Pixstat analyses pictures by classifying each pixel to either white = background or black = objects. This means that colour pictures are first transformed to grey tones and their pixels are then classified according to a threshold. In the first step, the brightness of each pixel is calculated from its three colour channel red, green and blue (RGB). The weight of each channel can be chosen between -100 and +100. The goal of the colour settings is to have the highest contrast possible between the objects and the background. Several combinations are preset:
grey (33% red + 33% green + 33% blue)
30% red + 59% green + 11% blue,
corresponding to the sensitivity of the human eye
red channel only
magenta (50% red + 50% blue)
green channel only
yellow (50% red + 50% green)
blue channel only
cyan (50% green + 50% blue)
The result of the current setting is visualised directly on a picture thumbnail.
If there is no particular reason to use a defined combination of coulours, it is possible to search for an optimum by running an Optimisation by example (via the Options menu).