计算机工程与应用 ›› 2023, Vol. 59 ›› Issue (11): 212-221.DOI: 10.3778/j.issn.1002-8331.2212-0023

• 图形图像处理 • 上一篇    下一篇

改进YOLOv5轻量级网络的柑橘检测方法

高新阳,魏晟,温志庆,于天彪   

  1. 1.东北大学 机械工程与自动化学院,沈阳 110000
    2.智能机器人工程研究中心 季华实验室,广东 佛山 528000
  • 出版日期:2023-06-01 发布日期:2023-06-01

Citrus Detection Method Based on Improved YOLOv5 Lightweight Network

GAO Xinyang, WEI Sheng,WEN Zhiqing, YU Tianbiao   

  1. 1.School of Mechanical Engineering and Automation, Northeastern University, Shenyang 110000, China
    2.Jihua Laboratory, Intelligent Robot Engineering Research Center, Foshan, Guangdong 528000, China
  • Online:2023-06-01 Published:2023-06-01

摘要: 针对现有的柑橘检测算法准确率低、模型参数量大、检测实时性差、不适用移动采摘设备等问题,提出一种基于改进轻量模型YOLO-DoC的柑橘检测方法。引入Bottleneck结构的ShuffleNetV2网络作为YOLOv5骨干网络模型,构造轻量化网络。同时加入无参型SimAM注意力机制提高复杂环境下对目标的识别精度。为了提高检测网络对于目标果实的边界框定位精度,通过引入Alpha-IoU边界框回归损失函数的方法来获取目标的边界框。实验显示,YOLO-DoC模型的P(precision)值和mAP(mean average precision)值分别为98.8%和99.1%,参数量缩减为YOLOv5网络的1/7,模型的大小为2.8?MB。改进后的模型相比于原网络模型具有识别速度快、定位准度高以及占用内存少的优势,在满足精准采摘工作要求的前提下可以提高采摘效率。

关键词: 神经网络, 注意力机制, YOLOv5, 柑橘检测, 损失函数, ShuffleNetV2, Alpha-IoU

Abstract: Aiming at the problems of the existing citrus detection algorithms, such as low accuracy, large amount of model parameters, poor real-time detection, and unsuitability for mobile picking equipment, a citrus detection method based on the improved lightweight model YOLO-DoC is proposed. This paper introduces the ShuffleNetV2 network of the Bottleneck structure as the YOLOv5 backbone network model to construct a lightweight network. At the same time, the non-parametric SimAM attention mechanism is added to improve the recognition accuracy of targets in complex environments. In order to improve the positioning accuracy of the bounding box of the target fruit by the detection network, the bounding box of the target is obtained by introducing the method of Alpha-IoU bounding box regression loss function. Experiments show that the P(precision) value and mAP(mean average precision) value of the YOLO-DoC model are 98.8% and 99.1%, respectively, and the number of parameters is reduced to 1/7 that of the YOLOv5 network, and the size of the model is 2.8?MB. Compared with the original network model, the improved model has the advantages of fast recognition speed, high positioning accuracy and less memory usage. It can improve the picking efficiency under the premise of meeting the requirements of precise picking work.

Key words: neural network, attention mechanism, YOLOv5, citrus detection, loss function, ShuffleNetV2, Alpha-IoU