Transplant Status Detection Algorithm of Cabbage in the Field Based on Improved YOLOv8s

doi:10.12133/j.smartag.SA202401008

Abstract

Abstract:

[Objective] Currently, the lack of computerized systems to monitor the quality of cabbage transplants is a notable shortcoming in the agricultural industry, where transplanting operations play a crucial role in determining the overall yield and quality of the crop. To address this problem, a lightweight and efficient algorithm was developed to monitor the status of cabbage transplants in a natural environment. [Methods] First, the cabbage image dataset was established, the cabbage images in the natural environment were collected, the collected image data were filtered and the transplanting status of the cabbage was set as normal seedling (upright and intact seedling), buried seedling (whose stems and leaves were buried by the soil) and exposed seedling (whose roots were exposed), and the dataset was manually categorized and labelled using a graphical image annotation tool (LabelImg) so that corresponding XML files could be generated. And the dataset was pre-processed with data enhancement methods such as flipping, cropping, blurring and random brightness mode to eliminate the scale and position differences between the cabbages in the test and training sets and to improve the imbalance of the data. Then, a cabbage transplantation state detection model based on YOLOv8s (You Only Look Once Version 8s) was designed. To address the problem that light and soil have a large influence on the identification of the transplantation state of cabbage in the natural environment, a multi-scale attention mechanism was embedded to increase the number of features in the model, and a multi-scale attention mechanism was embedded to increase the number of features in the model. Embedding the multi-scale attention mechanism to increase the algorithm's attention to the target region and improve the network's attention to target features at different scales, so as to improve the model's detection efficiency and target recognition accuracy, and reduce the leakage rate; by combining with deformable convolution, more useful target information was captured to improve the model's target recognition and convergence effect, and the model complexity increased by C3-layer convolution was reduced, which further reduced the model complexity. Due to the unsatisfactory localization effect of the algorithm, the focal extended intersection over union loss (Focal-EIoU Loss) was introduced to solve the problem of violent oscillation of the loss value caused by low-quality samples, and the influence weight of high-quality samples on the loss value was increased while the influence of low-quality samples was suppressed, so as to improve the convergence speed and localization accuracy of the algorithm. [Results and Discussions] Eventually, the algorithm was put through a stringent testing phase, yielding a remarkable recognition accuracy of 96.2% for the task of cabbage transplantation state. This was an improvement of 2.8% over the widely used YOLOv8s. Moreover, when benchmarked against other prominent target detection models, the algorithm emerged as a clear winner. It showcased a notable enhancement of 3% and 8.9% in detection performance compared to YOLOv3-tiny. Simultaneously, it also managed to achieve a 3.7% increase in the recall rate, a metric that measured the efficiency of the algorithm in identifying actual targets among false positives. On a comparative note, the algorithm outperformed YOLOv5 in terms of recall rate by 1.1%, 2% and 1.5%, respectively. When pitted against the robust faster region-based convolutional neural network (Faster R-CNN), the algorithm demonstrated a significant boost in recall rate by 20.8% and 11.4%, resulting in an overall improvement of 13%. A similar trend was observed when the algorithm was compared to the single shot multibox detector (SSD) model, with a notable 9.4% and 6.1% improvement in recall rate. The final experimental results show that when the enhanced model was compared with YOLOv7-tiny, the recognition accuracy was increased by 3%, and the recall rate was increased by 3.5%. These impressive results validated the superiority of the algorithm in terms of accuracy and localization ability within the target area. The algorithm effectively eliminates interferenced factors such as soil and background impurities, thereby enhancing its performance and making it an ideal choice for tasks such as cabbage transplantation state recognition. [Conclusions] The experimental results show that the proposed cabbage transplantation state detection method can meet the accuracy and real-time requirements for the identification of cabbage transplantation state, and the detection accuracy and localization accuracy of the improved model perform better when the target is smaller and there are weeds and other interferences in the background. Therefore, the method proposed in this study can improve the efficiency of cabbage transplantation quality measurement, reduce the time and labor, and improve the automation of field transplantation quality survey.

Key words: transplantation of cabbage, YOLOv8s, target detection, multi-scale attention, deformable convolution

WU Xiaoyan, GUO Wei, ZHU Yiping, ZHU Huaji, WU Huarui. Transplant Status Detection Algorithm of Cabbage in the Field Based on Improved YOLOv8s[J]. Smart Agriculture, 2024, 6(2): 107-117.

Figures/Tables 12

Fig. 1

Fig. 2

Fig. 3

Fig.4

Fig. 5

Fig. 6

Table 1

Fig. 7

Fig. 8

Table 2

Fig. 9

Table 3

References 26

1	崔志超, 管春松, 杨雅婷, 等. 蔬菜机械化移栽技术与装备研究现状[J]. 中国农机化学报, 2020, 41(3): 85-92.
	CUI Z C, GUAN C S, YANG Y T, et al. Research status of vegetable mechanical transplanting technology and equipment[J]. Journal of Chinese agricultural mechanization, 2020, 41(3): 85-92.
2	蒋展. 油菜毯状苗移栽机栽植质量在线监测系统设计与试验[D]. 北京: 中国农业科学院, 2021.
	JIANG Z. Design and experiment of on-line monitoring system for planting quality of rape blanket seedling transplanter[D].Beijing: Chinese Academy of Agricultural Sciences, 2021.
3	赵德安, 赵璜晔. 基于CNN算法的缺秧与漂秧图像识别技术研究[J]. 软件导刊, 2020, 19(8): 230-233.
	ZHAO D A, ZHAO H Y. Image recognition technology of seedling-lacking and drifting seedlings based on CNN algorithms[J]. Software guide, 2020, 19(8): 230-233.
4	王传宇, 郭新宇, 肖伯祥, 等. 基于图像拼接的苗期玉米植株缺失数量自动测量方法[J]. 农业工程学报, 2014, 30(12): 148-153.
	WANG C Y, GUO X Y, XIAO B X, et al. Automatic measurement of numbers of maize seedlings based on mosaic imaging[J]. Transactions of the Chinese society of agricultural engineering, 2014, 30(12): 148-153.
5	蒋展, 张敏, 吴俊, 等. 油菜毯状苗移栽漏栽实时监测方法—基于视频图像拼接[J]. 农机化研究, 2022, 44(9): 189-195.
	JIANG Z, ZHANG M, WU J, et al. Real-time monitoring method for rape blanket seedling transplanting and omission based on video image SSplicing[J]. Journal of agricultural mechanization research, 2022, 44(9): 189-195.
6	AL-QANESS M A A, ABBASI A A, FAN H, et al. An improved YOLO-based road traffic monitoring system[J]. Computing, 2021, 103(2): 211-230.
7	DEGADWALA S, VYAS D, CHAKRABORTY U, et al. Yolo-v4 deep learning model for medical face mask detection[C]// 2021 International Conference on Artificial Intelligence and Smart Systems (ICAIS). Piscataway, New Jersey, USA: IEEE, 2021: 209-213.
8	PENG H, ZHANG Y F, YANG S, et al. Battlefield image situational awareness application based on deep learning[J]. IEEE intelligent systems, 2020, 35(1): 36-43.
9	张秀花, 静茂凯, 袁永伟, 等. 基于改进YOLOv3-Tiny的番茄苗分级检测[J]. 农业工程学报, 2022, 38(1): 221-229.
	ZHANG X H, JING M K, YUAN Y W, et al. Tomato seedling classification detection using improved YOLOv3-Tiny[J]. Transactions of the Chinese society of agricultural engineering, 2022, 38(1): 221-229.
10	LIU Z D, WANG X, ZHENG W X, et al. Design of a sweet potato transplanter based on a robot arm[J]. Applied sciences, 2021, 11(19): ID 9349.
11	LI Y M, ZHU Y J, LI S S, et al. The extraction method of navigation line for cuttage and film covering multi-functional machine for low tunnels[J]. Inventions, 2022, 7(4): ID 113.
12	PERUGACHI-DIAZ Y, TOMCZAK J M, BHULAI S. Deep learning for white cabbage seedling prediction[J]. Computers and electronics in agriculture, 2021, 184: ID 106059.
13	LI Z B, LI Y, YANG Y B, et al. A high-precision detection method of hydroponic lettuce seedlings status based on improved Faster RCNN[J]. Computers and electronics in agriculture, 2021, 182: ID 106054.
14	HASAN A S M M, SOHEL F, DIEPEVEEN D, et al. A survey of deep learning techniques for weed detection from images[J]. Computers and electronics in agriculture, 2021, 184: ID 106067.
15	王明, 张倩. 中国基于深度学习的图像识别技术在农作物病虫害识别中的研究进展[J]. 中国蔬菜, 2023(3): 22-28.
	WANG M, ZHANG Q. Research progress of image recognition technology based on depth learning in identification of pest and disease in crops in China[J]. China vegetables, 2023(3): 22-28.
16	杨文庆, 刘天霞, 唐兴萍, 等. 智慧农业背景下的植物表型组学研究进展[J]. 河南农业科学, 2022, 51(7): 1-12.
	YANG W Q, LIU T X, TANG X P, et al. Research progress on plant phenomics in the context of smart agriculture[J]. Journal of Henan agricultural sciences, 2022, 51(7): 1-12.
17	LOU H T, DUAN X H, GUO J M, et al. DC-YOLOv8: Small-size object detection algorithm based on camera sensor[J]. Electronics, 2023, 12(10): ID 2323.
18	ZHENG Z H, WANG P, LIU W, et al. Distance-IoU loss: Faster and better learning for bounding box regression[J]. Proceedings of the AAAI conference on artificial intelligence, 2020, 34(7): 12993-13000.
19	OUYANG D L, HE S, ZHANG G Z, et al. Efficient multi-scale attention module with cross-spatial learning[C]// ICASSP 2023-2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). Piscataway, New Jersey, USA: IEEE, 2023: 1-5.
20	WANG W H, DAI J F, CHEN Z, et al. InternImage: exploring large-scale vision foundation models with deformable convolutions[C]// 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Piscataway, New Jersey, USA: IEEE, 2023: 14408-14419..
21	YANG Z M, WANG X L, LI J G. EIoU: An improved vehicle detection algorithm based on VehicleNet neural network[J]. Journal of physics: Conference series, 2021, 1924(1): ID 012001.
22	ZHANG Y F, REN W Q, ZHANG Z, et al. Focal and efficient IOU loss for accurate bounding box regression[J]. Neurocomputing, 2022, 506(C): 146-157.
23	REDMON J, FARHADI A. YOLOv3: An incremental improvement[EB/OL]. arXiv:1804.02767, 2018.
24	GIRSHICK R. Fast R-CNN[C]// 2015 IEEE International Conference on Computer Vision (ICCV). Piscataway, New Jersey, USA: IEEE, 2015: 1440-1448.
25	WANG C Y, BOCHKOVSKIY A, LIAO H Y M. YOLOv7: trainable bag-of-freebies sets new state-of-the-art for real-time object detectors[C]// 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Piscataway, New Jersey, USA: IEEE, 2023: 7464-7475.
26	LIU W, ANGUELOV D, ERHAN D, et al. SSD: single shot MultiBox detector[M]// Computer Vision-ECCV 2016. Cham: Springer International Publishing, 2016: 21-37.

	EMA	C3_DCNV3	Focal-EIoU	AP/%			R/%	FLOPs/G	mAP_0.5/%	T1/FPS
	EMA	C3_DCNV3	Focal-EIoU	埋苗	露苗	正常苗	R/%	FLOPs/G	mAP_0.5/%	T1/FPS
YOLOv8s	×	×	×	89.5	90.5	91.3	88.8	28.9	93.8	156
	√	×	×	88.5	93.6	92.0	90.2	29.1	94.2	148
	×	√	×	89.9	93.4	90.2	89.6	28.0	94.4	151
	×	×	√	90.5	94.9	93.6	88.9	28.9	94.6	149
	√	√	×	89.0	94.3	93.1	88.3	28.5	95.9	150
	√	×	√	88.7	93.9	92.8	90.6	29.1	94.8	147
	×	√	√	90.6	94.8	93.5	91.5	28.0	95.8	153
	√	√	√	91.2	95.3	94.1	92.2	28.5	96.2	146

网络模型	检测数/株			漏检数/株
网络模型	正常苗	埋苗	露苗	正常苗	埋苗	露苗
YOLOv8s	208	42	85	7	12	7
YOLOv8s-FDN	213	51	90	2	3	2

网络模型	模型大小/MB	R/%	mAP_0.5/%
YOLOv3-tiny	36.6	86.7	87.3
YOLOv5	24.4	88.9	94.2
Faster R-CNN	108.2	77.4	84.8
YOLOv7-tiny	12.3	89.9	92.7
SSD	101.0	84.4	90.1
YOLOv8s-FDN	22.6	90.4	96.2

[1]	JIN Xuemeng, LIANG Xiyin, DENG Pengfei. Lightweight Daylily Grading and Detection Model Based on Improved YOLOv10 [J]. Smart Agriculture, 2024, 6(5): 108-118.
[2]	YANG Lin, LIU Shuangyin, XU Longqin, HE Min, SHENG Qingfeng, HAN Jiawei. Dynamic Prediction Method for Carbon Emissions of Cold Chain Distribution Vehicle under Multi-Source Information Fusion [J]. Smart Agriculture, 2024, 6(4): 138-148.
[3]	ZHANG Yuyu, BING Shuying, JI Yuanhao, YAN Beibei, XU Jinpu. Grading Method of Fresh Cut Rose Flowers Based on Improved YOLOv8s [J]. Smart Agriculture, 2024, 6(2): 118-127.