基于改进的K-均值聚类算法的农作物图像分割  

Segmentation of Agriculture Images Based on Improved K-means Clustering Algorithm

作　　者：彭辉[1] 任继平[1] 吴兰兰[1] 陆尚平[1] 

机构地区：[1]华中农业大学理学院,武汉430070

出　　处：《农机化研究》2008年第6期57-60,共4页Journal of Agricultural Mechanization Research

摘　　要：图像分割是图像进行分析处理的首要步骤。为此,针对彩色农作物图像的特征,首先将RGB彩色图像转换到HIS色彩空间,运用均值-方差与粗糙集理论选取适当的初值聚类中心和聚类个数,再进行聚类计算,实现了色彩分量的快速自动化分割,较准确地从背景中提取出了目标物体,为农作物图像的识别与分析、后续计算和处理提供了可靠的基础。实验结果表明,改进的k-均值算法减少了运算量,提高了分类精度和准确性。Image segmentation is the first important step before image analysis. In this paper, according to the feature of color agricultural images, we firstly changed color space of RGB to HIS, then used improved algorithm to select appropriate number and the center of the clusters , which initialize the K - means clustering. And then the image is segmented by K -means clustering algorithm, the last , we achieve to rapid and automated segmentation, which extracted target objects form background accurately and provides a reliable basis for identification, analysis and follow - up treatment for agricultural images. Our experimental results demonstrated that our scheme can improve the computational speed of the k - means algorithm and can enhance precision and accuracy of clustering result.

关 键 词：农作物图像分割 K-均值聚类 HIS空间 

分 类 号：S375[农业科学—农产品加工] TP391.41[农业科学—农艺学]