Key Technologies on High-capacity Satellite Imagery Compression

With the continuous progress of science and technology, the accuracy of satellite remote sensing, detection and reconnaissance technologies is getting higher and higher, and the amount of data generated by them is also getting larger and larger. The amount of image data generated by high-resolution satellite imaging systems often reaches hundreds of Gb/s, which, coupled with the limited transmission bandwidth and storage space constraints, puts great pressure on data caching and transmission. The upper limit of transmission rate provided by traditional satellite data transmission methods has been difficult to meet the demand of inter-satellite and satellite-ground high-speed communication. To address this, a high-capacity satellite image data compression algorithm is proposed. Firstly, for the problem of redundancy of handover satellite image data, image fusion or stitching technology is used to extract valuable information from different images and synthesize high-quality images to reduce data redundancy. Secondly, to address the problem of repeated transmission of re-entry satellite data, multi-temporal data is analyzed to determine the change process of re-entry satellite image features and segment the change region in the image. In the re-entry process, only the information of the change region needs to be transmitted, avoiding multiple transmissions of duplicate information. Finally, to solve the problem of low compression efficiency of existing on-satellite compression algorithms, a new “image-information-image” data compression strategy is explored to achieve high compression ratio and high fidelity of satellite image compression. In order to achieve high capacity compression and efficient transmission of satellite images with limited bandwidth and storage space, and to lay the foundation for fast and efficient intelligent services for all-day, all-weather observation of the Earth.