Multispectral image compression method based on square matrix transformation

Abstract

The invention provides a multispectral image compression method based on square matrix transformation, and relates to the field of computer information technology, especially to the field of multispectral image compression and transmission. A mean square error coding principle is combined with tristimulus values of a view to establish a color sense distortion criterion, and the criterion is converted into a matrix with visual characteristics; an image is divided into multiple representative image subclasses according to a specific method; CDF9 / 7 wavelet transformation and SPIHT coding are carried out on class components, CDF9 / 7 wavelet transformation and SPIHT coding are also carried out on differential components, and two types of code streams are output and stored respectively; and a decoding process includes inversion operation of a compression process. An error compensation mechanism based on the differential components is established to improve the spectral precision and chroma precision of reconstructed image. Under the condition that high compression ratio and transmission speed are realized, the multispectral image compression method can be used to store more effective image information, improve the the expansibility and stability, and ensure the pixel-level spectral and chroma precision during cross-device storage and transmission.

Description

[0001] Field [0002] The invention relates to the field of computer information technology, in particular to the field of compression and transmission of multispectral images. technical background [0003] In real work, the analysis of multispectral images is of great significance. It is a three-dimensional data cube that combines two-dimensional spatial information and one-dimensional spectral features. It is obtained by a spectral imaging system and records the scene under a specific illumination. Response value, it obtains its spectral information while obtaining the spatial information of the object. The rich spectral information creates good conditions for the recognition of the boundary of the object and the object of the object. Compared with the monochrome image, the multispectral image has a great Superiority. With the improvement of the spatial resolution of multispectral images and the development of geographic information system technology, people's confidence is...

AI Technical Summary

Problems solved by technology

However, various schemes have different weaknesses, such as loss of reconstruction spectral accuracy at the cost of poor illumination stability, non-targeted coding schemes, high computational complexity and lack of scalability, and color stability under different illuminations. Weak, and the reconstructed image is not conducive to consistent reproduction across devices or to ensure pixel-level spectral chromaticity accuracy. It can only remove spatial redundancy, and has high time complexity. There is no specific encoding algorithm, etc.

Images