Table of Content

30 December 2021, Volume 3 Issue 4

Previous Issue    Next Issue

Topic--Agricultural Products Processing and Testing

Recent Advances on Application of Radio Frequency Heating in the Research of Post-Harvest Grain Storage and Processing | Open Access

LI Hongyue, LI Qingluan, ZHENG Jianjun, LING Bo, WANG Shaojin

2021, 3(4):  1-13.  doi:10.12133/j.smartag.2021.3.4.202106-SA001

Asbtract ( 788 )   HTML ( 69)   PDF (983KB) ( 966 )  

References | Related Articles | Metrics

The storage and processing of grain are the basis for economic and social stability and development. As a new heating treatment technology based on electromagnetic wave, radio frequency technology has the characteristics of large penetration depth, rapid heating, volumetric heating and no chemical residue. It has been widely used in post-harvest research of grain and has potential industrial application prospects in some fields. To expound the research progress of the application of radio frequency heating technology in grain storage and processing, this review briefly described the basic principle and characteristics of radio frequency heating as well as the current commercial radio frequency heating system including free oscillation type and 50 Ω type. The basic research of radio frequency heating in grain storage and processing was summarized from three aspects: Dielectric properties of grain and pests, heat resistance of stored grain pests and heating uniformity of sample. The dielectric properties refer to the interaction between materials and electromagnetic waves in an electromagnetic field and determines the absorption and conversion of electromagnetic energy. It can predict the heating characteristics of grain and provide basic data for computer simulation to optimize process during radio frequency treatment. The heat resistance data of pests are necessary for the establishment and optimization of dis-infestations technology, so the kinetic date of thermal death of common stored grain pests were reported in this review. As a main hinder in the commercial application of radio frequency treatments, the heating uniformity has significant effect on heat treatment quality and results in potential food safety problems. The major factors causing heating non-uniformity are the non-uniformity of electromagnetic field, runaway heating and the sample shape effect. The improvement methods of heating uniformity were summarized from three aspects in this article including changing the electromagnetic field distribution, sample position, and optimizing the radio frequency working parameters. Based on the above basic research of radio frequency technology and combining with the practical problems in grain storage and processing, the applications of radio frequency heating in the fields of dis-infestations, sterilizing, enzyme inactivation and drying were also summarized. Finally, some suggestions on the application of this technology in grain storage and processing and future research directions were proposed. This review may play a certain guiding role for the application of radio frequency technology in grain storage and processing.

Research Progress of Sensing Detection and Monitoring Technology for Fruit and Vegetable Quality Control | Open Access

GUO Zhiming, WANG Junyi, SONG Ye, ZOU Xiaobo, CAI Jianrong

2021, 3(4):  14-28.  doi:10.12133/j.smartag.2021.3.4.202106-SA011

Asbtract ( 1381 )   HTML ( 168)   PDF (959KB) ( 1729 )  

References | Related Articles | Metrics

Vegetable and fruit planting areas and products of China have always ranked first in the world, and the vegetable and fruit industry is respectively the second and third largest agricultural planting industry after grain. Vegetables and fruits are prone to quality deterioration during postharvest storage and transportation, resulting in reduced edible value and huge economic losses. To ensure fruit and vegetable quality and reduce the waste of resources caused by postnatal deterioration, this paper summarizes the latest research status of sensor detection and monitoring technology for fruit and vegetable quality deterioration and analyzed the principle, characteristics, advantages, and disadvantages of various detection technology. Among them, machine vision can detect the external quality and surface defects of fruits and vegetables, but fruits and vegetables are different from the standard machined products, and they are affected by many factors in the growth process, which seriously interfere with the image collection work and easily lead to misjudgment. An electronic nose equipped with expensive gas sensors can monitor the odor deterioration of fruits and vegetables but would require improved sensitivity and durability. Near-infrared can detect the internal quality and recessive defects of fruits and vegetables, but the applicability of the model needs to be improved. Hyperspectral imaging can visually detect the internal and external quality of fruits and vegetables and track the deterioration process, but the huge amount of data obtained leads to data redundancy, which puts forward higher requirements for system hardware. Therefore, low-cost multispectral imaging systems should be developed and characteristic wavelength extraction algorithms should be optimized. Raman spectroscopy can detect fruit and vegetable spoilage bacteria and their metabolites, but there is no effective Raman enhanced substrate production and accurate Raman standard spectrogram database. The comprehensive evaluation of fruit and vegetable deterioration can be realized by multi-technology and multi-information fusion. It can overcome the limitation of single sensor information analysis, improve the robustness and parallel processing ability of the detection model, and provide a new approach for high-precision detection or monitoring of fruit and vegetable quality deterioration. The Internet of Things monitoring system is constructed with various sensors as the sensing nodes to realize the intelligent real-time monitoring of fruit and vegetable quality deterioration information, provide a reference for solving the technical limitation of quality deterioration control in the processing of fruit and vegetable. This is of great significance for reducing the postpartum economic loss of fruits and vegetables and promoting the sustainable development of the fruit and vegetable industry.

Research Progress and Application Prospect of Electronic Nose Technology in the Detection of Meat and Meat Products | Open Access

LIU Yang, JIA Wenshen, MA Jie, LIANG Gang, WANG Huihua, ZHOU Wei

2021, 3(4):  29-41.  doi:10.12133/j.smartag.2021.3.4.202011-SA003

Asbtract ( 699 )   HTML ( 41)   PDF (1178KB) ( 809 )  

References | Related Articles | Metrics

With the continuous increase of import and export of various countries, people have put forward higher requirements on the efficiency and accuracy of meat and meat products safety indicators detection. Since electronic nose technology is simple to operate and allows rapid and nondestructive testing, it can meet today's need for efficient test of meat and meat products. In this paper, the detection principle of electronic nose technology was introduced firstly, and its development process was described from two aspects of hardware and software system. Then, the application research progress of electronic nose technology in meat and meat products detection in recent years from the aspects of freshness, adulteration, flavor evaluation and microbial contamination of meat and meat products was analyzed. Different electronic nose instruments and equipment or different pattern recognition algorithms result in different analysis results. Therefore, it highlighting the feasibility and advancement of electronic nose technology application in various aspects of meat and meat products detection. At the same time, in view of the application research results of electronic nose technology in the detection of meat and meat products, the paper pointed out the shortcomings of electronic nose technology, for example: The analysis effect of electronic nose technology was uneven, the price of electronic nose equipment was relatively expensive, and the application range of large electronic nose equipment was limited. Therefore, there were still some difficulties and problems of electronic nose technology in the aspects of universality and popularization. Finally, in view of the shortcomings of the current electronic nose technology, the development and application prospects of the electronic nose technology in the future were prospected. In terms of hardware system, with the research and development continuously of new gas sensitive materials, the durability and sensitivity to smell recognition of the electronic nose by improving the performance of the electrode film material of the electronic nose sensor array was enhanced. In terms of software system, with the upgrading continuously of computer systems, a supporting platform for the emerging and complex pattern recognition algorithms was provided. New pattern recognition algorithms in the pattern recognition system of electronic nose technology were explored and introduced, so that electronic nose technology can achieve faster and more accurate recognition and analysis of odors.

Rapid Detection of Imazalil Residues in Navel Orange Peel Using Surface-Enhanced Raman Spectroscopy | Open Access

ZHANG Sha, LIU Muhua, CHEN Jinyin, ZHAO Jinhui

2021, 3(4):  42-52.  doi:10.12133/j.smartag.2021.3.4.202106-SA002

Asbtract ( 565 )   HTML ( 24)   PDF (1209KB) ( 480 )  

References | Related Articles | Metrics

Imazalil, a preservative for navel orange in the process of postharvest processing, is easy to seep into the flesh through the peel and produce residues in the flesh, which is vulnerable to cause endanger to human body if it was eaten accidentally. Base on this, a fast detection method of imazalil residues in navel orange peel ,namely surface-enhanced Raman spectroscopy (SERS) was proposed in this study. Firstly, the SERS detection conditions of imazalil residues in navel orange peel were optimized, and the optimal detection conditions were determined as follows: Reaction time of 2 min, gold colloid of 400 μL, NaBr as electrolyte solution, NaBr dosage of 25 μL. Based on the above optimal conditions, 6 groups of spectral data processed by adaptive iterative penalized least squares (air PLS), air PLS combination with normalization, air PLS combination with baseline correction, air PLS combination with first derivative, air PLS combination with standard normal distribution (SNV), air PLS combination with multiplicative scatter correction (MSC) were used to establish support vector regression (SVR) models and compare the models prediction performance. And air PLS method was selected as the spectral pretreatment method, because the value of correlation coefficient computed value of prediction set (R_P) is the largest, and the value of root mean square error calculated value of the prediction set (RMSEP) is the smallest. Then, principal component analysis (PCA) was used to extract the features from spectral data, and the first seven principal component scores were selected as the input values of SVR prediction model. SVR, multiple linear regression (MLR) and partial least squares regression (PLSR) were used to analyze and compare the prediction performances. The R_P value of prediction set of SVR prediction model could reach 0.9156, the RMSEP value of their prediction set was 4.8407 mg/kg, and the relative standard deviation computation value (RPD) was 2.3103, which indicated that the closer the predicted value of imazalil residue on navel orange surface based on SVR algorithm was to the measured value, the more effective the prediction accuracy of the model could be. The above data indicated that the speedy detection of imazalil residues in navel orange peel could be emploied by SERS coupled with PCA and SVR modeling method.

Detection of Peel Puffing and Granulation in Citrus Based on Soft X-ray Imaging Technology | Open Access

XU Qian, CAI Jianrong, DU Can, SUN Li, BAI Junwen

2021, 3(4):  53-65.  doi:10.12133/j.smartag.2021.3.4.202106-SA009

Asbtract ( 556 )   HTML ( 39)   PDF (1718KB) ( 562 )  

References | Related Articles | Metrics

The internal quality of citrus is an important index for citrus grading, and the most common factors affecting the internal quality of citrus are peel puffing and granulation, which affect the fruit quality and lose the market value due to the large consumption of nutrients. In this study, a soft X-ray imaging device was developed, including a transmission device, a soft X-ray imaging device, a trigger device and a soft X-ray protection device, for the problem that traditional methods cannot detect citrus peel puffing and granulation efficiently and non-destructively. In this research, the detection parameters were determined according to the physical characteristics of wide peeled citrus, and the clarity, contrast and aberration rate of citrus images were used as the judging criteria. The best imaging parameters were determined by adjusting the parameters of the imaging device as follows: The tube voltage of X-ray source was 60 kV, the tube current was 1.3 mA, the integration time of line array detector was 5.5 ms, and the transmission speed of citrus conveyor belt was 10 cm/s. The aberrations in the column direction were detected by the circular hole metal plate, and the results showed that the transmission speed was stable and the aberrations in the column direction were negligible. The aberrations in the row direction ware detected by using the 70 mm stainless steel calibration sphere, and the projection aberration coefficients at different positions in the row direction were calculated, and the aberration correction model was established. The soft X-ray images of citrus were acquired under the above parameters, and Gaussian filtering was used to denoise the citrus images. The image enhancement algorithm was used to enhance the contrast of the denoised images. The fixed threshold segmentation method and morphological algorithm were fused to remove the background area, separate the flesh area and the peel area of the citrus images. Finally, the area percentage method was used to calculate the ratio of citrus flesh area to citrus fruit area to discriminate the degree of citrus peel puffing; the grayscale features of citrus fruit area were extracted to obtain the citrus withered area, and the ratio of citrus withered area to citrus flesh area was calculated to discriminate the degree of citrus granulation. Qingjiang Ponkan were taken as the experimental object, and the results showed that the overall recognition rate of the homemade soft X-ray imaging device were 96.2% and 86.9% for the peel puffing and granulation of Qingjiang Ponkan, respectively. The method proposed in this study may achieve nondestructive detection of peel puffing and granulation inside citrus.

Detection Method of Apple Mould Core Based on Dielectric Characteristics | Open Access

LI Dongbo, HUANG Lyuwen, ZHAO Xubo

2021, 3(4):  66-76.  doi:10.12133/j.smartag.2021.3.4.202102-SA035

Asbtract ( 664 )   HTML ( 26)   PDF (1520KB) ( 560 )  

References | Related Articles | Metrics

Apple mouldy core disease often occurs in the ventricle of apples and cannot be effectively identified by appearance. Near-infrared spectroscopy, nuclear magnetic resonance and other methods are usually used in traditional apple mouldy core disease detection, but these methods require complex equipment and high detection costs. In this research, a simple and fast nondestructive detection method of apple mouldy core disease was proposed by using a dielectric method to construct an apple mouldy core disease detection model. Japan's Hioki 3532-50 LCR tester was used to collect 108 dielectric indicators (12 dielectric indicators at 9 frequencies) of 220 apples as the original data. Due to the large differences in the distribution of data collected with different dielectric indexes and different frequencies, a standardized method was used for data preprocessing to eliminate the problem of large differences in dielectric data distribution. Afterwards, in order to eliminate the redundant information between the data, the principal component analysis algorithm was used to reduce the data dimensionality, and finally the three algorithms of BP neural network (BPNN), support vector machine (SVM) and random forest (RF) were used to construct the mouldy core disease detection model. After pre-experiment, the most effective parameters of each algorithm were selected, the test results showed that the apple mouldy core disease detection model based on the RF algorithm obtained the best performance, and the detection accuracy rate reached 96.66% and 95.71% in the training set (150 apples) and the test set (70 apples). The mouldy core disease detection model constructed by using BPNN was the second most effective, and the detection accuracy could reach 94.66% and 94.29%, respectively. The detection effect of the model built by using SVM was relatively poor, and the detection accuracies were 93.33% and 91.43%, respectively. The experimental results showed that the model constructed by using RF can more effectively identify mouldy core disease apples and healthy apples. This study could provide references for apple diseases and insect pests and non-destructive testing of apple quality.

Construction of Milk Purchase Classification Model Based on Shuffled Frog Leaping Algorithm and Support Vector Machine | Open Access

XIAO Shijie, WANG Qiaohua, LI Chunfang, ZHAO Limei, LIU Xinya, LU Shiyu, ZHANG Shujun

2021, 3(4):  77-85.  doi:10.12133/j.smartag.2021.3.4.202107-SA003

Asbtract ( 470 )   HTML ( 12)   PDF (1063KB) ( 389 )  

References | Related Articles | Metrics

Protein, fat and somatic cells are three important reference indicators in milk purchase, which determine the quality and price of milk. The traditional chemical analysis methods of these indexes are time-consuming and pollute the environment, while the mid-infrared spectrum has the advantages of fast, non-destructive and simple operation. In order to realize the rapid classification of milk quality and improve the production efficiency of dairy enterprises, 3216 Holstein milk samples were chosen as the research objects and mid-infrared spectroscopy technology was applied to realize the detection and classification of 4 different quality milks during the purchase process. The spectrum was preprocessed by using the first derivative and the first difference, and combined with the algorithm competitive adaptive reweighted sampling (CARS) and the shuffled frog leaping algorithm (SFLA), the effective characteristic variables that could represent different milks were selected, and the SVM model was established. Among them, the penalty parameter c and the kernel function parameter g which were the key parameters of the SVM model were optimized by using the grid search method (GS), genetic algorithm (GA) and particle swarm algorithm (PSO). The training time of GS, GA and PSO algorithms were compared, the results showed that the training time of GS was much longer than that of GA and PSO algorithms.The SFLA algorithm was generally better than the CARS algorithm, and the PSO optimized the SVM model the best. After the first-order difference preprocessing, the PSO-SVM established by using the SFLA algorithm to filter the characteristic variables, the accuracy of the training set, the accuracy of the test set and the AUC were 97.8%, 95.6% and 0.96489, respectively. This model has a high accuracy rate and has practical application value in the milk industry.

Intelligent Management and Control

Multi-Factor Coordination Control Technology of Promoting Early Maturing in Southern Blueberry Intelligent Greenhouse | Open Access

XU Lihong, LIU Huihui, XU He, WEI Ruihua, CAI Wentao

2021, 3(4):  86-98.  doi:10.12133/j.smartag.2021.3.4.202109-SA007

Asbtract ( 758 )   HTML ( 46)   PDF (1965KB) ( 1030 )  

References | Related Articles | Metrics

In order to get blueberries goes on sale in advance and obtain greater economic benefits, southern blueberries were moved to an intelligent greenhouse with controllable environment for experimental production. The early maturing production control technology of southern blueberry intelligent greenhouse was explored and studied. First, a detailed and comprehensive investigation and summary were conducted on the production factors of blueberry soilless cultivation, such as the production characteristics of various blueberry varieties, the pH and composition of the substrate, the key points of water and fertilizer irrigation, and the scope of the microenvironment climate. Then, the existing Venlo-type greenhouse was deployed for blueberry production, and the geography, climate and internal structural conditions of the greenhouse were briefly described, and the greenhouse blueberry full-cycle control goal was planned. Finally, the production control system was designed and implemented based on the Internet of Things technology, and the overall framework of the software layer, the hardware layer and the cloud were introduced. Based on multi-factor coordinated control model of greenhouse environment, according to the characteristics of blueberry growth environment, a set of blueberry greenhouse multi-factor coordinated control algorithms were proposed and used for environmental regulation. The experimental greenhouse is located in the southeast of Huaqiao Town, Kunshan city, Suzhou city, Jiangsu province. It has been verified that the overall control system has a significant effect, and the first wave of fruits was harvested in early May 2021, making the southern variety of blueberry enter the fruit picking period nearly one month earlier. Compared with the blueberry plants without cold storage, the yields per plant of "Star" "Emerald" "Lanmei No. 1", and "Coast" after cold storage increased by 51.5%, 85.5%, 43.8%, and 94.7%, respectively, and the weight of each fruit was increased 10.9%, 7.2%, 2.6%, and 5.3%, respectively. Experiments proved that the use of multi-factor coordinated control algorithms for regulation can increase the yield and quality of blueberries and achieve significant economic benefits and provide a demonstration for the industrialization of blueberry plants in southern greenhouses to promote early maturity production and management.

Information Processing and Decision Making

Strawberry Growth Period Recognition Method Under Greenhouse Environment Based on Improved YOLOv4 | Open Access

LONG Jiehua, GUO Wenzhong, LIN Sen, WEN Chaowu, ZHANG Yu, ZHAO Chunjiang

2021, 3(4):  99-110.  doi:10.12133/j.smartag.2021.3.4.202109-SA006

Asbtract ( 848 )   HTML ( 79)   PDF (2697KB) ( 652 )  

References | Related Articles | Metrics

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.

Fast Counting Method of Soybean Seeds Based on Density Estimation and VGG-Two | Open Access

WANG Ying, LI Yue, WU Tingting, SUN Shi, WANG Minjuan

2021, 3(4):  111-122.  doi:10.12133/j.smartag.2021.3.4.202101-SA002

Asbtract ( 680 )   HTML ( 57)   PDF (1631KB) ( 833 )  

References | Related Articles | Metrics

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.