Two-point non-uniformity correction method for infrared thermal image based on scene adaptation

Abstract

The invention claims a two-point non-uniformity correction method for an infrared thermal image based on scene adaptation and belongs to the field of infrared thermal imaging. A traditional two-pointnon-uniformity correction method can not adaptively correct an offset coefficient and a gain coefficient with the change of an ambient temperature, and so a non-uniformity correction error is large. The method of invention comprises a step of solving a gain coefficient and an offset coefficient of the two-point non-uniform correction of the infrared thermal image at different ambient temperatures,a step of obtaining expressions of the corresponding ambient temperatures of the gain coefficient and the offset coefficient by using a polynomial fitting technique, and a step of correcting the infrared thermal image in real time through the gain coefficient and the offset coefficient obtained by the expressions. Compared with the traditional two-point non-uniformity correction method, the method of the invention has a better correct effect and a lower residual non-uniformity of the infrared thermal image. The method has the advantages of a simple operation, low algorithm complexity and a good correction effect, and therefore the method has good application prospect and popularization value.

Description

technical field [0001] The invention belongs to the field of infrared thermal imaging, in particular to non-uniform correction processing of infrared thermal images based on scene adaptation. Background technique [0002] Under the uniform radiation of the black body, the output responses of each detection unit of the infrared focal plane array should be completely consistent, but in fact the output responses of each detection unit have certain differences, which is manifested as non-uniformity. From the perspective of noise, this non-uniformity is caused by spatial noise and transient noise, and its spatial noise is mainly caused by the inhomogeneity of the semiconductor material of the infrared focal plane array (doping concentration, crystal defects, and internal structure inhomogeneity, etc. ) and manufacturing process (mask unevenness and lithography error, etc.), transient noise is mainly manifested as dark current noise caused by circuit design noise and ambient tempe...

AI Technical Summary

Problems solved by technology

These algorithms have subtle differences in the effect of non-uniform correction. Generally speaking, these algorithms have the advantages of strong adaptive correction ability and less residual non-uniformity, but the algorithm also has slow convergence speed and high complexity. defect

These shortcomings severely limit the application of scene-based adaptive non-uniform correction algorithms in engineering practice.

[0004] According to the above analysis, it is concluded that the two-point non-uniform correction method cannot adaptively correct defects such as gain coefficients and offset coefficients with changes in the environment, resulting in large non-uniform correction errors

It is also concluded that although the neural network correction algorithm can adaptively correct the gain and offset coefficients, this algorithm needs a large number of image frames to converge, so the complexity is high, and it is difficult to use in engineering and other defects.

Images