Satellite-borne super-large-format color video image compression system and compression method

Abstract

The invention relates to a satellite-borne super-large format color video image compression system and compression method. The system comprises a video camera, a satellite-borne compression storage system, a data transmission system and a central machine, wherein the video camera is connected with the satellite-borne compression storage system, the satellite-borne compression storage system is connected with the data transmission system, and the satellite-borne compression storage system is connected with the central machine. Through tile cutting, compression can be completed by adopting industrial universal devices such as ADV212, and the development cost and the development difficulty of a satellite-borne compression storage system can be reduced. In the prior art, direct color compression is carried out after on-satellite bayer interpolation. Due to the fact that color precision loss can be caused due to the fact that an accurate algorithm with overlarge calculation amount cannot beadopted under the influence of calculation resources, spectrum segment separation belongs to a reversible process, color loss is avoided, after sub-spectrum segment data are obtained through ground decompression, an interpolation algorithm with higher precision can be adopted, and the color restoration effect is better.

Description

technical field [0001] The invention relates to the technical field of on-orbit compression of satellite-borne ultra-high-definition video color images, and relates to a space-borne ultra-large-frame color video image compression system and compression method. Background technique [0002] There are two main types of high-definition video compression technology, one is dynamic video compression represented by H.264, which has the advantage of high compression ratio, but the disadvantage is that it does not support lossless compression, except for key frames, the image information of other frames is too damaged , can only be used for video playback, and cannot use the difference in image information between frames for deep processing such as stereoscopic imaging and super-resolution. Video compression has strict restrictions on the frame size. Generally, it is required to meet standard formats such as 3840×2160, 1920×1080, and 1280×720. ;Another type of static image compressi...

AI Technical Summary

Problems solved by technology

[0002] There are two main types of high-definition video compression technology, one is dynamic video compression represented by H.264, which has the advantage of high compression ratio, but the disadvantage is that it does not support lossless compression, except for key frames, the image information of other frames is too damaged , can only be used for video playback, and cannot use the difference in image information between frames for deep processing such as stereoscopic imaging and super-resolution. Video compression has strict restrictions on the frame size. Generally, it is required to meet standard formats such as 3840×2160, 1920×1080, and 1280×720. ;Another type of static image compression represented by JPEG2000 has the advantage of retaining enough details in each frame, but the compression is relatively low, generally 2:1 for lossless, and generally no more than 10:1 for lossy, otherwise it will cause The image quality drops sharply. Generally, only the total pixels of the frame are limited. Taking the ADV212 chip as an example, the lossless mode requires that the frame size does not exceed 256k pixels, and the lossy compression does not exceed 1M pixels.

[0003] There are two methods for color video image compression. One is to directly obtain a color image by Bayer interpolation, and then convert it to a color space such as YCrCb for color image compression. However, this method will cause serious loss of color information, and the on-board processing algorithm is too complicated. , not very suitable for space-borne compression system; the second is to achieve spectrum separation on the star, compress each single-spectrum data according to panchromatic data, decompress it on the ground, and then restore it to Bayer color data, through the high-precision ground The Bayer color interpolation algorithm is used to obtain color images. This method has the advantages of simple on-board processing, complete color information preservation, and good color interpolation effects.

[0004] The ultra-high-definition video color image onboard has a frame size of 60 million pixels, which is difficult to directly realize by conventional dynamic video compression and static image compression, and the original image is arranged in RGGB Bayer encoding method, which cannot be directly compressed as a full-color image or a color image, so how to Realize the separation of Bayer color chromatogram segment and realize 60 million pixel ultra-large-format image data compression, is the main problem to be solved by the present invention

Images