基于最小类内方差的蛇形机器人多阈值分割

Multi-Threshold Segmentation of Snake-Like Robot Based on Minimum Interclass Variance

针对传统最小类内方差分割方法计算量大、效率低、单阈值分割、不能多目标分割的缺点,提出了一种改进的基于最小类内方差的蛇形机器人多阈值分割方法.通过提取整幅图像的感兴趣区域(ROI),有效减小算法搜索的范围和整体计算量;根据直方图的多峰值特点,把ROI区域划分成多个子区域,采用改进的最小类内方差分割法搜索各个局部最优阈值,最终实现蛇形机器人关节组的多阈值分割.实验结果表明,该方法分割效率高,分割效果明显,且在保证实时性的同时提高了目标识别对光线变化的鲁棒性,降低了对步态变化的敏感性.

In order to remove the drawbacks of the traditional single-threshold segmentation methods, namely heavy computational burden, low efficiency and incapability of multi-target segmentation, an improved multi-threshold segmentation method of the snake-like robot is proposed based on the minimum interclass variance （MIV）. In this method, first, the R0I （ Region of Interest） of the whole image is extracted to effectively reduce the search scope and hence lighten the computational burden. Then, according to the multi-peak characteristics of histograms, the ROI is divided into multiple sub-regions, and the local optimal thresholds of the sub-regions are obtained by means of the MIV segmentation. Thus, the multi-threshold segmentation of the joint groups of the snake-like robot is successfully implemented. Experimental results indicate that the proposed method is of high segmentation efficiency, good segmentation effect, constant real-time performance, stronger target recognition robustness to light intensity change and lower sensitivity to gait change.