The method is based on the fact that the coefficient of the wavelet transform signifies the resemblance between the wavelet and the signal, at the same time, the regularity of the signal is relative to the wavelet coefficient.
A transform matrix B[n×n] for decomposing the sampled signal F[n] into Harr Wavelets series coefficients C[n] is presented,where n is the exponent of 2.As the inverse transform matrix B -1 =B T,the wavelet transform matrix method is then quite convenient for decomposing C=B·F and reconstructing F=B T·C.
Based on the analyses of the directional protection scheme and adequate consideration of the characteristics of HVDC line,the paper proposes the setting value protection scheme of polarity comparison using the wavelet transform.
First, the wavelet transform is used to perform a multiscale decomposition of each image. Then, the wavelet coefficients of fused image are constructed using multiple operators according to different fusion rules.
Utilizing the ability of time-frequency analysis of the wavelet transform in signal processing and approximation of the neural networks towards any nonlinear function,a method of intrusion detection based on wavelet neural network is proposed.
The image fusion based on wavelet transform is a novel pixel-level image fusion scheme on multi-scale decomposition. The medical image CT and MRI were decomposed by means of the wavelet transform. Then the wavelet coefficients of the image fusion were formed with the fusion scheme.
The results show that the self-adaptive filter can be used to detect line spectrum , the wavelet transform and the Fourier transform can be used to detect the ship noise in modulating the anisotropic noise field.
For the unstable Gaussian noise of the CCD image, the conventional method of Fourier transform can not eliminate it,but the wavelet transform is the powerful tools of noise elimination,since the wavelet transform in both the time domain and frequency domain has the excellent localization features, and the arbitrary details of analyzed objects can be focused with gradually detailed time domain or frequency domain step at high frequency.
It is not resultful to remove fractal noises using traditional method. The wavelet transform is a kind of multi-resolution method on finite time and frequency and it can separate and remove fractal noises in effect.
The basic concepts and the recent developments in ther modynamic formalism and phase transition for multifractals,and the wavelet transform,are introduced in this paper.
The wavelet transform (WT) is used to analyze the fractional Brownian motion (FBM) and it is proved that the wavelet transform of FBM is a stable process. Two methods for estimating parameter H of FBM with the wavelet transform are proposed. These methods can be realized by a fast wavelet transform (FWT) algorithm.
An approach to feature extraction and recognition of the characteristic signal is studied. The wavelet transform and its basic properties are discussed. An approach to feature extraction based on wavelet modulus maxima is suggested and a global distance function is presented. An application to fuel pressure waveform in a diesel engine is given as an example.