Content Based Image Retrieval with Multiresolution Salient Points

This paper presents a robust technique for Content Based Image Retrieval (CBIR) using salient points of an image. The salient points are extracted from different levels of the unsegmented image. Local contrast information at different resolution is embedded along with shape information. Fuzzy compactness vector is computed from the signature obtained at different thresholds. The resemblance of two images is defined as the similarity between the computed feature vectors.

• Type: Conference
• Year: 2004
• Where: ICVGIP
• Authors: Minakshi Banerjee, Malay K. Kundu
• Keywords: Computed Feature Vectors, Content Based Image Retrieval, ICVGIP 2004, ICVGIP 2008, Salient Points