Semantic image classification with hierarchical feature subset selection

High-dimensional visual features for image content characterization enables effective image classification. However, training accurate image classifiers in high-dimensional feature space suffers from the problem of curse of dimensionality and thus requires a large number of labeled images. To achieve accurate classifier training in high-dimensional feature space, we propose a hierarchical feature subset selection algorithm for semantic image classification, where the feature subset selection procedure is seamlessly integrated with the underlying classifier training procedure in a single algorithm. First, our hierarchical feature subset selection framework partitions the high-dimensional feature space into multiple homogeneous feature subspaces and forms a two-level feature hierarchy. Second, weak image classifiers are trained for each homogeneous feature subspace at the lower level of the feature hierarchy, where the traditional feature subset selection techniques such as principal component analysis (PCA) can be used for dimension reduction. Finally, these weak classifiers are boosted to determine an optimal image classifier and the higher-level feature subset selection is realized by selecting the most effective weak classifiers and their corresponding homogeneous feature subsets. Our experiments on a specific domain of natural images have obtained very positive results.