Study On Nonuniformity Correction Technology Of IRFPA Based On DSP

Infrared Focal Plane Arrays(IRFPA) is a key part of modern infrared system and thermal imaging system, Imaging with IRFPA is future development direction of infrared and thermal imaging system. In this paper, the situation of IRFPA was presented, and the history, present state and tendency of IRFPA were generally expounded.With its widely used in military and civil aspect, the study on IRFPA will be profoundly significant. Based on studying on structure of many thermal systems, a master-slave IRFPA thermal system was developed, as the key parts are PC and Digital Signal Processor (DSP). Nonuniformity is intrinsic property of IRFPA, nonuniformity correction algorithms are deeply explored, and nonuniformity correction simulation experiments are carried.According to the characteristic of IRFPA, integrated design was realized on Infrared thermal system, and the DSP IRFPA thermal system was developed successfully. A master-slave IRFPA thermal system is developed as the key parts are PC and Digital Signal Processor (DSP), PC is the main control unit and DSP is the operation unit. The high speed of DSP was utilized to speed up the software algorithm. A Field Programmable Gate Array (FPGA) is used as sampling control unit in data sampling control part, which realizes the programmable of hardware circuit and advances system adaptability. Digital design was applied in most part of this system. The most important character of this system is high speed and precision, flexibility, stable performance, easily transplanted. Clear image is obtained from experiments using this imaging system.In present technology condition, it is impossible to make the video signal of IRFPA perfect uniformity. Nonuniformity is urgent to be realized, when IRFPA is used in staring imaging system. According to the nonuniformity of IRFPA, we research into the nonuniformity correction method of IRFPA. Now, the main correction methods are classified as two kinds: one is calibration correction and the other is self-adaptive correction. In this paper, we thoroughly study on memory correction, and advance memory optimized correction method. Using the optimized standard curves to realize correction, the nonuniformity correction precision is promoted and exposure space is wided. It is proved in experiment that remain nonuniformity is descended from 2.8% to 1.5% by the contrast to the old correction method and the memory optimized method.While used memory correction method, the calibration data is big. A new datacompression method-histogram amendment method is presented according to the distribution character of photoelectronic curves of IRFPA. While using this compression method, the capacity of memory is reduced, the dense space is wided and the precision of correction is promoted.In order to verify the validity of correction method, the simulation experiment is carried out using the IRFPA imaging system based on DSP. The CMOS image sensor is used to simulate the pyroelectricity IRFPA. Take use of the DSP imaging system, the contrast experiment is carried out by many kinds of correction method, such as two-point correction method, memory correction method, memory optimized correction method, histogram amendment compression correction method. It is verified that the new correction methods has higher correction precision and image effect after correction.