基于彩色图像分割和平面投影变换的机器人障碍检测

Robot obstacle detection with colored image segmentation and planar projection transformation

提出了一种应用到机器人障碍检测中的新方法,该方法综合了自适应性的颜色区域分割和基于双目立体视觉的平面投影变换两种算法.此方法不但避免了图像匹配和生成全景深度图,且算法简单,可靠性强.尤其适应地面相对平坦、地面颜色大体一致的室内或高速公路等环境.最后给出了将此方法应用到移动机器人上的结果.

A new method was presented for robot obstacle detection,where two algorithm of adaptive colored area segmentation and binocular stereovision-based planar projection transmission were used incorporately.Using this method,the image matching and the generation of panoramic depth of field were avoided.On the other hand,the algorithm was simple,reliable,and suitable especially for indoor or highway environment with relatively smooth and basically monocolored floor.Finally,applicatin result of this method to the mobile robot was given.