A robust stripe segmentation method for 3D measurement of structured light

Ambient noise and variation of illumination intensity can severely degrade the performance of the structured light sensor. To overcome these noises, we present a robust algorithm based on adaptive monochromatic space and minimum entropy model for the laser stripe segmentation in the structured light image. The proposed method not only makes high accuracy segmentation of the laser stripe, but also keeps the smoothness of the stripe. Experimental results show that the algorithm has great improvement to four classical algorithms.