| 1,046 | 0 | 316 |
| 下载次数 | 被引频次 | 阅读次数 |
为实现田间复杂环境下玉米幼苗与杂草的精准识别，通过对PP-YOLOE算法进行改进和优化，包括引入坐标注意力机制和ASPP(Atrous Spatial Pyramid Pooling)结构，提出了一种用于玉米苗间杂草检测的AAPP-YOLOE模型。试验结果表明，AAPP-YOLOE模型在玉米苗间杂草检测中取得了显著的性能提升。模型在召回率、准确率和F1-Score指标上表现更佳，证明了算法在玉米苗间杂草检测中的有效性和实用性，可为农业生产管理提供一定的帮助，有助于减轻杂草对玉米苗生长的影响。
Abstract：In order to realize the accurate identification of maize seedlings and weeds in complex field environment, according to the characteristics and requirements of this task, the PP-YOLOE algorithm was improved and optimized, including the introduction of coordinate attention mechanism and ASPP(Atrous Spatial Pyramid Pooling) structure, and an AAPP-YOLOE model for the detection of weeds between maize seedlings was proposed. The experimental results showed that the AAPP-YOLOE model had achieved significant performance improvement in the detection of weeds between maize seedlings. The model showed superior performance in recall, accuracy and F1-Score indicators, which proved the effectiveness and practicality of the algorithm in the detection of weeds between maize seedlings, which can provide some help for agricultural management and help reduce the impact of weeds on the growth of maize seedlings.
[1] PATELD D D.Weed and its management:a-major threats to crop economy [J].Journal of pharmaceutical science and bioscientific research,2016,6(6):753-758.
[2] 刘莫尘，高甜甜，马宗旭，等.基于MSRCR-YOLOv4-tiny的田间玉米杂草检测模型[J].农业机械学报，2022,53(2):246-255,335.
[3] 孙俊，何小飞，谭文军，等.空洞卷积结合全局池化的卷积神经网络识别作物幼苗与杂草[J].农业工程学报，2018,34(11):159-165.
[4] ELNEMR H A.Convolutional neural network architecture for plant seedling classification[J].International journal of advanced computer science and applications (IJACSA) ,2019,10(8):319-325.
[5] HAMUDA E,MC Ginley B,GLAVIN M,et al.Automatic crop detection under field conditions using the HSV colour space and morphological operations[J].Computers and electronics in agriculture,2017,133:97-107.
[6] NAKARMI A D,TANG L.Automatic inter-plant spacing sensing at early growth stages using a 3D vision sensor[J].Computers and electronics in agriculture,2012,82:23-31.
[7] DUTTA A,GITAHI J M,GHIMIRE P,et al.Weed detection in close range imagery of agricultural fields using neural networks[J].Publ.DGPF,2018,27:633-645.
[8] 李丹，张凯锋，李行健，等.基于Mask R-CNN的猪只爬跨行为识别[J].农业机械学报，2019,50(增刊1):261-266,275.
[9] 熊俊涛，郑镇辉，梁嘉恩，等.基于改进YOLOv3网络的夜间环境柑橘识别方法[J].农业机械学报，2020,51(4):199-206.
[10] 樊湘鹏，周建平，许燕，等.基于改进卷积神经网络的复杂背景下玉米病害识别[J].农业机械学报，2021,52(3):210-217.
[11] 张伟荣.基于机器视觉对靶施药系统的玉米苗与杂草识别技术研究[D].石河子：石河子大学，2021.
[12] 胡文泽，王宝聚，耿丽杰，等.基于Cascade R-CNN的玉米幼苗检测[J].农机化研究，2023,45(5):26-31.
[13] 郭敬东，陈彬，王仁书，等.基于YOLO的无人机电力线路杆塔巡检图像实时检测[J].中国电力，2019,52(7):17-23.
[14] HUANG X,WANG X,LV W,et al.PP-YOLOv2:A practical object detector[EB/OL].[2023-07-18].https://arxiv.org/abs/2104.10419.
[15] 赵文晖，伍昕忠，徐斌，等.目标检测在链条标准件动态抓取中的应用研究[J].制造业自动化，2023,45(4):213-216.
[16] HOU Q B,ZHOU D Q,FENG J S.Coordinate attention for efficient mobile network design[C]//2021 IEEE/CVF conference on computer vision and pattern recognition.IEEE,2021:13713-13722.
[17] 范亚臣.融入坐标注意力的多尺度Logo检测方法研究[D].西安：西安理工大学，2022.
[18] 龙洁花，赵春江，林森，等.改进Mask R-CNN的温室环境下不同成熟度番茄果实分割方法[J].农业工程学报，2021,37(18):100-108.
[19] 张宝才，高瑞翔.基于目标检测的玉米苗识别研究[J].农业开发与装备，2021(2):118-120.
[20] 刘伯圣.基于坐标注意力机制的弱监督目标检测研究 [D].临汾：山西师范大学，2022.
[21] 陆健强，梁效，余超然，等.基于坐标注意力机制与高效边界框回归损失的线虫快速识别[J].农业工程学报，2022,38(22):123-132.
[22] 杨其晟，李文宽，杨晓峰，等.改进YOLOv5的苹果花生长状态检测方法[J].计算机工程与应用，2022,58(4):237-246.
[23] CHEN L C,PAPANDREOU G,SCHROFF F,et al.Rethinking atrous convolution for semantic image segmentation[EB/OL].[2023-07-18].https://arxiv.org/abs/ 1706.05587.
[24] 李英萍.基于空间金字塔和注意力机制融合网络的影像语义分割[D].西安：西安电子科技大学，2021.
[25] 胡根生，吴继甜，鲍文霞，等.基于改进YOLOv5网络的复杂背景图像中茶尺蠖检测[J].农业工程学报，2021,37(21):191-198.
[26] 陈志伟，张万枝，张涛，等.基于YOLOv3算法的马铃薯种薯芽眼识别[J].农机化研究，2022,44(11):19-23,30.
基本信息:
DOI：10.13427/j.issn.1003-188X.2025.08.002
中图分类号:S451;TP183;TP391.41
引用信息:
[1]肖立同,王熙.基于改进YOLO算法的玉米苗间杂草检测模型研究[J].农机化研究,2025,47(08):10-16.DOI:10.13427/j.issn.1003-188X.2025.08.002.
基金信息:
“十三五”国家重点研发计划项目(2016YFD020060802); 黑龙江省农垦总局重点科研计划项目(HKKY190504)