Quality Recognition of Interphones by the LVQ Method
Nobuaki Kamimoto, Toshiro Abe, Lin Ken, Sigeru Omatu · IEEJ Transactions on Electronics Information and Systems · 2002
Automatic quality recognition of interphones has been well developed and it is important that the classifier must be highly accurate. Generally, the accuracy is represented by a recognition rate for sample data. To evaluate it more strictly, we will introduce a reliability criterion. In the pattern recognition, neural networks (NNs) have been adopted. Among them a competitive NN has a simple structure and can explain the input/output relation more easily than a layered NN based on the error back-propagation method. Thus, we adopt a competitive NN for the recognition of good and no good interphones where the learning vector quantization (LVQ) method is used for training. Finally, we show the effectiveness of the proposed method by applying it to real interphone data.