Automatic Image Brightness Adjustment

Due to the widely digital camera application, there is an enormous market demand for post digital photo processing, especially for adjusting the brightness of under-exposed photos. Brightness adjustment is an old problem in image enhancement. Some existing automatic image brightness adjustment algorithms are too simple to produce satisfying results. Recently, researchers proposed some complex adjusting algorithms, but they always need tedious user interaction or have high complexity which limit there application.We propose a new approach for automatically adjusting the brightness of underexposed digital images. We make the most of the dynamic range of digital images, and adjust the brightness through dynamic range remapping, with a specifically defined tone mapping operator. The operator is composed of two concatenate terms. The first one is a global operator that adjusts the brightness of underexposed image with a linear enhancement function as well as a non-uniform intensity reduction function. The second one is a local operator used for noise suppression and detail enhancement. Finally, we extend our algorithm to color images without chromatic bias. Our approach achieves real-time performance for the images with normal size (e.g. 0.3 second for 800*600 images).Meanwhile, in order to meet different needs, we give a gradient domain adjustment algorithm that can further improve the quality of image brightness adjustment. We first segment the input image, and then adjust the brightness of each region respectively. The constraint of gradient is derived from the region based adjustment, and then by solving a Poisson equation, we produce more satisfactory results ultimately.These two completely different algorithms automatically adjust the brightness to meet different application needs. The first real-time algorithm can ensure a great improvement against the input under-exposed image. The second gradient domain algorithm is fairly slow, but can produce even more desirable results.