Hdr Image Tone Mapping Algorithms Based On Bilateral Filtering

With the continuous development of computer science and technology, People have higher requirement for High-Resolution image and video. People have real, natural and clear feeling in High-Resolution image and video, which can make them experience rich and colorful image level like in the real world. High dynamic range (HDR) image is a kind of image type, which expresses more brightness range in the real scene than the general brightness of the display can be represented, it can show more detail features, have a good overall image contrast, HDR image pursue the quality of the image that is consistent with the natural real scene. Therefore, HDR image have a wide range of applications in many fields, such as satellite remote sensing, medical imaging, virtual reality and digital imaging. Due to dynamic range of the HDR image is generally able to reach nine orders of magnitude, while the ordinary display only two orders of magnitude dynamic range, how to make a HDR image displayed on the ordinary display become a problem for the further development of these applications. Domestic and foreign researchers proposed a lot of tone mapping algorithms for compressing the HDR image.The purpose of HDR image tone mapping operator is compressing the contrast of the image to make it can be displayed on the ordinary monitor, to improve the overall image contrast,to reserve the detail of the rich features, to get high quality images. This thesis describes the tone mapping research background and the basic theory of HDR images, meanwhile, introduces several classic tone mapping algorithms. Researches show that, most of the tone mapping algorithms is just to save a certain aspect of image quality, there may be poor in overall image contrast, detail is lost, brightness is turned, color is distortion, and flare phenomenon may appear. In this thesis, we detailed analysis of the bilateral filtering tone mapping algorithms, and improve the bilateral filtering tone mapping algorithms in image contrast compression on processing method. This algorithm according to image high light area and dark area automatically adjust the proportion of image contrast compression factor, rather than all pixel brightness values are the same compression ratio. This algorithm better realize the dynamic range compression and reshape the image of visual appearance.After several experiments, we discovered that this algorithm can produce a low dynamic range images which have a high image quality, and effectively give a solution to the halo phenomenon that may occur in tone mapping algorithms based on the same spatial variation. Meanwhile, it avoid the possible brightness inversion after tone mapping, and improve the overall contrast of the image, the image brightness changes have levels feeling, more close to the real natural scene.