基于改进YOLOv3的复杂环境下西红柿成熟果实快速识别

doi:10.13733/j.jcam.issn.2095-5553.2023.08.024

摘要/Abstract

摘要： 针对西红柿成熟果实快速识别问题，采集并标注西红柿成熟果实的图像数据集，用于深度神经网络模型的训练，并基于实际应用对经典的YOLOv3目标检测算法进行模型轻量化改进，使其方便地部署到采摘机器人的嵌入式控制器上，同时对激活函数、锚框的聚类、非极大值抑制和损失函数等进行优化，提高算法运行的效率和稳定性。经测试集的验证，所提出的改进型YOLOv3目标检测算法在包括不同密集程度、不同光照条件和不同遮挡程度情况的复杂环境下最终检测精度为92.11%，召回率为86.21%，F1得分为89%，mAP为84.58%，即试验结果证明所提方法的可行性、准确性和鲁棒性。

关键词: 西红柿, 复杂环境, 深度神经网络, YOLOv3目标检测, 果实识别

Abstract: Aiming at the problem of fast recognition of ripe tomato fruits, an image dataset of ripe tomato fruit is collected and labeled for deep neural network model training. In addition, the target detection algorithm of the classic algorithm YOLOv3 is improved in light weight based on practical applications, so that it can be easily deployed on the embedded controller of tomatopicking robot. The activation function, clustering of anchor frame, nonmaximum suppression and loss function are also optimized, the efficiency and the stability of the algorithm is also improved. Through verification of the test set, the proposed improved YOLOv3 target detection algorithm can effectively recognize tomato ripe fruit in complex environment, including different density, different lighting conditions and even different occlusion degree. The final detection accuracy was 92.11%, recall rate was 86.21%, F1 score was 89%, mAP was 84.58%. Thus the experimental results demonstrate the feasibility, accuracy and robustness of the proposed method.

Key words: tomato, complex environment, deep neural network, YOLOv3 target detection, recognition of fruits

中图分类号:

S24： TP391.4

高芳征, 汤文俊, 陈光明, 黄家才. 基于改进YOLOv3的复杂环境下西红柿成熟果实快速识别[J]. 中国农机化学报, 2023, 44(8): 174-183.

Gao Fangzheng, Tang Wenjun, Chen Guangming, , Huang Jiacai. Fast recognition of ripe tomato fruits in complex environment based on improved YOLOv3[J]. Journal of Chinese Agricultural Mechanization, 2023, 44(8): 174-183.

参考文献

［1］　刘凡, 杨光友, 杨康. 农业采摘机器人柔性机械手研究［J］. 中国农机化学报, 2019, 40(3): 173-178.
Liu Fan, Yang Guangyou, Yang Kang. Research on flexible manipulator for agricultural picking robot ［J］. Journal of Chinese Agricultural Mechanization, 2019, 40(3): 173-178.
［2］　初广丽, 张伟, 王延杰, 等. 基于机器视觉的水果采摘机器人目标识别方法［J］. 中国农机化学报, 2018, 39(2): 83-88.Chu Guangli, Zhang Wei, Wang Yanjie, et al. A method of fruit picking robot target identification based on machine vision ［J］. Journal of Chinese Agricultural Mechanization, 2018, 39(2): 83-88.
［3］　Koirala A, Walsh K B, Wang Z, et al. Deep learningmethod overview and review of use for fruit detection and yield estimation ［J］. Computers and Electronics in Agriculture, 2019, 162: 219-234.
［4］　Sa I, Ge Z, Dayoub F, et al. Deepfruits: A fruit detection system using deep neural networks ［J］. Sensors, 2016, 16(8): 1222.
［5］　Hamuda E, Mc Ginley B, Glavin M, et al. Improved image processingbased crop detection using Kalman filtering and the Hungarian algorithm ［J］. Computers and Electronics in Agriculture, 2018, 148: 37-44.
［6］　Linker R, Cohen O, Naor A. Determination of the number of green apples in RGB images recorded in orchards ［J］. Computers and Electronics in Agriculture, 2012, 81: 45-57.
［7］　Liu G, Nouaze J C, Touko Mbouembe P L, et al. YOLOtomato: A robust algorithm for tomato detection based on YOLOv3 ［J］. Sensors, 2020, 20(7): 2145.
［8］　Girshick R. FastRCNN ［C］. Proceedings of the IEEE International Conference on Computer Vision, 2015: 1440-1448.
［9］　Ren S, He K, Girshick R, et al. FasterRCNN: Towards realtime object detection with region proposal networks ［J］. arXiv preprint arXiv: 2017, 39(6).
［10］　Redmon J, Farhadi A. Yolov3: An incremental improvement ［J］. arXiv preprint arXiv: 2018. 1804.02767.
［11］　Liu W, Anguelov D, Erhan D, et al. SSD: Single shot multibox detector ［C］. European conference on computer vision. Springer, Cham, 2016: 21-37.
［12］　Bochkovskiy A, Wang C Y, Liao H Y M. Yolov4: Optimal speed and accuracy of object detection ［J］. arXiv preprint arXiv: 2004. 10934, 2020.
［13］　Tian Y, Yang G, Wang Z, et al. Apple detection during different growth stages in orchards using the improved YOLO-V3 model ［J］. Computers and Electronics in Agriculture, 2019, 157: 417-426.
［14］　赵德安, 吴任迪, 刘晓洋, 等. 基于YOLO深度卷积神经网络的复杂背景下机器人采摘苹果定位［J］. 农业工程学报, 2019, 35(3): 164-173.Zhao Dean, Wu Rendi, Liu Xiaoyang, et al. Apple positioning based on YOLO deep convolutional neural network for picking robot in complex background ［J］. Transactions of the Chinese Society of Agricultural Engineering, 2019, 35(3): 164-173.
［15］　刘芳, 刘玉坤, 林森, 等. 基于改进型YOLO的复杂环境下番茄果实快速识别方法［J］. 农业机械学报, 2020, 51(6): 229-237.Liu Fang, Liu Yukun, Lin Sen, et al. Fast recognition method for tomatoes under complex environments based on improved YOLO ［J］. Transactions of the Chinese Society for Agricultural Machinery, 2020, 51(6): 229-237.
［16］　薛月菊, 黄宁, 涂淑琴, 等. 未成熟芒果的改进YOLOv2识别方法［J］. 农业工程学报, 2018, 34(7): 173-179.Xue Yueju, Huang Ning, Tu Shuqin, et al. Immature mango detection based on improved YOLOv2 ［J］. Transactions of the Chinese Society of Agricultural Engineering, 2018, 34(7): 173-179.
［17］　吕梦棋, 张芮祥, 贾浩, 等. 基于改进ResNet玉米种子分类方法研究［J］. 中国农机化学报, 2021, 42(4): 92-98.
Lü Mengqi, Zhang Ruixiang, Jia Hao, et al. Research on seed classification based on improved ResNet ［J］. Journal of Chinese Agricultural Mechanization, 2021, 42(4): 92-98.
［18］　Simonyan K, Zisserman A. Very deep convolutional networks for largescale image recognition ［J］. arXiv preprint arXiv: 2014. 1409.1556.
［19］　Sengupta A, Ye Y, Wang R, et al. Going deeper in spiking neural networks: VGG and residual architectures ［J］. Frontiers in Neuroscience, 2019, 13: 95.
［20］　Wen L, Li X, Gao L. A transfer convolutional neural network for fault diagnosis based on ResNet-50 ［J］. Neural Computing and Applications, 2020, 32(10): 6111-6124.
［21］　Szegedy C, Ioffe S, Vanhoucke V, et al. Inceptionv4, inceptionresnet and the impact of residual connections on learning ［C］. Thirtyfirst AAAI Conference on Artificial Intelligence. 2017.
［22］　He K, Zhang X, Ren S, et al. Deep residual learning for image recognition ［C］. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2016: 770-778.
［23］　Wu Z, Shen C, Hengel A V D. Wider or deeper: Revisiting the resnet model for visual recognition ［J］. Pattern Recognition, 2019, 90: 119-133.

[1]	张磊, 刘琪芳, 聂红玫, 王晨, 牛帆. 基于改进YOLOV4网络模型的番茄果实检测[J]. 中国农机化学报, 2022, 43(12): 162-169.
[2]	杨英茹, 吴华瑞, 张燕, 朱华吉, 李瑜玲, 田国英, . 基于复杂环境的番茄叶部图像病虫害识别[J]. 中国农机化学报, 2021, 42(9): 177-186.
[3]	李小敏, 张日红, 陈天赐, 侯炳法, 张权, 李雄, . 基于深度学习的林下落果识别方法与试验[J]. 中国农机化学报, 2021, 42(9): 202-208.
[4]	刘丽娟;窦佩佩;王慧;. 自然环境下重叠与遮挡苹果图像识别方法研究[J]. 中国农机化学报, 2021, 42(6): 174-181.
[5]	冯俊惠;李志伟;戎有丽;孙志立;. 基于改进Hough圆变换算法的成熟番茄果实识别[J]. 中国农机化学报, 2021, 42(4): 190-196.
[6]	杨保海;任全会;李海生;. 复杂环境下果园机器人路径规划方法研究[J]. 中国农机化学报, 2021, 42(2): 134-138.
[7]	张星;高巧明;潘栋;张伟伟;. 基于改进YOLOv3的田间复杂环境下菠萝拾捡识别研究[J]. 中国农机化学报, 2021, 42(1): 201-206.
[8]	谢万里;李宏志;周辉;尹绍武;. 基于迁移学习与卷积神经网络的鱼濒死预警系统研究[J]. 中国农机化学报, 2019, 40(2): 186-192.