Fuzzy rule-based image processing

✍ Scribed by Kaoru Arakawa

• Publisher: John Wiley and Sons
• Year: 1997
• Tongue: English
• Weight: 164 KB
• Volume: 8
• Category: Article
• ISSN: 0899-9457
• DOI: 10.1002/(sici)1098-1098(1997)8:5<457::aid-ima7>3.0.co;2-9

✦ Synopsis

A novel image-processing technique based on fuzzy proposed here designs the membership functions automatically rules is proposed. This technique uses human knowledge about how by training. Strictly speaking, the whole system including the an image should be processed depending on the local characteristics membership functions is designed optimally by training. This of image. The knowledge is represented by fuzzy rules and is also fuzzy rule-based system can be applied to typical image-prorepresented by a multidimensional nonlinear function which can be cessing problems such as noise reduction and edge extraction.

optimized by training. Efficient noise reduction and edge detection As to the noise reduction, two types of noises are considered are realized by this fuzzy rule-based method. ᭧ 1997