Some camera makers reduce weak signals in order to reduce noise. However, as a side effect, the contrast of image is increased and precision of color description is decreased.

Without altering color description, noise may be reduced by one of the following methods:

• Standard filters, such as Gauss blur, that make photographs less sharp. This method is widely used in photo editors and in some cameras.
• Systems with pattern recognition that identify objects and blur colors only inside homogeneous (or quasi-homogeneous) areas. Most likely, a human brain uses this model.
• Adaptive algorithms that are sensitive to local change of intensity, but do not identify an object. This method is a cross between the first two methods.

KWEII has several own adaptive algorithms; some of them are already implemented in KWEII software and/or included in the KWEII compression.

The efficiency of KWEII noise reduction algorithms, implemented in KWEII software, can be seen from the comparison of the images below. Click on references KWEII Sony to see whole pictures.