Analysis of Hidden Stochastic Properties in Images or Curves by Fuzzy Measures and Functions and their Fusion by Fuzzy- or Choquet Integrals

摘要

The paper shows the separation of stochastic properties in images or curves by analysis with different methods. A fuzzy measure and a fuzzy function represent these stochastic properties. Related to their importance the such represented stochastic properties are fused by the fuzzy integral. The fuzzy integral can be used as a new fuzzy measure; an iteration is generated until the interesting information is separated. If during the procedure of isolating a selected property the result violates with the original aim, then the search procedure has to be interrupted and another aim has to be traced. This is organised with the help of the Choquet integral. An estimated threshold selects the different stochastic parts.