Infrared image non-uniformity correction method and device, equipment and storage medium

Abstract

The invention discloses an infrared image non-uniformity correction method and device, equipment and a storage medium. The method comprises the steps of: collecting two frames of images, carrying out the registration of the two frames of images, and obtaining the offset information; according to the offset information, carrying out subtraction operation on the overlapped part of the two frames of images to obtain a noise difference value between different pixels; accumulating the noise difference values, and extracting low-frequency noise information; performing iterative updating on the noise value according to the noise difference values and the low-frequency noise information to obtain a noise value obtained through current iterative operation; and subtracting the noise value obtained by the current iterative operation from the pixel value of the to-be-corrected image to obtain a corrected image. Therefore, fixed noise, high-frequency noise and low-frequency noise can be effectively eliminated, efficiency and accuracy of non-uniformity correction are improved, operation resources are saved, and a blocking piece is not needed to serve as a reference.

Description

technical field [0001] The invention relates to the technical field of thermal imaging, in particular to a method, device, equipment and storage medium for correcting infrared image non-uniformity. Background technique [0002] The infrared focal plane array is a detector chip integrated with hundreds of thousands or even millions of pixel units, which is used to present infrared images in real time. However, there is obvious noise in the original output image of the actual thermal imager, also known as non-uniformity. The response of each pixel is different; the non-uniformity introduced by the working state, such as the uneven temperature of the focal plane device, the change of the drive signal on the readout circuit, the 1 / f noise introduced by the semiconductor surface current, etc.; the optical system, etc. Effects, such as the uneven distribution of radiation caused by the lens, the radiation of the lens barrel itself, etc. [0003] At present, most thermal imagers ...

AI Technical Summary

Problems solved by technology

However, in the process of use, with the complex environment and the change of the detector itself, the actual non-uniformity and the calibration result will be different

Therefore, it is necessary to place a shutter in front of the detector, and use the shutter as a standard to perform frequent calibration for non-uniformity correction, but this will increase the complexity and cost of the system, require additional steps for the user, and cause scene image interruption

[0004] Another non-uniformity correction method is a scene-based correction algorithm, but the current correction algorithm can only eliminate high-frequency noise, because the high-frequency noise changes sharply in a small scale and has obvious differences compared to the scene. Therefore its information is easier to extract and eliminate

However, during the use of the thermal imager, there is another type of noise that seriously affects the observation. It is mainly manifested as the overall light and dark gradient of the image, halo, etc. This kind of noise is called low-frequency noise. Its strength is very small, and the existing methods are usually difficult to effectively eliminate low-frequency noise

Images