The Research Of Scene-based Non-uniformity Correction For Infrared Detector

Infrared imaging technology is widely used in many departments and imaging with Infrared Focal Plane Arrays is future development direction of infrared and thermal imaging system. However, because of the materials of detectors, manufacturing technology and the response difference among the blocks in infrared detector, there is non-uniformity in the infrared images which reduces the distinguish ability. So it is necessary to make non-uniformity correction to the infrared detector. This paper aims at the analysis and research of the non-uniformity correction on the infrared detector.First, this paper describes the present developing situation of infrared detector and the non-uniformity correction algorithm nowadays. In order to comparing with scene-based non-uniformity algorithms, calibration-based non-uniformity algorithms are simply introduced. Then the reason to the non-uniformity and the characteristic of non-uniformity are analyzed, so that we know the characters of noise in infrared images. This paper pays most attention to analyze several scene-based non-uniformity algorithms by amounts of program experiments on the software platform. Based on the tradition methods, this paper proposed an improved Temporal High-pass Filtering and an improved Kalman Filtering which corrects the non-uniformity in more wide dynamic bound. At the end of this paper, the non-uniformity correction test system and how the scene-based algorithms work is introduced. Long time experiments proved that the algorithms are stable with good results.