ACQUISITION OF HIGH DYNAMIC RANGE IMAGES BY MULTIPLE EXPOSURES

Peter Kragelj (2008) ACQUISITION OF HIGH DYNAMIC RANGE IMAGES BY MULTIPLE EXPOSURES. EngD thesis.

PDF Download (9Mb)

Abstract

There are more and more applications in the world, one of their constituent parts also being digital photography. Images are captured with modern cameras. However, most of these images uses the exact same saving-format. Eight bits are used to present each of colour channels for every single pixel of an image, meaning only 256 values are available to present whole dynamic range. This is not enough to present a lot of images, especially those containing dark, as well as very light parts of an image. Consequently a new type of images has been being developed – high-dynamic-range images. These are able to present a much higher dynamic range, eliminating problems with over- or under exposed image parts. High-dynamic-range images need to be captured first, which raises a lot of questions. The main question is how to do this in the first place. There are more answers: different advanced technologies, using special sensors, or entirely program solutions with multiple capturing at different exposure times. Capturing high-dynamic-range images with multiple exposures is made up of more algorithms, which complement each other and together compose a technique, which generates one high-dynamic-range image out of more low-dynamic-range images. First, images need to be aligned, then respond function is found, next images are merged and in the end potential errors are corrected.

Actions (login required)