Measuring image similarity using the geometrical distribution of image contents

摘要

To measure the similarity of images using spatial distribution of primary features such as colour, shape and texture is difficult because the image has to be segmented and features are extracted from localized areas. Few researches have been done in this area. However, such information is vital to the content-based image retrieval as it contributes to the similarity measurement in the human visual system. Based on our previously proposed signature using Radon transform, we propose a decimation to reduce the projections using the principal component analysis and use correlation in measuring the similarity in this paper.