摘要
提出一种改进的基于均值偏移的户外图像快速分割算法,使其能够满足基于视觉的户外移动机器人导航对图像处理的快速性要求。阐述了均值偏移算法的基本原理,给出了图像快速分割的具体实现方法:对图像进行尺度空间变换,用均值偏移算法在选定的颜色空间中对图像进行分割。通过对多种户外环境下采集到的图像进行实验,结果表明该算法对户外彩色图像快速分割取得了良好的效果。
This paper presented an improved fast outdoor color image segmentation algorithm based on Mean-Shift,which could satisfy the real-time requirement of the image processing for the visual navigation of mobile robots in outdoor environments.It introduced the basic theory of Mean-Shift algorithm firstly,and then processed fast color image segmentation algorithm：used a scale transformation to the original image,then segmented the transformed image by the Mean-Shift algorithm in the chosen color space.The experiment results on the outdoor color images show that good segmentation results can be achieved by the proposed algorithm.
出处
《计算机应用研究》
CSCD
北大核心
2012年第2期793-796,800,共5页
Application Research of Computers
关键词
均值偏移
户外环境
图像分割
尺度变换
颜色空间
Mean-Shift
outdoor environment
image segmentation
scale conversion
color space
