[1]刘志强,杨昭,王建伊,等.基于YOLOv5的复杂背景下植物叶片检测研究[J].计算机技术与发展,2024,34(08):49-56.[doi:10.20165/j.cnki.ISSN1673-629X.2024.0117]
　LIU Zhi-qiang,YANG Zhao,WANG Jian-yi,et al.Plant Leaf Detection Research in Complex Backgrounds Based on YOLOv5[J].,2024,34(08):49-56.[doi:10.20165/j.cnki.ISSN1673-629X.2024.0117]
点击复制

基于YOLOv5的复杂背景下植物叶片检测研究

《计算机技术与发展》[ISSN:1006-6977/CN:61-1281/TN]

卷:
34
期数:
2024年08期
页码:
49-56
栏目:
媒体计算
出版日期:
2024-08-10

文章信息/Info

Title:
Plant Leaf Detection Research in Complex Backgrounds Based on YOLOv5
文章编号:
1673-629X(2024)08-0049-08
作者:
刘志强1; 杨昭1; 王建伊1; 张旭2
1. 内蒙古工业大学 信息工程学院,内蒙古 呼和浩特 010080; 2. 内蒙古建筑职业技术学院,内蒙古 呼和浩特 010020
Author(s):
LIU Zhi-qiang1; YANG Zhao1; WANG Jian-yi1; ZHANG Xu2
1. School of Information Engineering,Inner Mongolia University of Technology,Hohhot 010080,China; 2. Inner Mongolia Technical College of Construction,Hohhot 010020,China
关键词:
叶片检测; 复杂背景; 多尺度融合; 小目标检测; 深度学习
Keywords:
leaf detection; complex background; multi-scale fusion; small object detection; deep learning
分类号:
TP751
DOI:
10.20165/j.cnki.ISSN1673-629X.2024.0117
摘要:
对植物叶片进行检测是研究植物表型性状的基础,但真实环境下叶片间相互遮挡、叶片边缘特征不明显、幼叶目标过小以及外部环境如光照条件等因素影响会对叶片检测效果造成很大的障碍。 针对复杂背景下的叶片检测,该研究提出了一种基于改进 YOLOv5 模型植物叶片检测方法。 通过在骨干网络中引入空洞卷积,使得网络可以捕获到更广阔范围的上下文信息;利用双向连接的加权特征金字塔网络,以增强目标叶片特征提取并更好地融合特征信息;利用注意力机制,通过动态地调整注意力分布,以提高边缘特征表达能力。 测试结果表明,在 Plant Village 数据集筛选的葡萄叶片图像以及自拍摄葡萄生长叶片上测试改进算法的可行性,改进的 YOLOv5 模型其叶片检测 mAP 比原生模型提高了 5. 8% ,遮挡叶片检测精度提高了 7. 09% 。 叶片检测效果有显著提升。 该研究提出的方法可以有效解决复杂背景下植物叶片检测效 果不佳的问题,为植物表型研究提供技术支撑。
Abstract:
Detecting plant leaves forms the basis for studying plant phenotypic traits. However,real-world conditions,such as mutual leaf occlusion,indistinct leaf edge characteristics,small target leaves,and external factors like varying lighting conditions,pose substantial obstacles to effective leaf detection. To address leaf detection challenges in complex backgrounds,we present an improved plant leaf detection approach based on the YOLOv5 model. The introduction of dilated convolutions in the backbone network extends the network's ability to capture a wider contextual information range. Leveraging a bidirectional connected and weighted feature pyramid network enhances the extraction of target leaf features and improves feature information integration. The incorporation of attention mechanisms dy-namically adjusts attention distributions,thereby enhancing edge feature representation. Test results confirm the feasibility of the enhanced algorithm on grape leaf images from the Plant Village dataset and self-captured grapevine leaf images. The improved YOLOv5 model a-chieves a 5. 8% increase in leaf detection mean average precision (mAP) compared to the original model,with a 7. 09% enhancement in occluded leaf detection accuracy. This study significantly enhances leaf detection performance in complex backgrounds. The proposed method effectively addresses the issue of subpar plant leaf detection in such scenarios,thereby providing valuable technical support for plant phenotypic research.

相似文献/References:

[1]李志杰,袁鹏泰,李博涵.基于透视变换的手机身份证字符分割算法[J].计算机技术与发展,2018,28(07):58.[doi:10.3969/ j. issn.1673-629X.2018.07.013]
　LI Zhi-jie,YUAN Peng-tai,LI Bo-han.A Segmentation Algorithm of Identification Card Characters of Mobile Phone Based on Perspective Transform[J].,2018,28(08):58.[doi:10.3969/ j. issn.1673-629X.2018.07.013]

更新日期/Last Update: 2024-08-10