A Context-Aware Similarity Search for a Handwritten Digit Image Database

In this paper, we propose a context-aware similarity search algorithm for a handwritten digit image database. Though we apply our algorithm to the search of handwritten digit images, the devised technique is generally applicable to other types of content-based image retrieval (CBIR). One of the central problems regarding CBIR is the semantic gap between the low-level features computed automatically from images and the human interpretation of image content. Many search algorithms that are used in CBIR have used the Minkowski metric (or L-p-norm) to measure similarity between image pairs. However those functions cannot adequately capture the aspects of the characteristics of the human visual system as well as the nonlinear relationships in contextual information given by images in a collection. Our new search algorithm tackles this problem by employing a new similarity measure and a ranking strategy that reflect the nonlinearity of human perception and contextual information in an image collection. Our search algorithm yields superior experimental results on a real handwritten digit image database and demonstrates its effectiveness.