On image classification: city images vs. landscapes

Grouping images into semantically meaningful categories using low-level visual features is a challenging and important problem in content-based image retrieval. Based on these groupings, eeective indices can be built for an image database. In this paper, we show how a speciic high-level classiication problem (city images vs. landscapes) can be solved from relatively simple low-level features geared for the particular classes. We have developed a procedure to qualitatively measure the saliency of a feature towards a classiication problem based on the plot of the intra-class and inter-class distance distributions. We use this approach to determine the discriminative power of the following features: color histogram, color coherence vector, DCT coeecient, edge direction histogram, and edge direction coherence vector. We determine that the edge direction-based features have the most discriminative power for the classiication problem of interest here. A weighted k-NN classiier is used for the classiication which results in an accuracy of 93:9% when evaluated on an image database of 2; 716 images using the leave-one-out method. This approach has been extended to further classify 528 landscape images into forests, mountains, and sunset/sunrise classes. First, the input images are classiied as sunset/sunrise images vs. forest & mountain images (94:5% accuracy) and then the forest & mountain images are classiied as forest images or mountain images (91:7% accuracy). We are currently identifying further semantic classes to assign to images as well as extracting low level features which are salient for these classes. Our nal goal is to combine multiple 2-class classiiers into a single hierarchical classiier.