Image Enhancement for Low-Light and Hazy Conditions Using Retinex Theory and Wavelet Transform Fusion

Image enhancement has played a significant role in analyzing and synthesizing images in the modern era. This study proposed a method to improve the visual information, discrete entropy, peak signal-to-noise ratio, and the visual quality of low-light and hazy image data. In this paper a technique for improving low-light and hazy images using illumination estimates, improved haze removal filter coefficients, and discrete wavelet transform fusion has been presented. The presented technique eliminates the low-light effect from the image to help improve the observation quality of low-light and hazy images. In order to eliminate the haze from the image, the optimal de-haze to the inverted image is applied. A better-quality image has been obtained by fusing both of the obtained images through discrete wavelet transform fusion. The simulation study showed a significant improvement in visible and quantitative aspects over previously proposed methods.