Combining image features for image classification

In this paper, we use neural networks and support vector machines (SVMpsilas) to compare the classification performances of four proposed image features. Of the four image features, two are developed by the authors, whereas the other two are well-known image features which we included for benchmark purposes. Indirectly the performances of the two image classifiers are compared. Based on the experiments that were carried out, it was found that our proposed combined image features gave the best performance amongst the four image features. In terms of the classifiers, SVM proved to be the better classifier.