Polar Radius-Haar Wavelet Descriptor for 2D Shape

Considering the low classification rate of the traditional feature extraction methods in recognizing the similar samples, on multi-classification and on multi-samples conditions, a polar radius-Haar wavelet descriptor for 2D shape is proposed by using the multi-resolution analysis. Following the process of the object segmentation, the breadth-first search is used firstly to track the binary image contour for the contour chain code, and then the p -order of the normalized polar radius signal is obtained. The multi-resolution´s character of shape description in detailed information and approximate information is used to obtain the invariants of the average energy and the average amplitudes. Then the eight invariants are proposed and the shifting, rotation and scaling invariance of it are proved theoretically. The experiments of the hand gesture show that the polar radius-Haar wavelet descriptor can classify the object shape rapidly and accurately, and more over, it has the better classification ability than the Hu invariant moment and polar-radius invariant moment.