Fuzzy min-max neural network with compensatory neuron architecture for invariant object recognition

Object recognition system is divided into two parts namely, feature extraction and classification. Feature Extraction part consists of rotation, translation and scale invariant features. Recognition part consists of fuzzy min-max neural network with compensatory neuron architecture (FMCN). MPEG7 shape database is used for experimentation. The performance of FMCN is compared with fuzzy min-max neural network (FMN) proposed by Simpson. FMN uses the contraction method to eliminate the problem of hyperbox overlaps. FMCN eliminates the contraction method, since it is found to be erroneous. The concept compensatory neurons are inspired from the reflex system of human brain which takes over the control in hazardous condition. Compensatory neurons are getting activated when the testing sample falls in the overlapped regions of different classes.