Structured light stripe boundary detection using steerable filters

Accurate stripe boundary detection is crucial for stripe coded structured light based 3-D reconstruction. In this paper, we propose an improved stripe boundary location scheme based on steerable filters. In the proposed scheme, adaptive median filter is first used to smooth the input image by suppressing impulsive and non-impulsive noise. Because of their better orientation selectivity than classical edge operators, steerable filters are further adopted to extract the edge features. Nonmaximum suppression and hysteresis thresholding are finally utilized to determine the edge points. Experimental results show that the proposed scheme is very effective for structured light stripe boundary detection.