Automatic Color Image Enhancement Using Double Channels

Digital cameras have been widely used in taking photos. However, some photos lack details and need enhancement. Many existing image enhancement algorithms are patch-based and the patch size is always fixed. Users have to tune the parameter to obtain the appropriate enhancement. In this paper, we propose an automatic consumer image enhancement method based on double channels and adaptive patch size. The method enhances an image pixel by pixel using both dark and bright channels. The local patch size is selected automatically by contrast feature. Our proposed method is able to automatically enhance both foggy and under-exposed consumer images without any user interaction. Experiment results show that our method can provide a significant improvement to existing patch-based image enhancement algorithms.