Abstract: A classify based on AP Clustering and biomimetic pattern recognition was proposed. It can relatively classify the samples by calculating the distance to the relative subspace. The training sample space was constructed by the AP algorithm and bionic pattern recognition theory. hhe posterior probabilities based on the class condition were estimated to reduce the reject rate caused by the space overlapping with low misclassification. Experiments were performed with Concordia University CENPARMI's handwritten digit database and Nanjing University of Science and Technology's handwritten amount database. Experimental results indicate that the proposed classifier has a higher recognition rate than the traditional classifiers.

Key words: Affinity propagation clustering, Biomimetic pattern recognition, Posterior probability, Class-conditional confidence transformation, Handwritten digit recognition