Research On Signal Detection And Image Recognition Based On Lifting Wavelet

Different from traditional Mallat construct method, Wavelet lifting scheme is wholly based on time-spatial domain. It relies on neither translation and dilation nor spectrum analytical tools. Keeping biorthogonal condition, the performance of wavelet and dual wavelet is improved through lifting and dual lifting. Yet lifting process of wavelet can't enhance vanishing moment of wavelet. Through the improvement of lifting coefficients, the thesis enhances vanishing moment and dual vanishing moment, accordingly wavelet possesses more concussive character.In order to detect railway frequency-shift signal, according to its characters, a method based on lifting wavelet and Fourier transform is proposed. For the homemade 4 information and UM71 railway-shift signal, lifting wavelet transform and its contrary transform are applied. Through dint of peak value, the abscissa of signal and frequency discontinuous points are resolved. Then through DFT of signal between two discontinuous points, all the parameters of railway frequency-shift signal are resolved. The different effect of detection on railway frequency-shift signal by Haar and biorthogonal 9/7 wavelet is also compared.Off-line handwritten signature verification (HSV) is hard because it depends on only static information of image and dynamic information in writing process nearly fully disappeared. Facing the characters of Off-line HSV, this thesis proposes a method based on lifting wavelet transform. It combines structure feature with statistical feature, then extracted eigenvector is compressed by K-L transform. At last, true signature and forge signature are distinguished through support vector machines (SVM). The experimental results demonstrate the algorithm presented here reaches satisfied identical rate for testing stylebook.Lifting wavelet transform is the leading focus of the whole thesis. The excellent performance of lifting wavelet transform on one-dimension signal (railway frequency-shift signal) and two-dimension signal (image) indicates the significance of lifting wavelet in signal process domain. Moreover, enhanced vanishing moment wavelet filter designed based on lifting scheme plays the important role in signal process.