Good Picture 2004: Digital De-Mystified!


Following a successful meeting last year the Imaging Science Group held the second of its, hopefully, annual symposia on digital imaging on Wednesday 15th December 2004 at the University of Westminster in London. Appropriately, the location was the old cinema in the University's Regent Street buildings which saw the first motion picture demonstration in the UK by the Lumière brothers over a century ago - and continues to be a venue for show-casing novel technology. Eight talks were given under the title "Good Picture 2004: Digital De-Mystified" by a number of experts in the field covering a variety of subjects. Over 100 delegates attended, professionals and amateurs, RPS members and non-members; their lively interest in the subjects added extra spice to the stimulating presentations.

The day began with a presentation of winning images from the 2004 Novartis/Daily Telegraph, Visions of Science competition by one of its judges Professor Ray Clark (RPS Past President). The stunning images, grouped under various subject headings, were a startling reminder that the beauty of science can be a visual as well as an intellectual experience.

The organisers of the competition were kind enough to provide an exhibition display of some of the images to support the slides, as well as a presentation envelope containing postcards of some of the pictures for each of the delegates attending the Symposium. The Imaging Science Group would like to take this opportunity to thank Novartis and the Daily Telegraph for their generosity and for allowing Professor Clark to give his presentation.

Dr. Phil Green of the London College of Communications, in his talk "From Shutter Click to Pixel - How Scene Colour Becomes Output Colour", described the transformations required to extract the required information from raw camera data to produce the acceptable images we see on various forms of colour display. Even more transformations are required, using colour management systems, to produce an acceptable print and he described these and the limitations inherent with various media.

In his presentation "Digital Cameras - Image Quality Matters", Dr. Lionel Baker described a convenient method of assessing the spatial quality of images produced by digital cameras. He has defined a parameter called "Optimum Print Width" which uses a sector star test chart in a simple, cheap and very practical test which can be carried out by even the most non-numerate user. Further details of the technique can be found here (PDF file) and on Lionel's website.

He presented some interesting results where he had compared a wide variety of cameras, making it clear that the much-quoted Megapixel number alone does not tell us everything we need to know about the relative resolution performance of digital cameras.

Paul Reynolds, of Sigma UK, in "From Film to Foveon" described the revolutionary new image sensor for digital cameras which uses light absorption at various depths in the sensor to respond to red, green and blue light in the same way as conventional film. He compared the image capture characteristics of this sensor with the more commonly used type which has a single layer of pixels and uses a colour filter matrix to record the three main regions of the spectrum. He showed examples of comparisons between images using the two types of sensor which indicated that the Foveon chip produced sharper images, better colour detail and less artefacts.

Dr. Rob Jenkin of the University of Cranfield gave one of his usual enthusiastic tutorials "DIY Photoshop 101" describing how some of the fundamental tools work in such software packages as Adobe PhotoshopTM and Paintshop ProTM. Of particular interest were his explanations of the effects produced by convolution which can be used to sharpen, blur or find the edges in an image. He introduced some very useful software, SCION IMAGE, which can be downloaded free from the Internet and used to explore these effects and enhance digital images. More on Scion Image (and IMATEST) can be found on our Toolbox page.

Dr. Jenkin then kindly - at very short notice - provided us with an additional entertaining talk "Apollo Moon Landing - Investigation of Photographic Anomalies". He produced a comprehensive debunking of the ideas of conspiracy theorists who believe that NASA never visited the moon and that all of the TV and film images were faked. By the use of models, simple scientific and photographic principles and more importantly, common sense, he was able to show that all of the "so-called" anomalies in the images were unfounded and could be easily explained.

Returning to the practical methods of digital image recording, the next talk was by Elizabeth Allen of the University of Westminster on "JPEG and JPEG 2000". Elizabeth introduced the concepts of compression and highlighted the dangers of loss in image information resulting from the reduction in file size. The artefacts produced by various compression techniques were also discussed.

The final talk of the day by Greg Duncan was entitled "The Truth is Out There!" and posed the very pertinent question of what amount of manipulation of a digital image is acceptable and what truth means in photography. Clearly there are some areas where absolutely no manipulation is acceptable, for example with forensic images, and at the other extreme where any amount of manipulation is acceptable, for example creative and artistic images. Unfortunately there is an enormous grey area between.

Following an entertaining talk, where he showed many interesting images with varying levels of manipulation, he did not answer his own question but turned it over to the audience to answer it in the light of their own experiences and uses of images.

At the end of the last session, RPS President Roger Reynolds praised the quality and content of the presentations, asking for a vote of thanks to the speakers, and urged the organisers to continue with this useful series of meetings.

An Open Forum session was then held where delegates and speakers could discuss any topics of particular interest:

  • Dr. Baker showed an unorthodox but effective approach to optical zoom combinations by demonstrating work he has carried out on linking a digital camera to a pair of binoculars.

  • There was continued discussion of the meaning of resolution and quality as applied to digital cameras.

Mike Christianson, Symposium Chairman

All photos © Harry Adam