Rapid detection of obstacle based on super pixels

For the mobile machine vision navigation process will encounter obstacles, this paper presents a fast obstacle recognition algorithm based on super pixels. First, reduce the resolution of the captured image in order to ensure real-time, and then using the SLIC super-pixel image processing algorithms. Finally, adopting the SAD method for classification estimates whether the pixel belongs to the obstacle. Experimental results show that the algorithm can quickly identify obstacles, playing an important role in visual navigation for mobile robots.