Design and Analysis of Efficient On-Star Compression Algorithms for FPGA Platforms

Image compression has always been a hot topic in the field of aerospace. The onboard processing unit, due to its background of deep space operation, has greatly limited performance in terms of data storage and data transmission. Satellite remote sensing images are developing towards larger size, higher frame frequency, and more frequency bands, which increasingly demand storage space and processing power. There is an urgent need for a set of efficient and lightweight image compression hardware and software. Therefore, in this work, we have improved the Context-Based Adaptive Lossless Image Compression (CALIC) algorithm, including error quantization and entropy coding, to meet the requirements for on-orbit processing in terms of compression time and compression ratio. At the same time, aiming at the working scenario of on-orbit processing, we chose Field-Programmable Gate Array (FPGA) as the platform to deploy the improved CALIC and designed the arithmetic coding Av value calculation process in parallel on this basis, which greatly reduced the processing time on the hardware side. We have carried out simulation verification of this paper’s image compression algorithm on the KODAK24 and SZTAKI_Airchange datasets. The experimental results show that our method strikes a good balance between compression ratio and compression rate compared with typical compression algorithms, making it suitable for on-orbit processing scenarios.