Image block feature vectors based on a singular-value information metric and color-texture description

Abstract

In this work, image feature vectors are formed for blocks containing sufficient information, which are selected using a singular-value criterion. When the ratio between the first two SVs are below a given threshold, the block is considered informative. A total of 12 features including statistics of brightness, color components and texture measures are used to form intermediate vectors. Principal component analysis is then performed to reduce the dimension to 6 to give the final feature vectors. Relevance of the constructed feature vectors is demonstrated by experiments in which k-means clustering is used to group the vectors hence the blocks. Blocks falling into the same group show similar visual appearances.

Recent development of perceptual image hashing

“…From Table 1, we conclude that reasonable modifications such as moderate brightness variations, small angle rotations, and high Q-factor JPEG coding do not alter the hash values, while large changes and malicious tampering can lead to considerable changes in the hash values. In order to develop a block-based image hashing scheme and a new CBIR system, we have proposed to form image feature vectors for blocks containing sufficient information, which are selected using a singular-value criterion [36] . Features including statistics of brightness, color components and texture measures are used to give intermediate vectors, which are then compressed to 6 by using a principal component analysis.…”