An Object Detection Algorithm Based on Angle  Spread Shape Context Descriptor

The R-table of generalized Hough transform(GHT) and classic Shape Context(SC) are good shape descriptors. They can locate non-deformed shape object， while it is difficult to solve the problem when the target is similar but not necessarily identical to the user， that to say， there are deformation in some sense. A target locate algorithm based on angle spread Shape Context for solving this problem is presented in this paper. A common problem for the classic shape context is that when two contours are in different angular bins， similar contours have very different histograms. To overcome this problem， overlap spans of adjacent angular bins is presented. Experiments show that our method is capable of detecting object effectively and is valuable in computer vision field’s applications.