基于网络瘦身算法的YOLOv4-tiny的甘蔗茎节识别

doi:10.13733/j.jcam.issn.2095-5553.2023.02.024

摘要/Abstract

摘要： 为提高智能甘蔗收获的准确性，降低算法对部署的高算力要求，利用轻量级目标检测算法YOLOv4-tiny相对YOLOv4算法更简化的网络结构、更高的推理速度等优点，提出基于MobileNet和网络瘦身的两种YOLOv4-ting识别算法方案，并比较二者的精度和模型复杂度。其中，基于网络瘦身算法的YOLOv4-tiny在精度较瘦身前(947%)下降0.6%的情况下，模型复杂度下降为原来的1/3，即瘦身后的FLOPs和Params分别为1.1 G和1 789 658。而以MobileNet为Backbone的YOLOv4-tiny在精度下降1.92%的情况下，它的FLOPs和Params为1.29 G、2 600 068，其在精度和模型复杂度上的表现都不如瘦身后的YOLOv4-tiny模型。结果表明：基于网络瘦身算法的YOLOv4-tiny甘蔗茎节识别模型可有效降低模型复杂度，其计算量对嵌入式设备和移动式设备友好。该研究可为智能甘蔗收割机构的开发提供技术参考。

关键词: MobileNet, 网络瘦身, YOLOv4-tiny, 甘蔗茎节

Abstract: In order to improve the accuracy of intelligent sugarcane harvesting, reduce the high computing power requirement for the algorithm deployment, the recognition algorithm of YOLOv4-tiny based on MobileNet and the recognition algorithm of YOLOv4-tiny based on network slimming are proposed, which make use of the advantages of the lightweight target detection algorithm YOLOv4-tiny compared with YOLOv4 algorithm, such as more simplified network structure and higher reasoning speed. The accuracy and complexity of the two models are compared. When the average precision of YOLOv4-tiny based on network slimming algorithm was only 0.6% lower than that before slimming (947%), its model complexity was reduced to 1/3 of the original, that was, the FLOPs and Params after slimming were 11 G and 1 789 658. However, when the average precision of YOLOv4-tiny with MobileNet as the backbone decreased by 192%, its FLOPs and Params were 1.29 G and 2 600 068. The performance of the latter in average precision and model complexity was not as good as that of the YOLOv4-tiny model after slimming. The results showed that the YOLOv4-tiny sugarcane stem node recognition model based on network slimming algorithm could effectively reduce the complexity of the model, and its computation was friendly to embed devices and mobile devices. The research can provide a technical method for the development of the intelligent sugarcane harvesting machine.

中图分类号:

S566.1

陈文, 余康, 李岩舟, 陈远玲, 胡珊珊, 乔曦. 基于网络瘦身算法的YOLOv4-tiny的甘蔗茎节识别[J]. 中国农机化学报, 2023, 44(2): 172-181.

Chen Wen, Yu Kang, Li Yanzhou, Chen Yuanling, Hu Shanshan, Qiao Xi.. Recognition of sugarcane stem node based on network slimming algorithm and YOLOv4-tiny algorithm[J]. Journal of Chinese Agricultural Mechanization, 2023, 44(2): 172-181.

参考文献

［1］　
Moshashai K, Almasi M, Minaei S, et al. Identification of sugarcane nodes using image processing and machine vision technology ［J］. International Journal of Agricultural Research, 2008, 3(5): 357-364.

［2］　
陆尚平, 文友先, 葛维, 等. 基于机器视觉的甘蔗茎节特征提取与识别［J］. 农业机械学报, 2010, 41(10): 190-194.

Lu Shangping, Wen Youxian, Ge Wei, et al. Recognition and features extraction of sugarcane nodes based on machine vision ［J］. Transactions of the Chinese Society for Agricultural Machinery, 2010, 41(10): 190-194.

［3］　
黄亦其, 乔曦, 唐书喜, 等. 基于Matlab的甘蔗茎节特征分布定位与试验［J］. 农业机械学报, 2013, 44(10): 93-97, 232.

Huang Yiqi, Qiao Xi, Tang Shuxi, et al. Localization and test of characteristics distribution for sugarcane internode based on Matlab ［J］. Transactions of the Chinese Society for Agricultural Machinery, 2013, 44(10): 93-97, 232.

［4］　
李尚平, 李向辉, 张可, 等. 改进YOLOv3网络提高甘蔗茎节实时动态识别效率［J］. 农业工程学报, 2019, 35(23): 185-191.

Li Shangping, Li Xianghui, Zhang Ke, et al.Increasing the realtime dynamic identification efficiency of sugarcane nodes by improved YOLO v3 network ［J］. Transactions of the Chinese Society of Agricultural Engineering, 2019, 35(23): 185-191.

［5］　
Meng Y, Ye C, Yu S, et al. Sugarcane node recognition technology based on wavelet analysis ［J］. Computers and Electronics in Agriculture, 2019, 158: 68-78.

［6］　
Zhou D, Fan Y, Deng G, et al. A new design of sugarcane seed cutting systems based on machine vision ［J］. Computers and Electronics in Agriculture, 2020, 175: 105611.

［7］　
Chen J, Wu J, Qiang H, et al. Sugarcane nodes identification algorithm based on sum of local pixel of minimum points of vertical projection function ［J］. Computers and Electronics in Agriculture, 2021, 182: 105994.

［8］　
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.

［9］　
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.

［10］　
Wong A, Famuori M, Shafiee M J, et al. YOLO Nano: A highly compact you only look once convolutional neural network for object detection ［C］. 2019 Fifth Workshop on Energy Efficient Machine Learning and Cognitive ComputingNeurIPS Edition (EMC2-NIPS). IEEE, 2019: 22-25.

［11］　
Jiang Z, Zhao L, Li S, et al. Realtime object detection method based on improved YOLOv4-tiny ［J］. arXiv preprint arXiv, 2011.04244, 2020.

［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］　
Howard A G, Zhu M, Chen B, et al. MobileNets: Efficient convolutional neural networks for mobile vision applications ［J］. arXiv preprint arXiv, 1704.04861, 2017.

［14］　
张宏鸣, 汪润, 董佩杰, 等. 基于DeepSORT算法的肉牛多目标跟踪方法［J］. 农业机械学报, 2021, 52(4): 248-256.

Zhang Hongming, Wang Run, Dong Peijie, et al. Beef cattle multitarget tracking based on DeepSORT algorithm ［J］. Transactions of the Chinese Society for Agricultural Machinery, 2021, 52(4): 248-256.

［15］　
高洋, 陈万米, 林城. 基于SSD-MobileNet的投篮机器人目标识别算法［J］. 工业控制计算机, 2021, 34(6): 51-53.

Gao Yang, Chen Wanmi, Lin Cheng. Target recognition algorithm of basketshooting robot based on SSD-MobileNet ［J］. Industrial Control Computer, 2021, 34(6): 51-53.

［16］　
Sandler M, Howard A, Zhu M, et al. MobileNetv2: Inverted residuals and linear bottlenecks ［C］. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2018: 4510-4520.

［17］　
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.

［18］　
Howard A, Sandler M, Chu G, et al. Searching for MobileNetV3 ［C］. Proceedings of the IEEE/CVF International Conference on Computer Vision, 2019: 1314-1324.

［19］　
Hu J, Shen L, Sun G. Squeezeandexcitation networks ［C］. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2018: 7132-7141.

［20］　
Liu Z, Li J, Shen Z, et al. Learning efficient convolutional networks through network slimming ［C］. Proceedings of the IEEE International Conference on Computer Vision, 2017: 2736-2744.

［21］　
Ma N, Zhang X, Zheng H T, et al. ShuffleNet v2: Practical guidelines for efficient cnn architecture design ［C］. Proceedings of the European Conference on Computer Vision (ECCV), 2018: 116-131.

［22］　
Tian H, Wang T, Liu Y, et al. Computer vision technology in agricultural automation—A review ［J］. Information Processing in Agriculture, 2020, 7(1): 1-19.

［23］　
Wu D, Lü S, Jiang M, et al. Using channel pruningbased YOLOv4 deep learning algorithm for the realtime and accurate detection of apple flowers in natural environments ［J］. Computers and Electronics in Agriculture, 2020, 178: 105742.

［24］　
景亮, 王瑞, 刘慧, 等. 基于双目相机与改进YOLOv3算法的果园行人检测与定位［J］. 农业机械学报, 2020, 51(9): 34-39, 25.

Jing Liang, Wang Rui, Liu Hui, et al. Orchard pedestrian detection and location based on binocular camera and improved YOLOv3 algorithm ［J］. Transactions of the Chinese Society for Agricultural Machinery, 2020, 51(9): 34-39, 25.

［25］　
崔博超, 郑江华, 刘忠军, 等. 无人机遥感影像的YOLOv3鼠洞识别技术［J］. 林业科学, 2020, 56(10): 199-208.

Cui Bochao, Zheng Jianghua, Liu Zhongjun, et al. YOLOv3 mouse hole recognition based on remote sensing images from technology for unmanned aerial vehicle ［J］. Scientia Silvae Sinicae, 2020, 56(10): 199-208.

［26］　
彭红星, 黄博, 邵园园, 等. 自然环境下多类水果采摘目标识别的通用改进SSD模型［J］. 农业工程学报, 2018, 34(16): 155-162.

Peng Hongxing, Huang Bo, Shao Yuanyuan, et al. General improved SSD model for picking object recognition of multiple fruits in natural environment ［J］. Transactions of the Chinese Society of Agricultural Engineering, 2018, 34(16): 155-162.