height, and maximum width were evaluated. The mean absolute percentage errors (MAPE) were 8.6%, 8.3% and 6.0%, respectively, while the root mean square errors (RMSE) were 1.371 pieces, 3.2 cm and 1.86 cm, respectively, and the determination coefficients (R2) were 0.93, 0.95 and 0.91, respectively. The research results indicated that the extracted phenotype parameters can accurately and efficiently reflect the growth status of potatoes. Combining the RGB image data of potatoes with three-dimensional laser point cloud data can fully exploit the advantages of the rich texture and color characteristics of RGB images and the volumetric information provided by three-dimensional point clouds, achieving non-destructive, efficient, and high-precision extraction of two-dimensional and three-dimensional phenotype parameters of potato plants. The achievements of this study could not only provide important technical support for the cultivation and breeding of potatoes but also provide strong support for phenotype-based research.
Beef cattle in the farm are active, which leads the collection of posture of the beef cattle changeable, so it is difficult to automatically measure the body size of the beef cattle. Aiming at the above problems, an automatic measurement method for beef cattle's body size under multi-pose was proposed by analyzing the skeleton features of beef cattle head and the edge contour features of beef cattle images. Firstly, the consumer-grade depth camera Azure Kinect DK was used to collect the top-view depth video data directly above the beef cattle and the video data were divided into frames to obtain the original depth image. Secondly, the original depth image was processed by shadow interpolation, normalization, image segmentation and connected domain to remove the complex background and obtain the target image containing only beef cattle. Thirdly, the Zhang-Suen algorithm was used to extract the beef cattle skeleton of the target image, and calculated the intersection points and endpoints of the skeleton, so as to analyze the characteristics of the beef cattle head to determine the head removal point, and to remove the beef cattle head information from the image. Finally, the curvature curve of the beef cattle profile was obtained by the improved U-chord curvature method. The body measurement points were determined according to the curvature value and converted into three-dimensional spaces to calculate the body size parameters. In this paper, the postures of beef cattle, which were analyzed by a large amount of depth image data, were divided into left crooked, right crooked, correct posture, head down and head up, respectively. The test results showed that the head removal method proposed based on the skeleton in multiple postures hads head removel success rate higher than 92% in the five postures. Using the body measurement point extraction method based on the improved U-chord curvature proposed, the average absolute error of body length measurement was 2.73 cm, the average absolute error of body height measurement was 2.07 cm, and the average absolute error of belly width measurement was 1.47 cm. The method provides a better way to achieve the automatic measurement of beef cattle body size in multiple poses.
Corn and soybean are upland grain in the same season, and the contradiction of scrambling for land between corn and soybean is prominent in China, so it is necessary to explore the price relations between corn and soybean. In addition, agricultural futures have the function of price discovery compared with the spot. Therefore, the analysis and prediction of corn and soybean futures prices are of great significance for the management department to adjust the planting structure and for farmers to select the crop varieties. In this study, the correlation between corn and soybean futures prices was analyzed, and it was found that the corn and soybean futures prices have a strong correlation by correlation test, and soybean futures price is the Granger reason of corn futures price by Granger causality test. Then, the corn and soybean futures prices were predicted using a long short-term memory (LSTM) model. To optimize the futures price prediction model performance, Attention mechanism was introduced as Attention-LSTM to assign weights to the outputs of the LSTM model at different times. Specifically, LSTM model was used to process the input sequence of futures prices, the Attention layer assign different weights to the outputs, and then the model output the prediction results after a layer of linearity. The experimental results showed that Attention-LSTM model could significantly improve the prediction performance of both corn and soybean futures prices compared to autoregressive integrated moving average model (ARIMA), support vector regression model (SVR), and LSTM. For example, mean absolute error (MAE) was improved by 3.8% and 3.3%, root mean square error (RMSE) was improved by 0.6% and 1.8% and mean absolute error percentage (MAPE) was improved by 4.8% and 2.9% compared with a single LSTM, respectively. Finally, the corn futures prices were forecasted using historical corn and soybean futures prices together. Specifically, two LSTM models were used to process the input sequences of corn futures prices and soybean futures prices respectively, two parameters were trained to perform a weighted summation of the output of two LSTM models, and the prediction results were output by the model after a layer of linearity. The experimental results showed that MAE was improved by 6.9%, RMSE was improved by 1.1% and MAPE was improved by 5.3% compared with the LSTM model using only corn futures prices. The results verify the strong correlation between corn and soybean futures prices at the same time. In conclusion, the results verify the Attention-LSTM model can improve the performances of soybean and corn futures price forecasting compared with the general prediction model, and the combination of related agricultural futures price data can improve the prediction performances of agricultural product futures forecasting model.
Convolutional neural network models have different advantages and disadvantages, it is becoming more and more difficult to select an appropriate convolutional neural network model in an actual fish identification project. The identification of underwater fish is a challenge task due to varies in illumination, low contrast, high noise, low resolution and sample imbalance between each type of image from the real underwater environment. In addition, deploying models to mobile devices directly will reduce the accuracy of the model sharply. In order to solve the above problems, Fish Recognition Ground-Truth dataset was used to training model in this study, which is provided by Fish4Knowledge project from University of Edinburgh. It contains 27,370 images with 23 fish species, and has been labeled manually by marine biologists. AlexNet, GoogLeNet, ResNet and DenseNet models were selected initially according to the characteristics of real-time underwater fish identification task, then a comparative experiment was designed to explore the best network model. Random image flipping, rotation and color dithering were used to enhance data based on ground-truth fish dataset in response to the limited number of underwater fish images. Considering that there was a serious imbalance in the number of samples in each category, the label smoothing technology was used to alleviate model overfitting. The Ranger optimizer and Cosine learning rate attenuation strategy were used to further improve the training effect of the models. The accuracy and recall rate information of each model were recorded and counted. The results showed that, the accuracy and recall rate of the fish recognition model based on DenseNet reached 99.21% and 96.77% in train set and validation set respectively, its F1 value reached 0.9742, which was the best model obtained in the experiment. Finally, a remote fish identification system was designed based on Python language, in this system the model was deployed to linux server and the Android APP was responsible for uploading fish images via http to request server to identify the fishes and displaying the identification information returned by server, such as fish species, profiles, habits, distribution, etc. A set of recognition tests were performed on real Android phone and the results showed that in the same local area net the APP could show fish information rapidly and exactly within 1 s.
Objects in farmlands often have characteristic of small volume and high density with variable light and complex background, and the available object detection models could not get satisfactory recognition results. Taking bees as research objects, a method that could overcome the influence from the complex backgrounds, the difficulty in small object feature extraction was proposed, and a detection algorithm was created for small objects irrelevant to image size. Firstly, the original image was split into some smaller sub-images to increase the object scale, and the marked objects were assigned to the sub-images to produce a new dataset. Then, the model was trained again using transfer learning to get a new object detection model. A certain overlap rate was set between two adjacent sub-images in order to restore the objects. The objects from each sub-image was collected and then non-maximum suppression (NMS) was performed to delete the redundant detection boxes caused by the network, an improved NMS named intersection over small NMS (IOS-NMS) was then proposed to delete the redundant boxes caused by the overlap between adjacent sub-images. Validation tests were performed when sub-image size was set was 300×300, 500×500 and 700×700, the overlap rate was set as 0.2 and 0.05 respectively, and the results showed that when using single shot multibox detector (SSD) as the object detection model, the recall rate and precision was generally higher than that of SSD with the maximum difference 3.8% and 2.6%, respectively. In order to further verify the algorithm in small target recognition with complex background, three bee images with different scales and different scenarios were obtained from internet and test experiments were conducted using the new proposed algorithm and SSD. The results showed that the proposed algorithm could improve the performance of target detection and had strong scale adaptability and generalization. Besides, the new algorithm required multiple forward reasoning for a single image, so it was not time-efficient and was not suitable for edge calculation.
Aiming at the real-time detection and classification of the growth period of crops in the current digital cultivation and regulation technology of facility agriculture, an improved YOLOv4 method for identifying the growth period of strawberries in a greenhouse environment was proposed. The attention mechanism into the Cross Stage Partial Residual (CSPRes) module of the YOLOv4 backbone network was introduced, and the target feature information of different growth periods of strawberries while reducing the interference of complex backgrounds was integrated, the detection accuracy while ensured real-time detection efficiency was improved. Took the smart facility strawberry in Yunnan province as the test object, the results showed that the detection accuracy (AP) of the YOLOv4-CBAM model during flowering, fruit expansion, green and mature period were 92.38%, 82.45%, 68.01% and 92.31%, respectively, the mean average precision (mAP) was 83.78%, the mean inetersection over union (mIoU) was 77.88%, and the detection time for a single image was 26.13 ms. Compared with the YOLOv4-SC model, mAP and mIoU were increased by 1.62% and 2.73%, respectively. Compared with the YOLOv4-SE model, mAP and mIOU increased by 4.81% and 3.46%, respectively. Compared with the YOLOv4 model, mAP and mIOU increased by 8.69% and 5.53%, respectively. As the attention mechanism was added to the improved YOLOv4 model, the amount of parameters increased, but the detection time of improved YOLOv4 models only slightly increased. At the same time, the number of fruit expansion period recognized by YOLOv4 was less than that of YOLOv4-CBAM, YOLOv4-SC and YOLOv4-SE, because the color characteristics of fruit expansion period were similar to those of leaf background, which made YOLOv4 recognition susceptible to leaf background interference, and added attention mechanism could reduce background information interference. YOLOv4-CBAM had higher confidence and number of identifications in identifying strawberry growth stages than YOLOv4-SC, YOLOv4-SE and YOLOv4 models, indicated that YOLOv4-CBAM model can extract more comprehensive and rich features and focus more on identifying targets, thereby improved detection accuracy. YOLOv4-CBAM model can meet the demand for real-time detection of strawberry growth period status.
In order to count soybean seeds quickly and accurately, improve the speed of seed test and the level of soybean breeding, a method of soybean seed counting based on VGG-Two (VGG-T) was developed in this research. Firstly, in view of the lack of available image dataset in the field of soybean seed counting, a fast target point labeling method of combining pre-annotation based on digital image processing technology with manual correction annotation was proposed to speed up the establishment of publicly available soybean seed image dataset with annotation. Only 197 min were taken to mark 37,563 seeds when using this method, which saved 1592 min than ordinary manual marking and could reduce 96% of manual workload. At the same time, the dataset in this research is the largest annotated data set for soybean seed counting so far. Secondly, a method that combined the density estimation-based and the convolution neural network (CNN) was developed to accurately estimate the seed count from an individual threshed seed image with a single perspective. Thereinto, a CNN architecture consisting of two columns of the same network structure was used to learn the mapping from the original pixel to the density map. Due to the very limited number of training samples and the effect of vanishing gradients on deep neural networks, it is not easy for the network to learn all parameters at the same time. Inspired by the success of pre-training, this research pre-trained the CNN in each column by directly mapping the output of the fourth convolutional layer to the density map. Then these pre-trained CNNs were used to initialize CNNs in these two columns and fine-tune all parameters. Finally, the model was tested, and the effectiveness of the algorithm through three comparative experiments (with and without data enhancement, VGG16 and VGG-T, multiple sets of test set) was verified, which respectively provided 0.6 and 0.2 mean absolute error (MAE) in the original image and patch cases, while mean squared error (MSE) were 0.6 and 0.3. Compared with traditional image morphology operations, ResNet18, ResNet18-T and VGG16, the method proposed improving the accuracy of soybean seed counting. In the testset containing soybean seeds of different densities, the error fluctuation was small, and it still had excellent counting performance. At the same time, compared with manual counting and photoelectric seed counter, it saved about 2.493 h and 0.203 h respectively for counting 11,350 soybean seeds, realizing rapid soybean seeds counting.
At present, the dynamic detection and monitoring of maize seedling mainly rely on manual observation, which is time-consuming and laborious, and only small quadrats can be selected to estimate the overall emergence situation. In this research, two kinds of data sources, the high-time-series RGB images obtained by the plant high-throughput phenotypic platform (HTPP) and the RGB images obtained by the unmanned aerial vehicle (UAV) platform, were used to construct the image data set of maize seedling process under different light conditions. Considering the complex background and uneven illumination in the field environment, a residual unit based on the Faster R-CNN was built and ResNet50 was used as a new feature extraction network to optimize Faster R-CNN to realize the detection and counting of maize seedlings in complex field environment. Then, based on the high time series image data obtained by the HTPP, the dynamic continuous monitoring of maize seedlings of different varieties and densities was carried out, and the seedling duration and uniformity of each maize variety were evaluated and analyzed. The experimental results showed that the recognition accuracy of the proposed method was 95.67% in sunny days and 91.36% in cloudy days when it was applied to the phenotypic platform in the field. When applied to the UAV platform to monitor the emergence of maize, the recognition accuracy of sunny and cloudy days was 91.43% and 89.77% respectively. The detection accuracy of the phenotyping platform image was higher, which could meet the needs of automatic detection of maize emergence in actual application scenarios. In order to further verify the robustness and generalization of the model, HTPP was used to obtain time series data, and the dynamic emergence of maize was analyzed. The results showed that the dynamic emergence results obtained by HTPP were consistent with the manual observation results, which shows that the model proposed in this research is robust and generalizable.
Yield estimation of fruit tree is one of the important works in orchard management. In order to improve the accuracy of in-situ yield estimation of apple trees in orchard, a method for the yield estimation of single apple tree, which includes an improved YOLOv5 fruit detection network and a yield fitting network was proposed. The in-situ images of the apples without bags at different periods were acquired by using an unmanned aerial vehicle and Raspberry Pi camera, formed an image sample data set. For dealing with no attention preference and the parameter redundancy in feature extraction, the YOLOv5 network was improved by two approaches: 1) replacing the depth separable convolution, and 2) adding the attention mechanism module, so that the computation cost was decreased. Based on the improvement, the quantity of fruit was estimated and the total area of the bounding box of apples were respectively obtained as output. Then, these results were used as the input of the yield fitting network and actual yields were applied as the output to train the yield fitting network. The final model of fruit tree production estimation was obtained by combining the improved YOLOv5 network and the yield fitting network. Yield estimation experimental results showed that the improved YOLOv5 fruit detection algorithm could improve the recognition accuracy and the degree of lightweight. Compared with the previous algorithm, the detection speed of the algorithm proposed in this research was increased by up to 15.37%, while the mean of average accuracy (mAP) was raised up to 96.79%. The test results based on different data sets showed that the lighting conditions, coloring time and with white cloth in background had a certain impact on the accuracy of the algorithm. In addition, the yield fitting network performed better on predicting the yield of apple trees. The coefficients of determination in the training set and test set were respectively 0.7967 and 0.7982. The prediction accuracy of different yield samples was generally stable. Meanwhile, in terms of the with/without of white cloth in background, the range of relative error of the fruit tree yield measurement model was respectively within 7% and 13%. The yield estimation method of apple tree based on improved lightweight YOLOv5 had good accuracy and effectiveness, which could achieve yield estimation of apples in the natural environment, and would provide a technical reference for intelligent agricultural equipment in modern orchard environment.
Dissolved oxygen is an important environmental factor for prawn breeding. In order to improve the prediction accuracy of dissolved oxygen concentration in prawn pond, and solve the problem of low prediction accuracy of different frequency domain modal classification after empirical modal decomposition of nonlinear time series data when there are few training samples, an combination prediction model based on empirical mode decomposition (EMD), random forest (RF) and long short term memory neural network (LSTM) was proposed in this research. Firstly, the time series data of prawn breeding dissolved oxygen concentration were decomposed at multiple scales by EMD to obtain a set of stationary intrinsic mode function (IMF). Secondly, with fewer training samples, poor predicts effects on the low-frequency were verified component by LSTM. Then, IMF1－IMF4 were divided into high-frequency components through test results and used for LSTM model. IMF5－IMF7, Rn were divided for RF model, the EMD-RF-LSTM combination model was constructed to improve the prediction accuracy. Modeled low-frequency and high-frequency components IMF using RF and LSTM, then predictions of each component were accumulated and the prediction value of dissolved oxygen of sequence data were got. Finally, the performance of the model was compared with the limit learning machine (ELM), RF, standard LSTM, EMD-ELM and EMD-RF, EMD-LSTM, etc. In the test based on real dataset, the EMD-ELM model contrasted with ELM model, reduced the mean absolute error (MAPE), root mean square error (RMSE) and mean absolute error (MAE) by 30.11%, 29.60% and 32.95%, respectively. The MAPE, RMSE, MAE for the proposed models were 0.0129,0.1156,0.0844, respectively. MAPE decreased by 84.07%, 57.57%, and 49.81% compared with EMD-ELM, EMD-RF and EMD-LSTM, respectively, the prediction accuracy was significantly improved. The results show that the proposed model EMD-RF-LSTM has good prediction performance and generalization ability, which is meets the actual demand of accurate prediction of dissolved oxygen concentration in prawn culture, and can provide reference for the prediction and early warning of prawn pond water quality.
With the change of global climate, extreme weather events such as drought and flood disasters occur frequently. These have a great impact on crop yields. As an important main grain producing area, the impact of drought and flood on the agricultural production of the three provinces in three northeast provinces of China cannot be ignored. Based on historic meteorological data such as daily precipitation, maximum temperature, minimum temperature, 2 m average wind speed, sunshine hours and relative humidity, etc., the standardized precipitation index (SPI) and standardized precipitation evapotranspiration index (SPEI) during 1988－2017 in three northeast provinces of China were calculated with different time scales. Through comparing with characterization of drought and flood disasters in history, SPEI was chosen to judge drought and flood in the growth season of spring maize. With the purpose of evaluating the effects of drought and flood on spring maize yield, based on the distance correlation analysis method, the index of reasonable time scale and key month were selected to analyze the relationship between the index and the relative meteorological yield of spring maize. The relationship between water conditions at different growth stages and the yield was also analyzed. The results showed that: (1) both SPI and SPEI could represent the drought and flood conditions in three northeast provinces of China. Compared with SPI, SPEI had higher correlation with the drought and flood disaster rate, and SPEI was more advantageous in characterizing the drought and flood conditions in the study area; (2) relative meteorological yield was significantly correlated with drought disaster rate in all three provinces (P<0.01), and reached 0.05 significant level with flood disaster rate in Liaoning province, but not significant in Jilin and Heilongjiang province; (3) the distance correlation coefficient between SPEI3-8 and relative meteorological in Liaoning province was the largest, and that between SPEI6-8 and relative meteorological yield in Jilin and Heilongjiang province was the largest. SPEI and relative meteorological yield showed a downward parabolic trend. Overall, the impact of waterlogging on the yield in Liaoning was slightly less than that of drought, mild drought or moderate wet could lead to a decrease in yield. The impact of drought disaster in Jilin and Heilongjiang was much greater than that of flood, but severe humidity could lead to a decrease in yield. Compared with other provinces, the maize yield in Liaoning province fluctuated more sharply with the change of dry and wet; (4) in Liaoning province, maize may reach the highest yield when the jointing-heading period was close to severe wet, which was mainly affected by drought. In the late growing season, the impact of flood disasters was more severe than that of the early growing season, and both drought and flood disasters had effects on the yield. In Jilin province, the highest yield of spring maize was reached when SPEI was about 1.0 during the period of emergence-jointing and jointing-heading, and the effect of drought was more serious during the period heading-milking. The key growth periods in Heilongjiang province were mainly affected by drought, and the maximum yield was reached in the normal-wet years of emergence-jointing and jointing-heading stages, but medium-scale size or more severe floods still led to the decrease of maize yield. The high yield could be achieved in the slightly wet years in period of heading-milking stage, while the decrease could be caused by flood when it was severely wet. This research can provide a reference for estimating the impact of drought and flood disasters on spring maize and taking disaster prevention measures in three northeast provinces of China.
To satisfy the communication needs of multiple robots working in orchards, an improved Ad Hoc on-demand distance vector routing protocol based on signal strength threshold and priority nodes (AODV-SP), and the prediction model of Wi-Fi signal reception in peach orchards, was proposed in this study. Different from the traditional AODV protocol, AODV-SP utilizes the idea of priority nodes and strength thresholds to construct a discovery routing algorithm and a selection routing algorithm by seeking priority nodes and calculating the maximum strength threshold between nodes, respectively. The discovery routing message and selection routing message of the AODV-SP protocol were designed according to the discovery routing and selection routing algorithms. To verify the performance of the AODV-SP protocol, the performance of the protocol with different maximum movement speeds of nodes was analyzed by using NS2 simulation software and the performance was compared with the traditional AODV protocol. The simulation results showed that the average end-to-end delay, route initiation frequency, and route overhead of AODV-SP protocol with the introduction of priority node and path signal strength thresholds were smaller than those of the traditional AODV protocol, and the packet delivery rate improved significantly compared with that of AODV protocol. Among them, when the maximum node movement speed was 5 m/s, the route initiation frequency and route overhead of AODV-SP protocol reduced by 3.65% and 7.09%, respectively, compared with AODV protocol. When the maximum node movement speed was 8 m/s, the packet delivery rate of AODV-SP protocol improved by 0.59% and the average end-to-end delay reduced by 13.09%. To further verify the simulation results of AODV-SP making AODV-SP protocol applicable to a multi-robot wireless communication system and ensure the normal operation of multi-robot wireless communication in orchards, a physical platform for multi-robot wireless communication was built in a laboratory environment, and software was designed to enable the physical platform to communicate properly under the AODV-SP protocol. And the physical platform for multi-robot wireless communication using the AODV-SP protocol was tested under static and dynamic conditions, respectively. The experiment results showed that, under static condition, when distance between nodes was less than or equal to 25 m, the packet loss rate of the robot was 0; when distance between nodes was 100 m, tthe packet loss rate of the robot was 21.01%, and the following robots could maintain the chain topology with the leader robot in dynamic conditions. Simulation and physical platform experiments results showed that the AODV-SP protocol could be used for the construction of multi-robot communication systems in orchard.
The development of convolutional neural networks(CNN) has brought a large number of network parameters and huge model volumes, which greatly limites the application on devices with small computing resources, such as single-chip microcomputers and mobile devices. In order to solve the problem, a structured model compression method was studied in this research. Its core idea was using knowledge distillation to transfer the knowledge from the complex integrated model to a lightweight small-scale neural network. Firstly, VGG16 was used to train a teacher model with a higher recognition rate, whose volume was much larger than the student model. Then the knowledge in the model was transfered to MobileNet by using distillation. The parameters number of the VGG16 model was greatly reduced. The knowledge-distilled model was named Distilled-MobileNet, and was applied to the classification task of 38 common diseases (powdery mildew, Huanglong disease, etc.) of 14 crops (soybean, cucumber, tomato, etc.). The performance test of knowledge distillation on four different network structures of VGG16, AlexNet, GoogleNet, and ResNet showed that when VGG16 was used as a teacher model, the accuracy of the model was improved to 97.54%. Using single disease recognition rate, average accuracy rate, model memory and average recognition time as 4 indicators to evaluate the accuracy of the trained Distilled-MobileNet model in a real environment, the results showed that, the average accuracy of the model reached 97.62%, and the average recognition time was shortened to 0.218 s, only accounts for 13.20% of the VGG16 model, and the model size was reduced to only 19.83 MB, which was 93.60% smaller than VGG16. Compared with traditional neural networks, distilled-mobile model has a significant improvement in reducing size and shorting recognition time, and can provide a new idea for disease recognition on devices with limited memory and computing resources.
With the development of smart agriculture, automatic question and answer (Q&A) of agricultural knowledge is needed to improve the efficiency of agricultural information acquisition. Agriculture named entity recognition plays a key role in automatic Q&A system, which helps obtaining information, understanding agriculture questions and providing answer from the knowledge graph. Due to the scarcity of labeled ANE data, some existing open agricultural entity recognition models rely on manual features, can reduce the accuracy of entity recognition. In this work, an approach of model distillation was proposed to recognize agricultural named entity data. Firstly, massive agriculture data were leveraged from Internet, an agriculture knowledge graph (AgriKG) was constructed. To overcome the scarcity of labeled named agricultural entity data, weakly named entity recognition label on agricultural texts crawled from the Internet was built with the help of AgriKG. The approach was derived from distant supervision, which was used to solve the scarcity of labeled relation extraction data. Considering the lack of labeled data, pretraining language model was introduced, which is fine tuned with existing labeled data. Secondly, large scale pretraining language model, BERT was used for agriculture named entity recognition and provided a pretty well initial parameters containing a lot of basic language knowledge. Considering that the task of agriculture named entity recognition relied heavily on low-end semantic features but slightly on high-end semantic features, an Attention-based Layer Aggregation mechanism for BERT(BERT-ALA) was designed in this research. The aim of BERT-ALA was to adaptively aggregate the output of multiple hidden layers of BERT. Based on BERT-ALA model, Bidirectional LSTM (BiLSTM) and conditional random field (CRF) were coupled to further improve the recognition precision, giving a BERT-ALA+BiLSTM+CRF model. Bi-LSTM improved BERT's insufficient learning ability of the relative position feature, while conditional random field models the dependencies of entity recognition label. Thirdly, since BERT-ALA+BiLSTM+CRF model was difficult to serve online because of the extremely high time and space complexity, BiLSTM+CRF model was used as student model to distill BERT-ALA+BiLSTM+CRF model. It fitted the BERT-ALA+BiLSTM+CRF model's output of BiLSTM layer and CRF layer. The experiment on the database constructed in the research, as well as two open datasets showed that (1) the macro-F1 of the BERT-ALA + BiLSTM + CRF model was improved by 1% compared to the baseline model BERT + BiLSTM + CRF, and (2) compared with the model trained on the original data, the macro-F1 of the distilled student model BiLSTM + CRF was increased by an average of 3.3%, the prediction time was reduced by 33%, and the storage space was reduced by 98%. The experimental results verify the effectiveness of the BERT-ALA and knowledge distillation in agricultural entity recognition.
Maize is an important food crop in China. In order to quickly and non-destructively estimate summer maize leaf area index (LAI) under different water stress conditions, in this study, maize samples with multiple irrigation treatments throughout the growth period were used for modeling analysis. Then, based on the unmanned aerial vehicle (UAV) multi-spectral remote sensing technology, combined with the summer maize LAI collected in the field during the same period, five kinds of vegetation indices, including the normalized difference vegetation index (NDVI), soil adjusted vegetation index (SAVI), enhanced vegetation index (EVI), green normalized difference vegetation index (GNDVI) and visible atmospherically resistant index (VARI) were selected in this research as model input parameters, and random forest regression algorithm was used to establish the relationship between the field maize canopy vegetation indices and LAI under different irrigation conditions during the entire growth period. The accuracies of the model were compared with that of the model established by the university linear regression and multiple linear regression algorithms. The results showed that under sufficient irrigation condition, the vegetation index using multiple linear regression model could well (R2 = 0.83, RMSE = 0.05) estimate LAI; under water stress conditions, the vegetation index using random forest regression model could well estimate LAI (R2 = 0.74~0.87, RMSE = 0.02~0.10), water stress factors had little effect on the random forest regression model, and NDVI and VARI contributed the LAI estimation model better. The spatial distribution map of LAI was generated based on the random forest regression algorithm. The above results showed that it was feasible to use the random forest regression algorithm to estimate the summer maize LAI under various irrigation conditions based on the UAV multi-spectral remote sensing technology. The results indicates that the model established has a good applicability. This research can provide technical and method support for the rapid and accurate monitoring of field summer maize LAI under different irrigation conditions during the entire growth period.
Soil organic matter (SOM) is an important source of crop growth, its content can reflect soil fertility status. In order to realize the fast and real-time estimation of the SOM, based on hyperspectral data, a rapid estimation model of SOM content in orchards was established. A total of 100 brown soil samples were collected from the apple orchard of Qixia county, Yantai city, Shandong province. After drying and grinding, the hyper-spectrum of the soil was measured in the laboratory using ASD FieldSpec. The spectral data was preprocessed by the method of moving average, and the spectral reflectance features of orchard soil were analyzed to study the correlation between spectral reflectance and its soil organic matter content. In order to enhance the correlation between relevant spectral parameters and soil indexes, the original data were processed by using the multivariate scattering correction, the first derivative and the first derivative of MSC. After the sensitive wavelengths of soil organic matter content were selected and the spectral indexes were constructed. Multiple linear regression models (MLR), support vector machines (SVM) and random forest (RF) models were respectively established. The estimation accuracy of the orchard soil organic matter estimation model was measured by the determination coefficient (R2), root mean square error (RMSE) and relative analysis error (RPD). The sensitive wavelengths of soil organic matter content selected were 678, 709, 1931, 1939, 1996 and 2201 nm. The spectral parameters were constructed using the selected wavelengths, which were NDSI(678, 709), NDSI(678, 1931), NDSI(678, 2201), NDSI(709, 1939), and NDSI(1939, 2201). These models established include MLR, SVM and RF model. The RF model had the best precision. The calibration sample R2 was 0.8804, the RMSE was 0.1423 and RPD reached 2.25; the R2 of the verification model was 0.7466, the RMSE was 0.1266, and the RPD was 1.79. The results showed that the fitting effect of the hyperspectral inversion model based on RF regression analysis was better than that based on MLR analysis and SVM regression analysis. As a promising and effective method, RF can play a vital role in predicting soil organic matter. The results can help understanding the distribution of soil nutrients, guiding farmers to apply fertilizer reasonably and improving the efficiency of orchard production and management.
In recent years, revenue protection crop insurance is an innovative insurance that has been prioritized in China. But it still lacks the support of the third-party yield data around crop harvest time. Aiming to provide objective yield data for revenue protection crop insurance, satellite remote sensing production estimation technology was employed to discuss its application mode and applicability. Taking the soybean revenue protection insurance in Jiaxiang county, Shandong province as an example, we first extracted soybean planting plots, calculated vegetation index and crop physiological parameters based on Sentinel-2 satellite images in 2018 . Combining to TRMM precipitation data from TRMM precipitation-monitoring radar satellite and MODIS land surface temperature data from Terra/Aqua satellite and site yield data, we established a multi-parameter linear regression model, and estimated soybean yield per unit area. The crop extraction results showed that the soybean planting area in the study area was 1.24 km2, which was in good agreement with the 1.27 km2 reported by the local agricultural bureau; and with using the actual measurement plots, the remote sensing identification accuracy of the planting distribution plots reached 90%. The yield estimation results showed that the NDVI of the soybean pod stage on August 23 and the leaf area index of the soybean seedling stage on September 7 explained the soybean yield per hectare the best, and the average estimated yield of the whole area was 244,500 kg/m2, which reflects the severely affected agricultural conditions, comparing to 299,800 kg/km2 in previous years.The regression coefficient between the estimated yield data and the measured data reached 0.92, which meet the application needs.With this results, the estimated yield of different towns can be summarized, and the regional yield was present, and was used as the real yield in 2018, multiplying with the average soybean price around October 11 to December 10 from the local price bureau, the real revenue was obtained. Compared the real revenue to the expected revenue in the contract of insurance, the claims work was decided. The results indicated that the Sentinel-2 satellite data could be used to identify the soybean planting distribution in the study area accurately, and to complete the yield estimation as soon as one week after the soybean harvest, which could guide the insurance company's claims work. The whole methodology is capable of aiding the claims work in revenue protection crop insurance.
Leaf wetness duration (LWD) is one of the important input variables of plant disease model, which is related to the infection of many leaf pathogens and affects the pathogen infection and developmental rate. In order to accurately predict the occurrence time and location of cucumber diseases in solar greenhouse, nine sampling points were set up in two different greenhouses located in Beijing in March and September 2019, according to the chessboard method to deploy temperature, humidity and light sensors. The fixed-point visual inspection method was used to collect the data every 1 h. From the leaf wetting to the leaf drying is the leaf wetness duration of a day. The relative humidity model (RHM) and back propagation neural network model (BPNN) were used to quantitatively estimate and analyze the LWD, the input layer of BPNN was temperature, humidity, radiation and location, the hidden layer was 10, and the output layer was location and whether the leaf surface was wet. The results showed that BPNN obtained similar accuracy ACC = 0.90 and 0.92 under the experimental conditions of two greenhouses, which was higher than RHM ACC = 0.82 and 0.84 in estimating of LWD, the mean absolute errors MAE were 1.81 h and 1.61 h, root mean squared error RMSE were 2.10 and 1.87, and coefficient of determination R2 were 0.87 and 0.85. In sunny and cloudy conditions, the spatial distribution of LWD was generally in the South > the Middle > the North. In the South, the average LWD was the longest, 12.17 h/d; from the east to the west, the spatial distribution of LWD was generally in the East > the West > the Middle. In the Middle, the average LWD was the shortest of 4.83h/d. The average LWD in rainy days was longer than that in sunny days and cloudy days, the average LWD in spring and autumn rainy days were 17.15 h/d and 17.41 h/d. These changes and differences had an important impact on the distribution of leaf wetness duration in the horizontal direction of cucumber population in greenhouse, which was closely related to the occurrence rule of most high humidity cucumber diseases. In this research, the method of regional analysis of the wet duration of cucumber leaves in greenhouse was proposed, which could provide a reference for simulating the spatial distribution of LWD in greenhouse, and also had a certain reference significance for the establishment of cucumber disease early warning system.
In order to study the development mode and realization way of smart agriculture, the technical route of agricultural information fusion of agricultural machinery in different production stages before, during and after wheat production was designed. Pre-production: use Beidou precision navigation technology and motion planning optimization method to realize the full area coverage path planning of the field operation of the automatic navigation tractor, combine the laser leveling equipment to realize the accurate and standardized land leveling and laser leveling, and realize the accurate and standardized operation of the land. On this basis, the spatial interpolation technology was used to make the variable fertilization prescription map and combining variable rate fertilizer machine and realized variable rate precise application of fertilizer and precise seeding. At the same time, combining with the optimal design of planting scheme, based on the prenatal database and knowledge base, it optimizes the decision-making of variety configuration and sowing time and seeding amount were optimized, and the software intelligent decision-making technology was used to recommend the varieties and sowing time and seeding amount suitable for planting at the decision-making point, and constructs the wheat and maize prenatal information service recommendation system based on WebGIS was constructed. In production: based on the image technology of automatic segmentation and color feature extraction of wheat image in the field environment, a remote monitoring model of wheat nutritional status with the function of wheat population image segmentation and nutritional estimation was established to realize the non-destructive monitoring of wheat nutritional status in the field environment. After production, the integrated measurement sensor, speed sensor, header height sensor and GPS were adopted, and controller area network bus was adopted with wireless communication technology, a real-time wheat yield measurement system was developed, which was installed on a large-scale combine harvester to carry out the real-time prediction service of wheat yield, so as to realize the synchronous process of wheat harvest and yield measurement, with the error less than 5%. The intelligent transformation of common agricultural machinery equipment and the research and development of sowing and harvesting equipment adapted to agricultural production were completed and realized, and the small scale with high-efficiency utilization of light and heat resources, increase of output and green development were studied. The model of wheat planting production was optimized .A real time measurement and prediction system for postpartum yield was developed, which included the selection of sowing date, fertilization recommendation, seedling growth and nutrition diagnosis. The experimental results show that the adoption of agricultural information fusion technology can increase wheat yield by 18.4%, input-output ratio by 16.6% and 7.9%, which shows that the intelligent agriculture of Henan province is effective and feasible.
Agriculture of China is a typical agricultural system of small producer and large market, producers are too scattered. Agricultural foundation remains weak in the vast rural areas, especially poverty-stricken areas. Blockchain technology has good complementarity and applicability with China's agricultural products trading system, because of its distributed storage, transaction information transparency and product information traceability. However, the agricultural product trading system has characteristics of product diversity, commercial process complexity, user group widespread, decentralized, privacy protection and so on. It is difficult to apply the traditional blockchain technology directly to China's agricultural products trading information network. In view of the above problems, the design idea of alliance chain was adopted, and the technology of multi-chain agricultural product transaction information, which includes transaction information blockchain, user information blockchain and agricultural products information blockchain was put forward. The product information blockchain provided the detailed information of agricultural products and guarante that the traceability and non-tamperability of the information. The blockchain node access mechanism was introduced in the user information chain to provide real-name voucher registration and management functions for the agricultural product trading platform. The transaction information blockchain recorded the results of all transaction smart contracts, and through the addition of channel technology, different transaction information could be isolated from each other, which could meet the privacy protection of transaction information and user data and the rapid processing of transaction data. The profit of the transaction was automatically divided by the smart contract, which improved the efficiency of execution and reduces the transaction cost. Finally, a transparent, efficient and applicable blockchain framework for agricultural product transactions was established.
