摘要
面对玉米作物害虫检测中目标体积较小、形态多变且种类分布不均的情况,现有检测器会出现误检、漏检等问题;针对以上问题,提出了基于YOLOv7的玉米作物虫害检测算法SPD-YOLOv7;制作收集玉米害虫数据集,采用数据增强方法扩充数据集;引入SPD-Conv模块,替换原先骨干和头部网络中的部分跨步卷积层,减少随着网络加深细节信息的丢失,提高模型获取小目标特征和位置信息的能力;将ELAN-W模块与CBAM注意力机制结合,使网络更好地学习害虫特征,抑制背景信息,关注目标本身;改进后的YOLOv7网络模型准确率达到了98.38%,平均精度均值达到了99.4%;相较于原始的YOLOv7模型,准确性和平均精度均值分别提高了2.46、3.19个百分点,与Faster-RCNN、YOLOv3、YOLOv4、YOLOv5和YOLOv6主流算法的检测精度相比更具优势,且满足实时性;实验结果说明改进算法有利于快速识别玉米作物的虫害分布,可用于实际农田间的害虫实时监测。
Faced with the challenges of small target volumes,diverse morphologies,and uneven distributions of pests in maize crop pest detection,existing detectors suffer from issues such as false positives and false negatives.In response to above challenges,a maize crop pest detection algorithm SPD-YOLOv7 based on YOLOv7 is proposed.The maize pest dataset is created and collected,and the data augmentation technique is used to enlarge the data set.The SPD-Conv module is introduced to replace some step convolution layers in the original backbone and head networks,and mitigate the loss of detailed information as the network deepens,thereby enhancing the model s ability to capture the features and positional information of small targets.By integrating the ELAN-W module with the CBAM attention mechanism,it enables the network to better learn pest features,suppress background noise,and focus on the target itself.The improved YOLOv7 network achieves an accuracy of 98.38%and mean average precision of 99.4%.Compared to the original YOLOv7 model,the accuracy and mean average precision improve by 2.46 and 3.19%,respectively.The enhanced algorithm is superior to mainstream algorithms such as Faster-RCNN,YOLOv3,YOLOv4,YOLOv5,and YOLOv6 in the detection accuracy,while maintaining real-time performance.Experimental results indicate that the proposed algorithm has the rapid identification of maize crop pest distributions and can be applied for real-time pest monitoring in practical agricultural fields.
作者
宫妍
翟俊杰
王凯
李玉
GONG Yan;ZHAI Junjie;WANG Kai;LI Yu(College of Light Industry,Harbin University of Commerce,Harbin 150028,China)
出处
《计算机测量与控制》
2024年第9期58-65,共8页
Computer Measurement &Control
基金
黑龙江省哲学社会科学规划项目(23YSD245)
哈尔滨商业大学博士启动项目(2019DS087)。
