Quadri-Histogram Equalization for infrared images using cut-off limits based on the size of each histogram

In infrared images, the pixels representing the objects are hidden in a large number of background pixels with low contrast. Several effective contrast enhancement techniques exist in the state of the art today, however they cause the noise level added to the images to increase. The improvement of contrast is an indispensable procedure for the analysis of infrared images, due to the scarce temperature difference between the objects and the background, captured by the surveillance systems using infrared sensors. Therefore, a contrast enhancement algorithm for infrared imaging based on histogram equalization using clipping is presented in this article. The proposed algorithm divides the histogram into 4 subhistograms, then each subhistogram is modified with a cut limit based on the size of the subhistogram in order to limit the improvement of the contrast. The experimental results prove that the algorithm improves the contrast of infrared images by 99%, especially the contrast between the objects and the background of the infrared images preserving the mean brightness and decreasing the aggregate noise level of them. With the proposed algorithm, the background of the infrared image is restricted while the objects are visually contrasted.