Fixed pattern noise removal method based on minimum entropy

The invention discloses a fixed pattern noise removal method based on minimum entropy, and belongs to the field of image processing. Aiming at the defects of an existing image noise reduction method in processing fixed-mode noise, the implementation method comprises the following steps of: firstly, modeling the fixed-mode noise into a product of a noise mode obtained by calibration and an unknown scale parameter, and effectively processing non-Gaussian and non-stationary noise; secondly, automatically estimating a scale parameter by using a minimum entropy principle, and ensuring the superiority of estimation; and finally, subtracting the noise mode scaled by the scale parameter from the image with the noise to obtain a denoised image, thereby achieving the purpose of effectively removing the fixed-mode noise in the image. The method is suitable for the field of image processing and the like, can effectively correct the fixed mode noise in the image, greatly improves the imaging quality under the dark light c