| 改进YOLOv5s和迁移学习对番茄穴盘苗的分级检测 |
| |
| 引用本文： | 李韬,任玲,胡斌,王双,赵明,张玉泉,杨苗.改进YOLOv5s和迁移学习对番茄穴盘苗的分级检测[J].农业工程学报,2023,39(23):174-184. |
| |
| 作者姓名： | 李韬 任玲 胡斌 王双 赵明 张玉泉 杨苗 |
| |
| 作者单位： | 1. 石河子大学机械电气工程学院, 石河子 832003;2. 农业农村部西北农业装备重点实验室, 石河子 832003 |
| |
| 基金项目： | 石河子大学高层次人才项目（RCZK2021B17）；石河子大学自主资助支持科研项目（ZZZC202105） |
| |
| 摘 要： | 为了提高番茄穴盘苗分级检测精度，该研究提出了改进YOLOv5s目标检测模型，并通过迁移学习对番茄穴盘病苗识别精度进行优化。采用轻量级网络EfficientNetv2的Backbone部分作为特征提取网络，保留YOLOv5s中的SPPF空间金字塔池化模块，压缩模型参数数量以减少计算量；更改模型Neck部分原始上采样模块为CARAFE轻量级上采样模块，在引入很少参数量的情况下提高模型精度；同时将PANet替换为BiFPN，引入特征权重信息，增强不同尺度特征融合能力；引入有效多尺度注意力机制(efficient multi-scale attention,EMA)，提高对番茄苗的关注，减少背景干扰；替换CIoU损失函数为SIoU损失函数，考虑真实框与预测框之间的方向匹配，提高模型收敛效果。试验结果表明，改进的YOLOv5s目标检测模型经过迁移学习训练后，平均精度均值达到95.6%，较迁移学习前提高了0.7个百分点；与原YOLOv5s模型相比，改进YOLOv5s模型平均精度均值提升2.6个百分点；改进YOLOv5s模型的参数量、计算量和权重大小分别为原YOLOv5s模型的53.1%、20.0%...
|
| 关 键 词： | 机器视觉 分级检测 番茄穴盘苗 EfficientNetv2 迁移学习 YOLOv5s |
| 收稿时间： | 2023/8/10 0:00:00 |
| 修稿时间： | 2023/12/9 0:00:00 |
Grading detection of tomato hole-pan seedlings using improved YOLOv5s and transfer learning |
| |
| LI Tao,REN Ling,HU Bin,WANG Shuang,ZHAO Ming,ZHANG Yuquan,YANG Miao.Grading detection of tomato hole-pan seedlings using improved YOLOv5s and transfer learning[J].Transactions of the Chinese Society of Agricultural Engineering,2023,39(23):174-184. |
| |
| Authors: | LI Tao REN Ling HU Bin WANG Shuang ZHAO Ming ZHANG Yuquan YANG Miao |
| |
| Institution: | 1. College of Mechanical and Electrical Engineering, Shihezi University, Shihezi 832003, China;2. Key Laboratory of Northwest Agricultural Equipment, Ministry of Agriculture and Rural Affairs, Shihezi 832003, China |
| |
| Abstract: | Grading is one of the most important steps in the tomato hole-pan seedlings. In this study, an improved YOLOv5s target model was proposed to optimize the recognition accuracy of tomato hole-pan diseased seedlings using transfer learning. The backbone part of the lightweight network (EfficientNetv2) was used as the feature extraction. The spatial pyramid pooling fusion (SPPF) module in YOLOv5s was retained to compress the number of model parameters, in order to reduce the amount of computation; The lightweight up-sampling module of CARAFE was used to introduce a small number of parameters in the Neck part of the model; And the PANet was replaced with BiFPN to optimize the accuracy of tomato hole-pan seedling identification using transfer learning. The feature weight information was introduced to enhance the fusion of features at different scales; The efficient multi-scale attention mechanism (EMA) was introduced to increase the attention for the tomato hole-pan seedlings, and reduce the background interference; The CIoU loss function was replaced by the SIoU loss function to improve the model accuracy. Direction matching between the real and predicted frame was considered to improve the convergence of the model. The mean average accuracy of the improved YOLOv5s target model reached 95.6% after training by transfer learning, which was 0.7 percentage points higher than before; The improved YOLOv5s model was 53.1% of the number of parameters in the original model, and the computation was 20.0% of the original with only 3.2G, while the weights were 53.6% of the original and the weight size was only 7.35MB, and the mean average accuracy was improved by 2.6 percentage points, compared with the original; The visualization show that the stronger feature extraction was achieved in the improved model. The feature weights were centrally distributed in the center of the burrow and the edge of the burrow holes. Meanwhile, the heat map drawn with the GradCam showed that the improved YOLOv5s model was focused mainly on the burrow seedling itself and the edge of the burrow hole, which reduced the interference of the substrate; The ablation test verified that the improved YOLOv5s model was performed the best in the detection accuracy and the frame rate, compared with the Faster-RCNN, CenterNet, and YOLO series. The findings can provide a strong reference for the grading detection and subsequent deployment of hole tray seedlings. |
| |
| Keywords: | machine vision grading detection tomato cavity seedling EfficientNetv2 Transfer learning YOLOv5s |
|
| 点击此处可从《农业工程学报》浏览原始摘要信息 |
| 点击此处可从《农业工程学报》下载免费的PDF全文 |
|
