农田蛾类图像分割方法研究  

The Segmentation about Digital Image of Moths in Agricultural Field

作　　者：李延频[1] 顾波[1] 张红涛[1] 刘新宇[1] 邱道尹[1] 

机构地区：[1]华北水利水电学院,郑州450011

出　　处：《农机化研究》2009年第7期125-128,共4页Journal of Agricultural Mechanization Research

基　　金：中国科学院模式识别国家重点实验室基金项目(NLPR2003);河南省自然科学基金项目(200510078009)

摘　　要：介绍了已有的农田蛾类昆虫检测装置和相关研究技术。在已有技术的基础上,介绍了图像的预处理和杂质点的消除方法。在对提取的图像进行分割时,首先介绍使用Otsu方法来对图像进行分割,接着采用sobel算子提取蛾类对象的边缘,以边缘像素为种子点,使用区域生长来对蛾类图像进行分割。通过比较这两种分割方法可知,使用区域生长方法对蛾类图像分割的效果较好。This paper first to introduce the technology of the detection equipment about moths in agricultural field, base on these technology, it introduces the previous processing of image and the filtering of impurity points, when segmenting the moths digital image,. This paper first to introduce the method of Otus for segmentation about digital image, after to introduce another method that to use sobel algorithm to get the boundary pixels of moths in digital image and to use these boundary pixels as seed points, to use the region growing to segment the digital image of moths, comparing these two methods, the region growing method is better in image segmentation about moths digital image.

关 键 词：农田蛾类 图像分割 区域生长 边缘表示 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]