Bright and dark distance-based image decomposition and enhancement

High contrast images are common in the scenes with dark shadows and bright light sources. It is difficult to simultaneously enhance the details in both dark and light areas on most wide dynamic range images. Recently, several image enhancement methods have been proposed to solve this problem. However, most of them are not consistent, may produce unnatural artifacts, and exhibit poor results when images having wide dynamic range are processed. In this paper, a novel method for contrast enhancement of wide dynamic range images is presented. Our method is based on an innovative image decomposition framework. Minimum cross-entropy between bright and dark image components is used to decompose an image into dark and bright image components. Visual and extensive quantitative analysis show that the proposed method outperforms the state-of-the-art algorithms, including the well-known Retinex, Histogram equalization, and Gamma Correction methods. Moreover, the new algorithm can be used in real-time image processing systems due to its simplicity and low computational complexity. The proposed method has various applications such as video door phone, security video cameras, and others. It is possible to be utilized in electronic products and image related instrumentation.