Contrast enhancement by adaptive mapping function with local information

Real scenes usually have a wider dynamic range than that of image acquisition devices, which makes most of the image acquired in under/over exposed conditions. Although the existing contrast enhancement algorithms such as contrast limited adaptive histogram equalization (CLAHE) are capable of restoring some of the details degraded in the acquisition, they may fail to detect others. This is mainly due to the complex degradation cases. It is necessary that any new mapping function should be modulated according to the local information. On this line, a new algorithm based on the CLAHE is proposed in this paper. The image quality improvement is focused on the enhancement of the local contrast in an image, in which the signal degradation could result in homogeneous sections in its histogram. It is achieved by discriminative clipping of the pixels based on its intensity levels and redistributing the pixels to maximize the available dynamic range. The contrast enhancement is performed locally based on the information of the local pixels. The clipped pixels are redistributed more assertively near histogram peaks to improve the image quality subjectively with maximum local contrast enhancement. The simulation result shows that the proposed method significantly enhances the image quality.