Nonlinear Sequence Transformation-Based Continuous-Time Wavelet Filter Approximation

Data denoising and compression of intelligent transportation system based on two‐dimensional discrete wavelet transform

“…Due to certain restrictions on the resources of a specific application, such as storage space or computing power, the code stream of a large image cannot be completely decoded, and the application is only interested in a specific part of the image or code stream, so the extraction in the region of interest It is an important function of the latest image compression 12 . Because the wavelet transform itself is based on the “micro” characteristics of multi‐resolution analysis and data analysis, the overall correlation between the created code streams is very small, so as long as the corresponding part of the code, it is not necessary to correctly retain all the code streams 13 . In practical applications, you can use a low compression rate for the required parts of the image to improve the image effect, and use a high compression rate for other parts to save storage space.…”

“…12 Because the wavelet transform itself is based on the "micro" characteristics of multi-resolution analysis and data analysis, the overall correlation between the created code streams is very small, so as long as the corresponding part of the code, it is not necessary to correctly retain all the code streams. 13 In practical applications, you can use a low compression rate for the required parts of the image to improve the image effect, and use a high compression rate for other parts to save storage space. This important information will not be lost, and the image data is effectively compressed.…”

Analog Continuous-Time Filter Designing for Morlet Wavelet Transform Using Constrained L2-Norm Approximation

“…In [30], [31] and [32], Padé approximation is introduced to yields a rational expression for wavelet function. Some other approximation algorithms include Levin"s transformation [33,34], Bessel approximation [35] and Chebyshev polynomials [36]. However, these methods make a poor-fitting for wavelet function in time domain.…”