Improving the Contrast of Dark Images with Fusion Blending of Fraction-Order Fusion Model and Bright Channel Prior

Abstract

The photos were taken in the dark light or poor environment always affects the quality of the images as these images are not able to understand for humane eyes and machines for experimental analysis. These images are hard to understand and identify objects with the help of precise details of the images. Sometimes machines get confused about these details of the images as image quality is degraded due to images taken in a poorly illuminated or dark environment. There are many existing techniques available for the contrast enhancement of the images. Some of these techniques have disadvantages. Disadvantages as a blurred image, a noise present in the image, the image gets distorted, etc. to overcome such disadvantages this paper proposed contrast enhancement techniques based on the simple weight blending of the bright channel prior(BCP) and Fraction-Order Fusion Model (FFM). For this experimentation exclusively dark image dataset is used and for the evaluation of the quality of the images entropy values of images are calculated. The outcomes of this experimentation give a better result compared to the individual output of bright channel prior (BCP), Fraction-Order Fusion Model (FFM), and other existing methods.