A method has been developed to calculate 2-D motion in a sequenceof time-varying images. The method, called motion detection usingrandomized Hough transform (MDRHT), is based on the randomized Houghtransform (RHT). The RHT decreases considerably the time consumption andmemory requirements of the Hough transform. The idea of the MDRHT is topick randomly point pairs from two images and calculate the translationwith them. The points can be e.g. edge points of the original images.This approach can avoid difficulties of standard segmentation methodslike overlapping and covering, and has the advantages provided by theRHT. The method can be generalized by picking more than two points.After a brief review of the RHT applied to motion detection, theextended algorithm to calculate both translation and rotation isrepresented in this paper
展开▼
