Image hashing method based on data sensing

Abstract

The invention discloses an image hashing method based on data sensing. The method comprises the following steps of: extracting feature vectors of an image; performing main ingredient analysis on the feature vectors to obtain a feature vector matrix with the feature values ranged in a descending order; performing ordinal measure on a feature sequence subjected to the main ingredient analysis to obtain the values in the vectors in the position sequence matrix of the whole vectors; performing locality preserving projection on the position sequence matrix of the vectors to obtain a dimension-reduction feature sequence; and performing binaryzation processing on the dimension-reduction feature sequence to obtain a 01-bit sequence. The method has the advantages that the time complexity is lower; obtained code words have higher efficiency; the semantics-preserving performance is better; the main ingredient analysis, the ordinal measure and the locality preserving projection are creatively combined; in the locality preserving projection, a distance metric function in the ordinal measure is used for constructing an adjacency graph, so that a projection matrix can better adapt to the feature vector, and the semantics-preserving performance of the code words is improved.

Description

technical field [0001] The invention belongs to the technical field of information retrieval, and more specifically relates to an image hashing method based on data perception. Background technique [0002] With the development of computer networks and the widespread use of digital processing technology, the multimedia data in the Internet has shown explosive growth, making the dissemination of multimedia data on the network increasingly popular. This makes the organization and management of multimedia data very important, and how to implement fast similarity search on large-scale data sets is the core key technology to complete the above functions. [0003] At present, similarity search techniques are mainly divided into methods based on space partitioning and methods based on hashing. The method based on space division mainly uses tree structure to store data, such as kd-Tree, R-Tree, X-Tree and SS-Tree. For space division, a suitable distance measure method is introduce...

AI Technical Summary

Problems solved by technology

However, due to the following problems in the tree structure: when the vector dimension is high, more space is required for representation, so that frequent splitting is required in the tree construction process and the height of the tree is significantly increased; in the process of space division, Due to the correlation between high-dimensional data, the space overlaps, so more subtrees need to be queried during the query process

However, due to the large amount of multimedia data and high-dimensional feature vectors on the Internet, similarity search algorithms need to face high-dimensional and massive data search problems

Images