Research On Tone Mapping For High Dynamic Range Images

Recently, high dynamic range images are becoming more and more important. High dynamic range images are more realistic than low dynamic range images. Now, high dynamic range technology is widely used in many fields: film processing, satellite meteorology, medical treatment and so on. However, most of display devices can only support low dynamic range output. In order to solve the problem, researchers purpose many tone mapping algorithms for high dynamic range images.The goal of tone mapping for high dynamic range images is trying its best to preserve the detailed information of original images, while compressing luminance of high dynamic range images. The article introduces basic theories about high dynamic range images and some classical algorithms. By researching, there is a fact that most of existing tone mapping algorithms can't render high dynamic range images well. So the article purposes a new tone mapping algorithm for high dynamic range images based on Meylan's. It improves the original algorithm. First, instead of PCA, it uses YCbCr color space to compress luminance component. Second, instead of Canny operator, it uses a new contour detection to do edge detection. Third, it processes chrominance component in HSI color space.Experiments show that, the new algorithm can perform tone mapping well for high dynamic range images. It can increase local contrast and enhance the saturation of images, while preventing halo aircraft. It is a better tone mapping algorithm.