Generalized cellular neural network for novelty detection
Giovanni Martinelli, R. Perfetti · IEEE Transactions on Circuits and Systems I Fundamental Theory and Applications · 1994
A cellular neural network (CNN) for novelty detection is proposed. Each cell is connected to its neighboring inputs via an adaptive control operator, and interacts with neighboring cells via nonlinear feedback. In the learning mode, the control operator is modified in correspondence to a given set of patterns applied at the input. In the application mode, the CNN behaves like a memoryless system, which evidences those components of the input pattern that cannot be explained as a linear combination of the learned patterns.>