Complex Fuzzy Integral And Its Applications

Multi-classifiers fusion is a powerful solution to the difficult pattern recognition problem,but who can improve the accuracy of classification and the robustness of systems ,we often use used as an integration tool for fuzzy integral .In the model of multi-classifiers fusion based on fuzzy integrals ,the fuzzy results are heavily dependent on fuzzy densities which represent the important of individual classifier to the fusion results. In this thesis, we proposed complex fuzzy integral which based on the application of multiple classifier fusion, and from the following aspects of the fusion of multiple classifiers were studied: the concepts of complex fuzzy integral,and discuss their properties and some converge theorems,methods to confirm the complex fuzzy measure, tested on the relevant samples. Firstly, the concept of complex fuzzy integral are given, and promoted some theorems of the classical integral to the complex fuzzy integral. Secondly, methods to confirm the complex fuzzy measure are given, and the fusion process of Complex Fuzzy integral in multi-classifier fusion applications are proposed. Finally, several examples were discussed in detail and the methods are verified to be practical.