Image compression unit for large visual field TDICCD camera

Abstract

The invention discloses an image compression unit for a large visual field TDICCD camera, which includes an image sparse representation part, a bit plane entropy coding part and a rate control part, wherein the image sparse representation part includes the steps of firstly performing 3-level 9 / 7 wavelet transformation to the image, generating 1 low-frequency subband and 9 high-frequency subbands, and performing the post transformation to the 9 high-frequency subbands; the bit plane entropy coding part includes the steps of performing BPE coding to post transformation factors, and performing entropy coding to the coding result, coding a wavelet low-frequency subband through prediction, and performing entropy coding to a predicated residual error. According to the invention, the Hadamard single-machine dictionary post transformation strategy in the wavelet domain can be adopted to remove dependency among wavelet factors in the wavelet subband, the defect that the wavelet transformation is poor in sparse representation capability of remote sensing images with texture complex degree can be made up, rich information of the space CCD image borders and textures can be well protected; the image compression unit has the characteristics of excellent compression performance, moderate complex rate and high reliability, and is suitable for being used on satellites.

Description

technical field [0001] The invention belongs to the field of space CCD camera image compression, in particular to an image compression device for a space TDICCD camera with a large field of view. Background technique [0002] With the rapid development of aerospace technology, the field of view and spatial resolution requirements of space cameras based on TDICCD imaging technology have been continuously improved, and the number of CCD splicing pieces and readout rate used are also increasing and improving, so that digital After the CCD image data volume increases exponentially. It brings great challenges to the storage and transmission of satellite images, so image compression is very necessary. [0003] Space CCD images are usually used for target detection and recognition. At present, the judgment of important targets in space CCD images mainly depends on geometric contour features for confirmation. Therefore, space CCD cameras have a large imaging coverage, a large amoun...

AI Technical Summary

Problems solved by technology

If a unified code rate allocation strategy is adopted, the area with rich texture and smooth area must be assigned the same code rate, which may lead to blurring of edges and texture areas

Therefore, the traditional CCSDS IDC algorithm cannot sparsely represent remote sensing images well, and the code rate control efficiency is poor, and ultimately cannot protect the edge and texture-rich information of spatial CCD images well.

Images