A Method for Adaptively Enhancing Foggy Low-Illumination Image

Abstract

Due to the influence of the foggy weather, the increase of dust particles in the air affects the atmospheric scattering, resulting in the image degradation in the foggy low-illumination environment, such as blurred content and low contrast. This study proposes an adaptive image enhancement method for foggy low-illuminance image based on the proposed adaptive gamma correction and adaptive multiscale retinex (MSR). The proposed enhancement method firstly converts the foggy low-illumination image into HSI space and then performs the proposed adaptive gamma correction on its luminance component through estimating the light component of the image based on the retinex theory. After converting the obtained intermediate image back into RGB space, the proposed adaptive MSR is performed on it to obtain the final enhanced image. For the proposed adaptive gamma correction and adaptive MSR, an effective fitness function is designed to select the optimal scale value by virtue of particle swarm optimization (PSO) algorithm. The experiments show that the proposed method has better effect on improving the sharpness and contrast of the image, compared with the typical algorithms. The processing results of the proposed method have an average increase of 12.16% in information entropy, an average increase of 220.52% in average gradient, and an average increase of 111.63% in visual contrast, compared with the original foggy image.