摘要
目的探究基于聚类分析k-means算法的图像分割以及在遥感图像中的应用。方法首先研究聚类分析算法,其次基于RGB和HSV颜色空间下通过软件编程实现算法,最后应用于遥感图像。结果与结论优点是可以比较好地实现遥感图像分割的效果,HSV颜色空间下的图像分割效果较好;缺点是由于算法自身的限制,只能达到局部最优,便利性较差。
Objective—To study the image segmentation based on clustering analysis and its application in remote-sensing images.Methods—Firstly,clustering analysis algorithm is studied.Then software coding is designed to realize the algorithm under the colorful space of RGB and HSV.Finally,the system is applied to remote-sensing images.Results and Conclusion—The advantage is that good results of remote-sensing image segmentation can be achieved.Compared with RGB color space,the result of HSV is a little better,while the disadvantage is that because of the algorithm＇s limit,only local optimum can be reached and its convenience is so bad.
出处
《宝鸡文理学院学报（自然科学版）》
CAS
2014年第2期66-69,共4页
Journal of Baoji University of Arts and Sciences(Natural Science Edition)
