Fuzzy-similarity-based noise cancellation for real-time image processing
Gustav Tolt, Ivan G. Kalaykov · 2002
We introduce a new algorithm for image noise cancellation based on fuzzy similarity and homogeneity. The proposed method allows simple tuning of fuzzy filter properties and it is very convenient for high-speed real-time image processing. A detailed analysis of the filter properties is presented to support tuning its parameters for a particular application. Test examples and comparisons with other image noise cancellation techniques show the advantages of the method.