Fuzzy Neural Network Models For Multispectral Image Analysis

Fuzzy neural networks (FNNs) provide a new approach for classification of multispectral data and to extract and optimize classification rules. Neural networks deal with issues on a numeric level, whereas fuzzy logic deals with them on a semantic or linguistic level. FNNs synthesize fuzzy logic and neural networks. Recently, there has been growing interest in the research community not only to understand how FNNs arrive at particular decisions but how to decode information stored in the form of connection strengths in the network. In this paper, we propose fuzzy neural network models for classification of pixels in multispectral images and to extract fuzzy classification rules. During the training phase, the connection strengths are updated. After training, classification rules are extracted by backtracking along the weighted paths through the FNN. The extracted rules are then optimized using a fuzzy associative memory (FAM) bank. The data mining system described above is useful in many practical applications such as mapping, monitoring and managing our planet’s resources and health, climate change impacts and assessments, environmental change detection and military reconnaissance.