Image contrast enhancement using entropy scaling in wavelet domain

• We present an entropy-based contrast enhancement method in the wavelet domain.
• The intensity components in the HSI color space are modified by the entropy of the contrast.
• The saturation components are linearly scaled by using the enhanced intensity component.
• Compared to conventional methods, our method gave reasonable outputs without any post-processing.

In this paper, we present an entropy-based contrast enhancement method in the wavelet domain. The proposed method is used in the HSI color space. The low-frequency coefficients in the wavelet domain are modified by the global histogram-based approach. The high-frequency coefficients are scaled by magnifying the entropy of the contrast defined in the wavelet domain. The contrast of the intensity component is enhanced first. Then, the saturation component of the HSI color space is linearly scaled by using the enhanced intensity component. Simulation results show that the image enhancement performance obtained by using the proposed algorithm is superior to the performance of the existing methods. In particular, the proposed approach can reveal the details and color information of low-light images without any post-processing.