计算机应用 ›› 2012, Vol. 32 ›› Issue (03): 752-755.DOI: 10.3724/SP.J.1087.2012.00752

• 图形图像技术 • 上一篇    下一篇

基于分水岭算法的作物病害叶片图像分割方法

任玉刚1,2,张建1,李淼1,袁媛1   

  1. 1.中国科学院 合肥智能机械研究所,合肥 230031;
    2.中国科学技术大学 信息科学技术学院,合肥 230026
  • 收稿日期:2011-07-20 修回日期:2011-11-21 发布日期:2012-03-01 出版日期:2012-03-01
  • 通讯作者: 任玉刚
  • 作者简介:任玉刚(1986-),男,河南信阳人,硕士研究生,主要研究方向:图像处理、模式识别;张建(1954-),男,陕西延安人,研究员,主要研究方向:人工智能、农业知识工程;李淼(1955-),女,安徽庐江人,研究员,博士生导师,主要研究方向:人工智能、农业知识工程;袁媛(1981-),女,安徽肥东人,助理研究员,硕士,主要研究方向:农业知识工程、农业信息化。

Segmentation method for crop disease leaf images based on watershed algorithm

REN Yu-gang1,2, ZHANG Jian1, LI Miao1, YUAN Yuan1   

  1. 1.Institute of Intelligent Machines, Chinese Academy of Sciences, Hefei Anhui 230031, China;
    2.School of Information Science and Technology, University of Science and Technology of China, Hefei Anhui 230026, China
  • Received:2011-07-20 Revised:2011-11-21 Online:2012-03-01 Published:2012-03-01

摘要: 为了提高作物病害叶片图像分割的准确性,采用了一种改进的基于标记的分水岭图像分割算法。首先,通过对二值图像进行距离变换和分水岭分割来获取背景标记,并通过提取数学形态学重建后的梯度图像中的区域极小值得到初步的前景标记,接着对前景标记进行进一步过滤,消除部分伪前景标记;然后,通过强制极小值方法将背景标记和前景标记叠加在梯度图像上;最后,对修改后的梯度图像进行分水岭变换。采用该方法对多幅黄瓜病害叶片进行图像分割,实验结果表明:该方法能够较好地将病斑部分分割出来,分割结果不受叶片纹理的干扰,平均分割正确率能够达到90%以上,具有一定的有效性和实用价值。

关键词: 分水岭算法, 标记, 数学形态学, 图像分割

Abstract: A new method based on watershed algorithm was proposed to raise the segmentation accuracy of the crop disease leaf images. At first, distance transformation and watershed segmentation were conducted on the binary crop disease leaf images to get the background marker, and the preliminary foreground markers were generated by extracting the regional minimum from the reconstructed gradient images, and then some fake foreground markers were eliminated by the further filter. In the next step, both background markers and foreground markers were imposed on the gradient image by the compulsive minimum algorithm. At last, the watershed transformation was carried out on the modified gradient image. Lots of cucumber disease leaf images were segmented effectively using the method. The results of experiment indicate that disease spots can be separated precisely from the crop leaf images. Additionally, the segmentation results are not influenced by leaf texture and its accuracy is up to more than 90 percent, so the method has certain validity and practical value.

Key words: watershed algorithm, marker, mathematical morphology, image segmentation

中图分类号: