Planar region depth filling using edge detection with embedded confidence technique and Hough transform

This paper proposes a method to fill in the missing range information of a planar region in the depth map of an image obtained from a commercial stereo vision system. The Digiclops Stereo Vision System, a fast three-camera module range measurement device, is used to provide the initial depth image. The edge information of the reference image from the system is extracted by the embedded confidence edge detection technique. Hough Transform (HT) is then applied to the binary edge image to extract the straight-line boundary edges and the planar region is determined by the intersections of the edges. A 3-D planar equation is then used to determine the missing depth information of the region. The range data in the world coordinate system is projected back into the image coordinate system using a pixel-to-pixel projection algorithm. Results demonstrate the accuracy of the method in filling the missing depth information in a corrupted region in a depth map.