| 面向复杂田间收获作业的轻量化三七目标检测方法 |
| |
| 引用本文： | 王法安,何忠平,张兆国,解开婷,房汉玉.面向复杂田间收获作业的轻量化三七目标检测方法[J].农业工程学报,2024,40(8):133-143. |
| |
| 作者姓名： | 王法安 何忠平 张兆国 解开婷 房汉玉 |
| |
| 作者单位： | 昆明理工大学现代农业工程学院，昆明 650500;云南省高校中药材机械化工程研究中心，昆明 650500;云南省高校中药材机械化工程研究中心，昆明 650500;昆明理工大学机电工程学院，昆明 650500;黑龙江佳木斯汤原县农业技术推广中心，佳木斯154000 |
| |
| 基金项目： | 国家重点研发计划项目（2022YFD2002004）；云南省基础研究专项-青年基金项目：（202401AU070196） |
| |
| 摘 要： | 针对目前三七检测算法在复杂田间收获工况下检测精度低、模型复杂度大、移动端部署难等问题，该研究提出一种基于YOLOv5s的轻量化三七目标检测方法。首先，采用GSConv卷积方法替换原始颈部网络的传统卷积，引入Slim-neck轻量级颈部网络，降低了模型复杂度，同时提升了模型精度；其次，使用ShuffleNetv2轻量型特征提取网络对主干网络进行轻量化改进，提升了模型实时检测性能，并采用角度惩罚度量的损失（SIoU）优化边界框损失函数，提升了轻量化后的模型精度和泛化能力。试验结果表明，改进后的PN-YOLOv5s模型参数量、计算量、模型大小分别为原YOLOv5s模型的46.65%、34.18%和48.75%，检测速度提升了1.2倍，F1值较原始模型提升了0.22个百分点，平均精度均值达到了94.20%，较原始模型低0.6个百分点，与SSD、Faster R-CNN、YOLOv4-tiny、YOLOv7-tiny和YOLOv8s模型相比能够更好地平衡检测精度与速度，检测效果更好。台架试验测试结果表明，4种输送分离作业工况下三七目标检测的准确率达90%以上，F1值达86%以上，平均精度均值达87%以上，最低检测速度为105帧/s，实际收获工况下模型的检测性能良好，可为后续三七收获作业质量实时监测与精准分级输送提供技术支撑。
|
| 关 键 词： | 神经网络 目标检测 轻量化 复杂收获作业工况 三七 YOLOv5s |
| 收稿时间： | 2024/1/17 0:00:00 |
| 修稿时间： | 2024/3/16 0:00:00 |
Lightweight object detection method for Panax Notoginseng in complex field harvesting |
| |
| WANG Fa'an,HE Zhongping,ZHANG Zhaoguo,XIE Kaiting,FANG Hanyu.Lightweight object detection method for Panax Notoginseng in complex field harvesting[J].Transactions of the Chinese Society of Agricultural Engineering,2024,40(8):133-143. |
| |
| Authors: | WANG Fa'an HE Zhongping ZHANG Zhaoguo XIE Kaiting FANG Hanyu |
| |
| Institution: | Faculty of Modern Agricultural Engineering, Kunming University of Science and Technology, Kunming 650500, China;Research Center on Mechanization Engineering of Chinese Medicinal Materials in Yunnan Universities, Kunming 650500, China;Research Center on Mechanization Engineering of Chinese Medicinal Materials in Yunnan Universities, Kunming 650500, China;Faculty of Mechanical and Electrical Engineering, Kunming University of Science and Technology, Kunming 650500, China; Tangyuan County Agricultural Technology Extension Service Center, Jiamusi 154000, China |
| |
| Abstract: | Intelligent harvesting is often required for object detection under complex field conditions, especially in the process of the real-time monitoring of harvesting quality and accurate grading conveyor. Taking Panax Notoginseng as the research plant, this study aims to propose lightweight object detection using YOLOv5s. The complex field conditions included the large variations in the light intensity, the difficulty in separating the roots from the soil, easy entanglement of roots, variable lifting speed, vibration amplitude, and frequency. The optimal model was also obtained with the high accuracy, and low complexity of a large model suitable for the deployment of mobile terminals. Firstly, a sample dataset was collected from the Panax Notoginseng in the complex field. The influence parameters of transportation and separation were also determined for the complex root-soil system; Secondly, real-time detection was realized under complex field conditions. Slim-neck lightweight neck network was introduced into the lightweight convolution of GSConv. The original SPPF feature fusion module was retained, while the ShuffleNetv2 lightweight feature extraction network was used to improve the original backbone network, which greatly reduced the model complexity with the model accuracy; Finally, the loss function with angular penalty metric (SCYLLA-IoU, SIoU) was used to optimize the bounding box loss function, in order to enhance the detection accuracy and generalization performance of the lightweight improved model. Ablation experiments were carried out to verify three improvement strategies, namely the Slim-neck neck feature extraction network, ShuffleNetv2 backbone feature extraction network, and SIoU bounding box loss function. The experimental results showed that the improved lightweight model (PN-YOLOv5s) had 3.27×106 M parameters, 5.4 G computational complexity, 6.85 MB weight size, and a detection speed of 108 frames per second. The number of parameters and weight size were approximately half of the original YOLOv5s, while the computational complexity was about one-third of the original model, and the detection speed increased by 1.2 times. Additionally, the precision of the improved model reached 93.15%, which was almost the same as the original model. The recall reached 89.46% with an improvement of 0.48%, compared with the original model. The F1 score reached 91.27% with an improvement of 0.22%. The mean average precision reached 94.20%, only 0.6% lower than the original. Compared with the mainstream SSD, Faster R-CNN, YOLOv4-tiny, YOLOv7-tiny, and YOLOv8s models, the improved lightweight model greatly reduced the complexity of the model, indicating better overall performance, in terms of model accuracy and real-time detection. The best performance was achieved in the actual harvesting, indicating the stronger robustness more suitable for deployment into mobile terminals. The bench tests showed that when the lifting inclination angle and vibration amplitude remained unchanged, the detection performance of the improved model decreased with the increase in lifting speeds and vibration frequencies. On the whole, the target detection of Panax Notoginseng was achieved with a precision of over 90%, an F1 score of over 86%, and a mean average precision of over 87% under four conditions of conveying and separation operating. There was little difference in the detection speed. But the improved model can fully meet the requirements of real-time detection under actual harvesting conditions. The finding can provide technical support to the subsequent monitoring of harvesting quality and adaptive grading conveyor in the Panax Notoginseng combine harvesters. |
| |
| Keywords: | neural network object detection lightweight complex harvesting conditions Panax Notoginseng 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 |
|
