基于改进轻量化YOLOv4模型的虾只肉壳辨识方法  

Improved lightweight YOLOv4 model-based method for the identification of shrimp flesh and shell

作　　者：陈学深[1] 吴昌鹏 党佩娜 梁俊 刘善健 武涛[1] CHEN Xueshen;WU Changpeng;DANG Peina;LIANG Jun;LIU Shanjian;WU Tao(College of Engineering,South China Agricultural University,Guangzhou 510642,China)

机构地区：[1]华南农业大学工程学院,广州510642

出　　处：《农业工程学报》2023年第9期278-286,共9页Transactions of the Chinese Society of Agricultural Engineering

基　　金：广东省重点领域研发计划项目(2021B0202060002)。

摘　　要：为实现虾只机械剥壳环节裸肉虾与带壳虾自动分选,该研究提出一种基于改进YOLOv4模型的虾只肉壳辨识方法。将YOLOv4模型中CSP-Darknet53网络替换为GhostNet网络,增强模型自适应特征提取能力及简化模型参数计算量。在YOLOv4主干特征提取网络Resblock模块中引入轻量级注意力机制,增强主干特征提取网络的特征提取能力。将YOLOv4模型中GIoU损失函数替换为CIoU损失函数,提高模型预测框的回归效果。为检测改进效果进行了不同模型对比验证,轻量化结果表明改进YOLOv4模型参数量最少、计算量最小;消融试验表明改进YOLOv4模型的平均精度均值为92.8%,比YOLOv4模型提升了6.1个百分点。不同场景下应用改进YOLOv4模型进行虾只肉壳辨识性能试验。结果表明:同品种不同环境的虾只肉壳辨识总体平均准确率为95.9%,同品种不同剥壳方式的虾只肉壳辨识准确率平均值为90.4%,不同品种虾只肉壳辨识准确率平均值为87.2%。研究结果可为裸肉虾与带壳虾自动分选提供技术支撑。An improved lightweight YOLOv4 model was proposed to realize the accurate,real-time,and robust automatic sorting of bare and shelled shrimp in the shrimp mechanical shelling process under complex scenarios.The CSP-Darknet53 network was replaced by the GhostNet in the YOLOv4 structure.The ability of the model was then improved to extract the features adaptively.The calculation of model parameters was also simplified after improvement.The GhostNet network was used for the YOLOv4 backbone feature extraction,in order to reduce the network model complexity,and the model parameters for better storage capacity and detection efficiency.A lightweight attention mechanism was introduced into the Resblock module of the YOLOv4 backbone feature extraction network,in order to enhance the feature extraction capability of the backbone feature extraction network.The SE attention mechanism module was used to enhance the attention between feature channels.The attention of the network model was improved to the shrimp shell by fitting the relevant feature information to the target channel and suppressing invalid information.The model recognition accuracy was improved to reduce background interference.The original GIoU loss function was replaced with a CIoU loss function to improve the regression effect of the prediction frame.The CIoU loss function made the data obtained from non-maximal suppression more reasonable and efficient.Furthermore,the prediction frame was more accurate to minimize the distance between the centroids of the detection frame and the labelled frame.The lightweight GhostNet-YOLOv4 model was compared with the YOLOv7,EfficientNet Lite3-YOLOv4,ShuffleNetV2-YOLOv4,and MobilenetV3-YOLOv4 models.The results showed that the GhostNet-YOLOv4 model shared the lowest number of parameters and computational effort.An ablation comparison experiment was designed to verify that replacing the backbone feature extraction network and embedding the SE attention mechanism optimized for the module.The replacement of the CSP-Darknet53 b

关 键 词：机器视觉 目标检测 虾只 深度学习 YOLOv4 

分 类 号：S985.2[农业科学—捕捞与储运]