Research And Implementation Of A Full-temperature Non-uniformity Correction Algorithm For Infrared Images

Infrared imaging devices can detect infrared radiation that is not visible to the human eye and convert it into electrical signals related to the temperature of the target.They have strong environmental adaptability and are therefore widely used in military,medical,firefighting,and industrial fields.However,due to the limitation of the manufacturing process level,the response curves of different pixels in the infrared focal plane array are inconsistent,resulting in serious non-uniformity in the output infrared image,which not only reduces the imaging quality,but also increases the difficulty of subsequent processing.Therefore,it is of great significance to correct the non-uniformity of infrared images.Based on this,from the perspective of engineering practice,this article conducts research on non-uniformity correction technology,improves the defects of commonly used infrared image non-uniformity correction algorithms,and implements them in hardware.We take Hikvision H6417 CIS detector as the research object,analyze its performance indicators,designs a hardware architecture with FPGA as the core and LPDDR and FLASH as external memory,and build a Qsys-based infrared image processing software architecture for algorithm verification.We research and analyze the currently used non-uniformity correction technology.In view of its defects,we follow the principle of facilitating hardware implementation,and propose a shutterless correction technology based on ambient temperature compensation,an inter-frame registration correction algorithm based on the adaptive learning rate based on the combination of space-time domain and single-frame image de-stripe algorithm based on local midway equalization and simulate and implement them on FPGA.The results show that the improved algorithm proposed in this paper has obvious advantages over traditional algorithms and improves the imaging quality of infrared cameras.