Thursday, December 13, 2007

High-Dynamic-Range (HDR) Photographic Survey

Mark FairchildLast August, Prof. Mark D.Fairchild, Professor of Color Science and Director of the Munsell Color Science Laboratory in the Chester F. Carlson Center for Imaging Science at the Rochester Institute of Technology released an image database for research in High-Dynamic-Range (HDR) imaging. This database is called the HDR Photographic Survey.

In these days, when camera sensors have a bit depth of 12 or 14 bits and LCD panels are at 12 bits, developing good compression and rendering algorithms for HDR images is increasingly important. Also, most SLR cameras nowadays feature automatic exposure bracketing, so it is relatively easy to create HDR images. The problem is having well characterized reference images that everybody can use, so algorithms can be compared.Mark Fairchild in 9 brackets

One aim of Prof. Fairchild's HDR Photographic Survey is to provide such images in the public domain to researchers working on HDR systems and perception with a key feature being the inclusion of camera characterization data to allow conversion to accurate device-independent image data, colorimetric measurements of original scene elements, color appearance scaling of scene elements, and other scene data allowing increased utility of the images.

There are 106 images in all. Twenty-eight have accompanying colorimetric and appearance data. The remaining images have various data associated with them, but as a minimum have an absolute luminance calibration. The link is http://www.cis.rit.edu/fairchild/HDR.html.

The images are saved as OpenEXR files. The OpenEXR format retains the image data as 32-bit floating-point values. These minimally-processed — and photometrically linear — OpenEXR files are what is available in the database. Each HDR image retains a dynamic range of about 800,000:1 (19-20 stops, or bits, of real image information).

The advantages of knowledge of the original scenes, both direct and through the appearance scaling and colorimetric data, cannot be overstated. It is already evident that having such data combined with HDR images can greatly enhance research on human visual perception and imaging techniques.

Image content includes many natural landscapes, but also portraiture and indoor/outdoor scenes of man-made objects. Lastly, some extra images are also available that are simply nice photographs.

The database images and data are in the public domain for research purposes. It is only requested that they not be reproduced for commercial purposes and that the source of the images be acknowledged in any publications or presentations resulting from research in which they are used.

Monday, December 10, 2007

Income level: economists are wrong

In traditional economic models of decision-making, the most important determinant of individual well-being is the absolute level of income. A recent study based on brain activity observed using functional magnetic resonance imaging (fMRI) proves these models wrong. Indeed, social comparison affects individuals' subjective well-being, and thus behavior.

Dr. Armin FalkThe study was done at the Neuroeconomics Lab at the University of Bonn located at the Life&Brain Research Center by Dr. Armin Falk and his co-workers and is reported in the 23 November 2007 issue of Science, Vol. 318. no. 5854, pp. 1305 - 1308 in the paper Social Comparison Affects Reward-Related Brain Activity in the Human Ventral Striatum.

The team in Bonn had access to two MRI machines placed side by side and was able to concomitantly give the same task to two subjects while rewarding them differently in case of coincident success. They included nineteen subject pairs, and analyzed data from 33 subjects.

The task involved estimating the number of dots on a screen. At the end of each of 300 trials, both subjects received a feedback. This feedback provided information about both subjects' performance (whether the estimates were correct or incorrect), as well as about both subjects' payments in a given trial. Subjects solved the estimation task correctly in 81 percent of the trials.

Analysis of variance suggested that the importance of relative comparison is independent of the level of payment. In addition, there was no significant impact of the side of the activation or the scanner type. Thus, the results provide neurophysiological evidence for the importance of social comparison on reward processing in the human brain.

Compensation boards should keep this in mind, if they want their organization's success to be sustainable.

Thursday, December 6, 2007

Bit rot antidote

PDF 1.7 is now a Draft International Standard, soon to become ISO 32000.

In my October 30 post on Photo permanence and durability I had written about the problem of bit rot. For documents, since 1993 the best antidote for bit rot has been PDF (Portable Document Format). In fact, since then I always keep a copy of my documents in its original format and in PDF. Most of my tutorial materials are created in FrameMaker, but since Adobe abandoned it on the Mac platform, I often end up making small changes directly in the PDF file, which I can edit in Illustrator or in PitStop, depending on the edit's nature.

Now PDF 1.7 has become an even more potent antidote to bit rot, because the ISO has promoted it to a Draft International Standard. To learn more about this, read Jim King's PDF blog entry of December 4 with title ISO Ballot for PDF 1.7 Passed!.

Above I wrote explicitly about documents, not about pictures. If you put your pictures in a PDF file you will always be able to read them, but note that PDF is a structured file format, not a file format for images. Therefore, inside a PDF file, the image will be encoded for example in GIF, JPEG or JPEG-2000. These encodings are also ISO standards, so there will always be decoders for them. However, JPEG does not specify a file format, so wrapping JPEG images in a PDF file is better against bit rot. This is not an issue for JPEG-2000, which specifies also file formats.