Fuzzy Clustering Paradigm and the Shape-Based Image Retrieval

This paper presents a strategy for shape-based image retrieval in which moment invariants form a feature vector to describe the shape of an object. Fuzzy k-means clustering is used to group similar images in an image collection into k-clusters whereas neural network is used to facilitate efficient retrieval of similar images against a given user-provided query image. Retrieval results and performance of the proposed scheme are compared with a setup involving k-means clustering.

• Type: Conference
• Year: 2008
• Where: FLAIRS
• Authors: Nan Xing, Imran Shafiq Ahmad
• Keywords: Artificial Intelligence, FLAIRS 2008, K-means Clustering, Shape-based Image Retrieval, Similar Images