Abstract:
In this paper, we propose a support vector machines (SVMs) method of classifying image regions hierarchically based on their semantics rather than on low-level features. First, image regions are segmented using the hill-climbing method. And then, the support vector machines classify these regions. The SVMs learn the semantics of specified classes from a database of image regions. A support vector machine was used as the classifier. We developed a new way to assign probability after multi-class SVM classification. Our approach achieved approximately 90% accuracy on a collection of images with minimal noise. A support vector machine (SVM) is used to classify the feature vectors. To reduce the computation time and improve the classification accuracy. We also developed a new way to compute probabilistic outputs from a multi-class support vector machines.
