Image classification using Partitioned-Feature based Classifier model

Classification of image data by using Partitioned-Feature based Classifier (PFC)is proposed in this paper. The PFC does not use concatenated feature vectors extracted from the original data at once to classify each datum, but uses extracted feature vectors to classify data separately. In the training stage, the contribution rate calculated from each feature vector group is drawn throughout the accuracy of each feature vector group and then, in the testing stage, the final classification result is obtained by applying weights corresponding to the contribution rate of each feature vector group. Experiments and results on Caltech image data set demonstrate that conventional clustering algorithms can improve their classification accuracy when the PFC model is used with them.