| Wavelet analysis, which is able to capture both spatial and frequency localization and provide global approximation information and local detail information at different scales, has become an effective tool for signal processing and image understanding. Neural network is widely used in pattern recognition for the stronger ability of self-organization,revising,robust and non-linearity-handling. Moment invariants, with simple mathematical representation and perfect theoretical framework are well-suited for invariant target recognition due to their invariance to general affine transforms and insensitivity to noise. Vehicles' License Plate Recognition(LPR) system is an important application of computer vision and pattern recognition in the intelligent transportation. So the application research on moments character and neural network based on wavelet analysis in LPR system has a certain theorical significance as well as practical values.In this paper we look into the application of moment function and neural network in feature extraction and pattern classification of car plate character recognition respectively .The main works are as follows:1. After reading amount of literature and datum about wavelet analysis, the well spatial-frequency characteristic of wavelet transform is summarized in this paper.2. The traditional invariant moments are briefly introduced and their characteristics are analyzed and compared in detail. And the invariants of Hu moments, Zernike moments are analyzed.3. Aim at the limitations of the traditional invariant moments, the wavelet moment invariants are introduced and applied to the car plate character recognition . Simulation results show that the wavelet moments has the image object's invariance.4. The wavelet neural network is used in system of car plate image recognition. Experiments show that the wavelet neural network is better than thetraditional BP algorithm. |
PDF Full Text Request