Research On Scene - Based Infrared Focal Plane Nonuniformity Correction System

In recent years,infrared focal plane array has been developed rapidly as the core technology of infrared thermal imaging.It is widely used in the fields of aviation,industry,agriculture,medicine,and scientific research with high sensitivity and reliability.However,because of the manufacturing technique,the photo-response of each detector within the IRFPA is different.As a result,the infrared image has non-uniformity.It has greatly restricted the application and research of the infrared thermal imaging system.So it is urgent to research the non-uniformity correction technology.With diving into the non-uniformity correction technique,a non-uniformity correction system based on FPGA platform is realized in this paper.In this paper,the reasons of the non-uniformity of the IRFPA and some common non-uniformity correction methods are studied.Following the principle of FPGA implementation and combining with the matching algorithm,a non-uniformity correction algorithm based on template matching algorithm is proposed.By using diamond search pattern,the algorithm selects image in a region as the benchmark module,calculates the motion vectors,and corrects gain and offset coefficients by the method of steepest descent.Because of inaccurate matching,the ghosting problem is appeared in the correction system.So a ghosting restraining method is proposed.Through experiment simulation and real time correction of FPGA hardware,the non-uniformity correction algorithm based on template matching is implemented to realize the real time fast matching between images.The algorithm has significant results in convergence speed and correction effect,and suppresses the ghosting problem effectively.