low-light-level image enhancement method and device based on Retinex

Abstract

The invention discloses a low-light-level image enhancement method and device based on Retinex and belongs to the technical field of image processing. The low-light-level image enhancement method based on the Retinex comprises selecting a minimum valve in three channels of R, G and B of every pixel from an original color image to obtain a minimum pixel value image and saving the channel where a minimum pixel value of the pixel is located, namely a minimum channel table; subjecting the minimum pixel value image to multi-scale Retinex processing to obtain a Retinex-processed image; and computing processed valves of the other two channels except a minimum channel of every pixel of the original image according to the Retinex-processed image, the minimum channel table and the proportion of every component of every pixel to obtain the enhanced color image. According to the embodiment of the low-light-level image enhancement method and device based on the Retinex, the enhancement effect of relatively dark areas of an image can be improved, and the image quality is improved; meanwhile, the computation is reduced greatly and the processing efficiency is improved.

Description

technical field [0001] The invention relates to the technical field of image processing, in particular to a Retinex-based low-light image enhancement method and device. Background technique [0002] Common image enhancement processing methods include grayscale transformation represented by histogram equalization method, homomorphic filtering method based on illumination-reflection model, gradient domain enhancement method and Retinex enhancement method, etc. The grayscale transformation method represented by the histogram equalization method can make the image grayscale distribution more uniform and enhance the image contrast, but it does not consider the frequency information and detail information of the image, and it is easy to produce an over-enhanced image. The homomorphic filtering method based on the illumination-reflection model divides the image into high and low frequency parts, and then performs filtering to enhance the contrast of the image while compressing the ...

AI Technical Summary

Problems solved by technology

The disadvantage is that it will sharpen the image to a certain extent, and reconstructing the image in the gradient domain requires a certain numerical algorithm, which is not suitable for real-time use

The Retinex enhancement method uses a Gaussian smoothing function to estimate the brightness component of the original image, and applies the illumination compensation method to approximate the reflection image, which can enhance the information in the dark part of the image while maintaining the brightness of the image. Compared with other image enhancement algorithms, Retinex has sharpening, It has the advantages of large dynamic range compression, constant color, and high color fidelity, including single-scale Retinex and multi-scale Retinex algorithms. Among them, the multi-scale Retinex algorithm has better results, but has a large amount of calculation and low processing efficiency

Images