The statistics of natural images have attracted the attention of researchers in a variety of fields and have been used as a means to better understand the human visual system and its processes. A number of algorithms in computer graphics, vision and image processing take advantage of such statistical findings to create visually more plausible results.
We have studied the statistical regularities of both conventional as well as high dynamic range images, and find that several commonly held wisdoms regarding natural image statistics do not directly apply to high dynamic range data. These results have implications for both the study of human vision, as well as for the design of algorithms in computer graphics and computer vision that rely on such statistical regularities.
For this study we created an extensive collection of images of 4 different categories: natural, manmade indoors, manmade outdoors and manmade night. Each category contains HDR and LDR versions of each scene, allowing us to compare the two directly. The HDR images were created by merging 9 exposures for each of the scenes. The LDR datasets consist of the best exposure out of the 9 (i.e. fewer over/under-exposed pixels). If you are interested in the dataset please send me an email or otherwise contact me and I will make them available.
Paper - Tania Pouli, Douglas Cunningham, Erik Reinhard, 'Statistical Regularities in Low and High Dynamic Range Images', ACM Symposium on Applied Perception in Graphics and Visualization (APGV), July 2010.
Course notes - Erik Reinhard, Tania Pouli and Douglas Cunningham, 'Image Statistics: From Data Collection to Applications in Graphics', SIGGRAPH Course, Los Angeles, 2010.
Dataset (~ 2GB zip) - The image collection described above. Please contact me if you cannot access this file.