RECOGNITION OF 2D OBJECT CONTOURS USING STARTING-POINT-INDEPENDENT WAVELET COEFFICIENT MATCHING

In this paper, a new recognition algorithm for 2D object contours, bas ed on the decimated wavelet transform, is presented, emphasizing the s tarting point dependency problem. The proposed matching algorithm cons ists of two parts: Firstly, we present new data structures for the dec imated wavelet representation and a searching algorithm to estimate th e misalignment between the starting points for the reference model and unknown object. We also adopt a polynomial approximation technique an d propose a fast searching algorithm. And then, matching is performed in an aligned condition on the multiresolutional wavelet representatio n. By employing a variable-rate decimation scheme, we can achieve fast and accurate recognition results, even in the presence of heavy noise . We provide an analysis on the computational complexity, showing that our approach requires only less than 25% of the computational load re quired for the conventional method [1]. Various experimental results o n both synthetic and real imagery are presented to demonstrate the per formance of the proposed algorithm. The simulation results show that t he proposed algorithm successfully estimates the misalignment and clas sifies 2D object contours, even for the input SNR = 5 dB. (C) 1998 Aca demic Press.