GPU fast restoration of non-uniform illumination images

This paper presents a GPU based parallel implementation for the non-uniform illumination image restoration method, which uses a retinex based algorithm to decompose the original image into brightness and reflectance components, and adjusts the brightness value through an adaptive gamma correction and nonparametric mapping to achieve the restoration. Specifically, we parallelize the improved retinex algorithm on GPU to extract the brightness value of each pixel. After that, the probability of different brightness range is counted through each block to the entire image to reduce the competition of memory access. Finally, we use two different parallel reduce methods to calculate the probability density and cumulative density of brightness value and generate the mapping curve. The experiment conducted on three different GPUs and two CPUs with different resolution images shows that our method can process a 1024 x 2048 image in 1.024 ms on RTX2080Ti, indicates a great potential for real-time application.

Automated summary

This paper introduces a GPU-based parallel implementation of the non-uniform illumination image restoration method, utilizing an improved retinex algorithm for brightness value extraction and two different parallel reduce methods for probability density and cumulative density calculation. The experiment demonstrates the potential for real-time application, with a processing time of 1.024 ms for a 1024 x 2048 image on an RTX2080Ti GPU.