Image sending and receiving method and device

Abstract

The invention provides an image sending and receiving method and device. The sending method comprises steps: an original image is subjected to blocking processing; an approach curve of each macro-block is determined according to the texture complexity; each macro-block is subjected to positive transformation according to the determined approach curve and through a smoothlet algorithm, pixel points with gray value mutation on the transition zone of the approach curve are subjected to filtering processing, and the minimum coding unit is extracted; the minimum coding unit comprises position information of the macro-block, information of the approach curve and a positive transformation coefficient; the minimum coding unit is coded and then sent out. The sending method is improved based on the Smoothlet transformation, an image coding technology based on Multi-smoothlets transformation is provided, combined with an image coding theory, the compressed image quality is raised while few of the special numbers are occupied and compression coding of an image is achieved well.

Description

technical field [0001] The invention belongs to the field of wireless communication, in particular to the technical field of image transformation coding. Background technique [0002] Wavelet transform has been widely used since the 1980s. In the field of signal and information processing, it has become one of the mathematical tools that must be mastered like the Fourier transform. Although the wavelet transform has achieved good results in the fields of data compression and denoising, the separable two-dimensional wavelet transform (not directly constructed) uses a one-dimensional wavelet transform for the row and a one-dimensional wavelet transform for the column Extended, or two-dimensional transformations directly constructed from two separable one-dimensional function bases, but from a mathematical point of view, they are not real two-dimensional functions. The support area of ​​the basis function is extended from the interval to a square, and the directionality of th...

AI Technical Summary

Problems solved by technology

Smoothlet transform is a kind of ultra-wavelet transform. It fits an approximation curve in each sub-block through traversal algorithm according to the distribution of gray values ​​in the block (except for Degenerate smoothlet), that is, each Smoothlet can only be used for one benchmark Although Smoothlet has approximated the macroblock to a great extent, it has blurred edges in natural images and there may be multiple or even interlaced edges locally. In low-bit In the case of high rate, it still cannot be well approximated, so that the reconstructed image has a good subjective visual effect, and the numerical value is that the reconstructed image cannot reach a high PSNR value.

Images