人工智能在作物叶片病害图像分割中的应用  

Application of Artificial Intelligence in Image Segmentation of Crop Leaf Diseases

作　　者：王从胜 邱秀荣[1] WANG Congsheng;QIU Xiurong(School of Information and Electronic Engineering,Shangqiu Institute of Technology,Shangqiu 476000,China)

机构地区：[1]商丘工学院信息与电子工程学院,河南商丘476000

出　　处：《种业导刊》2025年第2期76-80,共5页Journal of Seed Industry Guide

基　　金：河南省高等学校重点科研项目(24B413009)。

摘　　要：随着精准农业和智慧农业的快速发展,图像分割技术在小麦和玉米叶片病害识别中的应用愈发重要。为实现对小麦和玉米叶片病斑图像的有效分割和高效分析,基于深度学习模型DeepLabv3+和U-Net对小麦和玉米叶片病害图像进行分割,同时对叶片病斑进行分割和量化评估。结果显示,2种模型均能准确地分割出叶片区域和背景区域,其中采用U-Net模型生成的轮廓更清晰、更完整,但仍出现分割不足、分割粘连等问题。进一步优化图像分割技术,提高叶片病斑图像分割的准确性,深入研究图像分割技术在农业生产中的应用,对小麦和玉米叶片病害的精确诊断与防治具有重要意义。With the rapid development of precision agriculture and intelligent agriculture,the application of image segmentation technology in the identification of wheat and maize leaf diseases is becoming more and more important.In order to realize the effective segmentation and efficient analysis of wheat and corn leaf disease spot images,based on the deep learning models DeepLabv3+and U‑Net,the wheat and corn leaf disease image was segmented,and the leaf disease spot was segmented and quantitatively evaluated.The results showed that the two models could accurately segment the leaf area and the background area.The contour generated by the U‑Net model was clearer and more complete,but there were still problems such as insufficient segmentation and segmentation adhesion.It is of great significance for the accurate diagnosis and prevention of wheat and maize leaf diseases to further optimize the image segmentation technology,improve the accuracy of leaf lesion image segmentation,and further study the application of image segmentation technology in agricultural production.

关 键 词：小麦 玉米 图像分割 叶片病害 人工智能 

分 类 号：S432[农业科学—植物病理学]