A New Shape Distance for Content Based Image Retrieval

Turning Angles Representation (TA) is considered one of the most interestingmethods for representing object shapes in content-based image retrieval systems. Nevertheless, the distance commonly used to measure the similarity between shapes represented by TAs, the Euclidean one, is generally too sensitive to small variations in the shapes. In this paper we present a new distance between shapes represented by TA, namely the Median distance, specially devised to minimize the e ects of small variations in the shapes. Its analytical properties are discussed and experimental results are provided by comparison against those obtained by applying traditional techniques based on Euclidean distance. The Median distance has been implemented in the Automatic Image Storage and Retrieval (AISR) system, which allows storage and content-based retrieval of 2-D images.