An improved non-uniformity correction algorithm based on calibration

The imaging performance and measurement accuracy of infrared radiation measurement systems are seriously affected by the non-uniformity of the focal plane array. Therefore, the non-uniformity of the raw infrared image needs to be corrected by the image processing algorithm. In order to further improve the Non-Uniformity Correction (NUC) effect of cooled infrared detectors, an improved non-uniformity algorithm based on calibration is proposed in this paper. The algorithm is based on the single-point calibration and the two-point calibration NUC methods, which not only retains the consistency advantage of the two-point calibration NUC method in gain correction coefficient, but also combines the stability of the single-point calibration in the offset correction coefficient. The improved algorithm has a better correction effect. In order to verify the correction effect of the improved algorithm, a cooled medium wave infrared detector with a size of 640 pixel×512 pixel is taken as the research object, and an infrared imaging system with a pupil diameter of 25 mm is used to verify the performance of the proposed algorithm. The experimental results show that under the 1ms integral time, the single-point calibration method, the two-point calibration method and the improved algorithm correct the image's non-uniformity to 1.7833%, 0.2190% and 0.1481%, respectively. And under the 2 ms integral time,they correct the image's non-uniformity to 1.8257%, 2.2474% and 1.6546%, respectively. The improved algorithm further reduces the image's non-uniformity more effectively, so it's correction effect is better and the accuracy is higher.