A Novel Technique for Tone Mapping of HDR Images

Abstract

The images are quite useful for human understanding. It is an easy medium for visualizing the things. One picture is more worth than ten thousand words. High Dynamic Range imaging is used to capture the scene of the real world, so that the bright or dark areas will not be saturated in the output image. A high dynamic range (HDR) image uses a large bit depth up to 32-bit per pixel per color channel. Due to hardware limitation, it can neither be captured by conventional camera in a single photo, nor be displayed on a conventional monitor. The tone mapping method is used so that high dynamic range images can be displayed on the conventional display devices. Different methods of tone mapping have been proposed in the past. In this thesis a new tone mapping technique is proposed. In this technique a High Dynamic Range (HDR) image is taken as input and it is decomposed into two layers i.e. base layer and a detail layer. The details of HDR image will be preserved because only the contrast of the base layer is reduced. To obtain the base layer a bilateral filter is used. On the base layer the logarithmic compression is applied and a bias power function is used so that the logarithmic base can be varied adaptively and the contrast remains preserved. In the final step gamma correction is applied to improve the contrast in the dark areas of the image. This tone mapping technique produces a high quality tone mapped images as specified in experimental results.