基于阈值分割算法的水稻叶片病斑图像分割  

Segmentation of the Scab in Rice Leaf Image Based on Threshold Segmentation Algorithm

作　　者：高婷[1] 宋少忠 张佳环[3] 蔡利平 徐静 黎萍 董迎雪 车伟康 GAO Ting;SONG Shaozhong;ZHANG Jiahuan;CAI Liping;XU Jing;LI Ping;DONG Yingxue;CHE Weikang(Jilin Business and Technology College,Changchun Jilin 130507,China;Jilin Engineering Normal University,Changchun Jilin 130052,China;Jilin Agricultural University,Changchun Jilin 130118,China;Hunan Agricultural University,Changsha Hunan 410128,China)

机构地区：[1]吉林工商学院,吉林长春130507 [2]吉林工程技术师范学院,吉林长春130052 [3]吉林农业大学,吉林长春130118 [4]湖南农业大学,湖南长沙410128

出　　处：《吉林省教育学院学报》2021年第10期183-186,共4页Journal of Jilin Provincial Institute of Education

基　　金：吉林省发改委创新能力建设项目(2019C053);吉林省发改委创新能力建设项目(2021C044-10);吉林省发改委创新能力建设项目(2020C019-5);吉林省教育厅“十三五”产业培育规划项目(JJKH20200181KJ);吉林省自然科学基金项目(中央引导地方资金支持自由探索项目)(2020122348JC)。

摘　　要：图像分割是对图像特征提取与目标识别的前提,不同分割方法的精确度对特征提取及识别有很大影响。本研究对采集到的新鲜的感染稻瘟病的水稻叶片拍照,采用阈值分割法方对图像进行分割。实验结果表明,在直方图波谷处选取不同阈值进行分割,都能将水稻叶片图像中的病斑与叶片分割开,但阈值选取的差异会对分割结果产生很大影响。Image segmentation is the premise of image feature extraction and target recognition.Different segmentation methods have great influence on the accuracy of feature extraction and recognition.In this paper,through the threshold segmentation method,the collected fresh rice leaves infected with rice blast were photographed,and the threshold segmentation method was used to segment the image.The experimental results show that different thresholds at the trough of histogram can separate the diseased spots from the leaves,but the difference of threshold selection will have a great impact on the segmentation results.

关 键 词：图像分割 阈值法 直方图 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]