Adaptive classifier based on K-means clustering and dynamic programming

Generally speaking, a recognition system should be insensitive to translation, rotation, scaling and distortion found in the data set. Non-linear distortion is difficult to eliminate. This paper discusses a method based on dynamic programming which copes with feature normalisation subjected to small non-linear distortions. Combining with k-means clustering results in a statistical classification algorithm suitable for pattern recognition problems. In order to assess the classifier, it has been integrated into a hand-written character recognition system. Dynamic features have been extracted from a database of 1248 isolated Roman character. The recognition rates are, on average, 91.67% (first choice) and 94.55% (first two choices). The classifier might also be tailored to any pattern recognition application.