移动机器人级联超像素行人目标分割算法

针对移动机器人视觉应用中,复杂室内外环境下行人目标提取因背景干扰而导致主体轮廓失真的问题,提出一种基于超像素的级联式行人目标分割算法。利用超像素对目标边缘轮廓的吸附特性,第一级超像素在获取全局超像素区块的基础上,结合行人显著区域检测,计算第二级超像素区块的平均颜色距离和中心点空间位置距离相关度,从而获取行人目标轮廓的分割结果。仿真结果表明,该算法精确度与召回率统计平均为0.98,高于当下流行的其他显著目标分割算法,对行人目标检测分割性能具有良好效果,为行人目标跟踪等应用提供必要的预处理基础。

Cascaded Superpixel Pedestrian Object Segmentation Algorithm for Mobile Robot

For the human-being’s body contour distortion problem of the pedestrian object segmentation at the complex indoor and outdoor environment for mobile robot visual applications, a cascaded superpixel pedestrian object segmentation algorithm was proposed considering background interference. Based on acquiring the global superpixel blocks with the primary stage superpixel computation, the second stage superpixel achieved the correlation degree of the average color and center point Euclidean distance of each superpixel blocks between inside and outside of the pedestrian saliency detection region, in order to obtain the segmentation of the up-right person. With the simulation results, this proposed algorithm is 0.98 in precision and recall statistical average and has excellent target extraction performance compared to state-of-the-art saliency object segmentation algorithms, so that this method can provide the support for the pedestrian object tracking applications.