A fuzzy filter for dynamic range reduction and contrast enhancement

When an image with a large dynamic range is recorded on media with a smaller dynamic range, details that represents in very high and/or very low luminance regions become invisible. In this paper a fuzzy logic based solution to this problem is proposed. The algorithm estimates the image brightness and hence common sense rules determine the amount of dynamic range reduction. These rules are based on global (i.e. image level) and local (i.e. pixel level) brightness characteristics. To extract visual details, a fuzzy process calculates the contrast amplification factor according to the luminance reduction performed.<
>