摘要
针对彩色图像中的目标进行快速、精确的分割问题是计算机视觉和图像分析的重点和难点。为提高图像识别质量,提出了一种基于区域的彩色图像分割方法。首先选择合适的彩色空间,提取出图像中的每个像素点的颜色、纹理、位置等综合特征,形成特征向量空间;在特征空间中,用ISODATA算法求出最佳聚类数目和初始聚类中心,后利用K-均值聚类算法进行聚类和图像区域分割,从中抽取出图像区域的特征,并与相类似的方法进行了比较实验。实验结果表明,图像分割算法速度较快,分割结果较精确。
How to fast, accurately and effectively segment objects in the color images is the key point in the computer vision and image analysis. This paper introduces a method of region - based color image segmentation. This method first extracts color, texture, and location features for each pixel to form integrative feature vectors by selecting suitable color space, so that the feature space is formed. Then, in this feature space, first of all, the best cluster number and the initial cluster center are obtained by ISODATA alogorithm and then an image is clustering and is sep- arated into regions by K - means clustering alogorithm. Finally, the features of regions are extracted. The experiment results and the comparision results with the similar approach are provided. Experiment results show the proposed method has the quickly segmentation speed and good sementation results.
出处
《计算机仿真》
CSCD
北大核心
2010年第6期271-274,282,共5页
Computer Simulation
基金
辽宁省高校重点实验室资助项目(2008s115)
关键词
图像区域分割
迭代自组织算法
区域描述
Image segmentation
Iterative self - organizing algorithm
Region description
