Font Size: a A A

Distortion Correction And Dead Pixels Compensation Of Uncooled IRFPA Imaging System

Posted on:2006-01-26Degree:MasterType:Thesis
Country:ChinaCandidate:X T FanFull Text:PDF
GTID:2178360152975035Subject:Optics
Abstract/Summary:PDF Full Text Request
The more portable and fine Uncooled Infrared Imaging System has great foreground in the future. Firstly, the infrared optical system was usually designed to large-sized relative aperture system by reason of the target detected by the system was far, and weak radiant energy. However, the infrared image was suffered from geometric distortions due to wide-angle design of the infrared optical system. The geometric distortions could be corrected by the digital image processing technology, if it was not corrected by optical method. Secondly, there are billions or millions elements in the uncooled IRFPA (infrared focal plane array). The responsibility of the detectors of the IRFPA is not consonant with each other and it will cause the serious non-unification in the IRFPA. Some detectors are even ineffective and we call them dead pixels. The number and distribution type of dead pixels impact on performance of IRFPA greatly and make the visual effect of the image become very bad. It is necessary to build a digital image-processing unit for improving the imaging quality and getting more information. The main content of this paper includes two parts: (1) The distortion correction of infrared imaging system was discussed and the mathematical model of distortion correction was established. The space-variant linear image interpolation method as an efficient way was applied in the gray level interpolation of geometric distortion correction. The computational costs of the proposed approach are very little and the precision is high. It has been implemented by VC++6.0. (2) It was presented that adaptive dead pixels of 128×128 IRFPA compensation algorithm of four directions based on variable window. The dead pixel was not only an isolated pixel for the position distribution of dead pixels was diverse. The value of dead pixel was compensated by weighted interpolation of nearest normal pixels. According to the MATLAB experimental result, different distribution type of dead pixels could be corrected effectively.
Keywords/Search Tags:Uncooled infrared imaging system, Geometric distortion correction, Dead pixels, Interpolation compensation algorithm
PDF Full Text Request
Related items