This course intends to support engineers who need to develop and design color imaging applications. The participants will get an idea about how differently image sensors behave in comparison to our human visual system and how the processing pipeline has to deal with the issues involved. Examples are the Auto White Balance, Color Demosaicing, Color Matrixing, Vignetting. But not only these items will be discussed, a major part of the course will be devoted to the correction of artifacts introduced by the image sensor. Examples are Dark Current, Defects, Fixed-Pattern Noise, Temporal Noise. "There's more to the picture than meets the eye (Neil Young, 1977)", this could be the title of the course as well. Many processing steps are taking place on the raw data delivered by the image sensor before the RGB data is shown to the end-user. And the course deals with all these processing steps.

In the past, the quality of a picture taken, for instance by a DSC, was determined to a large extent by the quality of the lens and of the image sensor, but digital-signal-processing power has made a lot of progress over the last year. The understanding of the physics behind the various defects and limitations of the components that make up a DSC has also grown rapidly. The combination of these two factors allows many defects in a DSC to be corrected. Along the road from photons IN to digital numbers OUT, the signal can pass through several calculation and correction cycles to improve the quality of the end result. This course will study the main artifacts that are introduced by the lens and the image sensor, and show how they can be corrected or compensated. Examples are lens vignetting, white balance, color sampling, non-ideal color filters, temporal noise, fixed-pattern noise, dead pixels, dark current, etc. Applications can range from consumer cameras (mobile imaging, DSC, cam-corders, etc.) to professional or scientific applications (medical, broadcast, astronomy, metrology, etc.). All artifacts and correction algorithms will be demonstrated by means of images. A basic understanding of the working principles of image sensors will be reviewed very briefly in the class.