Up to now we have seen how light and colors can be measured. But what happens when light reaches our eye? It hits the photoreceptors on the retina, and they send the signal through nerves to the brain, where an image is formed. As stated before, it is possible that each one of us creates different image from the same stimuli. As we can not describe our sensation, it can not be quantified in any way, either. The complete vision system is not completely understood yet, but there are many known human vision characteristics. A lot of research is done in the laboratories, under special conditions. E.g. the observer has adapted to a certain light level (it can even take half an hour or more to fully adapt to some conditions), he or she sees only a small portion of the whole visual angle and so on. Sometimes, results of such experiments are applied on computer generated images, viewed in a complex environment, that is far from the ideal laboratory settings. That is the fact, that should always be in our mind, when evaluating various tone mapping methods. In this section we are going to describe various characteristics used in various tone mapping techniques. There are a lot more known human vision characteristics, but they are beyond the scope of this work.
The light intensity range that we experience every day is huge. The ratio of light at noon on a sunny day and the moonlight can be as much as 10 million. As stated before this light hits photoreceptors in our retina, namely rods and cones. Rods are extremely sensitive to light and provide achromatic vision at scotopic levels of illumination ( to ). They provide achromatic vision, and that is the reason why we can not see colors in dark surroundings. The cones (there are three types of them) are less sensitive, but provide color vision at photopic levels of illumination (). Note that both systems are active at light levels between 0.01 and . This range is called the mesoptic range. Unfortunately the mesoptic range is the poorest researched, and this is the range that is exercised by computer-based office environments with CRT monitors and subdued lighting.
What happens after the light hits photoreceptors? The signal travels by neural units to the brain where an image is formed. It is interesting, that despite the fact that incoming light can have a dynamic range of nearly 14 log units, the neural units can transfer the signal having the dynamic range of only about 1.5 log units. It is obvious that there is some adaptation mechanism involved in our vision. It means that we adapt to some luminance value, and then we can perceive data in a certain dynamic range near the adaptation level. One of the most important characteristics that changes with different adaptation levels is the just noticeable difference.