A Composition-Based Image Retrieval Method for Environment-Visualization with Images and Spatio-Temporal Information

摘要

It is promising to combine multiple images for expressing query-image and detecting situations and phenomena in environmental image databases. This paper presents a composition-based image retrieval method for environment-visualization with images and spatio-temporal information. A query-image is created by combining multiple images based on color and composition. This method realizes to detect the particular situation of environment and find out the images similar to that situation of environment. This method uses multiple divided-part-images in a single image and Euclid distance calculation for detecting highly-related images to a query-image. In this paper, we present several experimental results to clarify the feasibility and effectiveness of the proposed method.