Image retrieval method based on weight self-learning hypergraph and multivariate information fusion

Abstract

The invention discloses an image retrieval method based on weight learning hypergraphs and multivariate information combination. The method includes firstly, extracting multivariate characteristics of images of an image database, and establishing distance matrix of different characterized spaces; presenting the image database in a manner of hypergraph structures according to the distance relations of images, and calculating parameters of the hypergraphs; focusing on specific retrieval sample images, calculating the initial label vectors according to the semantic space distances of images of the database and sample images, and combining the correlation of the images of the hypergraph structures, the consistence of the image ranking result and the initial label vectors and hyperedge weight learning performance of the hypergraphs through a normalized frame, and adjusting the effects thereof through the normalized parameters; finally figuring out the optimization problem of the normalized frame by the alternating optimization algorithm, and completing the hyperedge weight learning update while acquiring the optimization retrieval ranking results of the sample images.

Description

technical field [0001] The invention relates to an image retrieval and sorting algorithm in the field of computer technology, in particular to an image retrieval method based on weight self-learning hypergraph and multi-element information fusion. Background technique [0002] In recent years, with the rapid development of Internet technology and multimedia technology, network multimedia data has shown explosive growth in the Internet, and hundreds of millions of network pictures are uploaded and shared every day on social networking sites and multimedia sharing sites. Massive network images provide a large amount of information, but also become the difficulty of effective organization and management of image data. Therefore, how to efficiently and accurately retrieve the rapidly growing network image data to return the results that users are interested in has become the core issue of many practical applications in the multimedia field. [0003] Most of the traditional imag...

AI Technical Summary

Problems solved by technology

[0004] Although the use of image visual information and semantic information provides an important means for network image retrieval, the existing methods generally have the following two problems: firstly, the label information of network images is provided by various users, and there are many of them related to the Labeling images with completely irrelevant "noise" tags, and often misspellings in user-generated tags, makes image retrieval algorithms that utilize both visual and semantic information unsatisfactory in practical applications

However, when this method constructs the hypergraph structure, the weight assigned to each hyperedge in the hypergraph is fixed, so the hypergraph structure cannot be adjusted according to the specific analysis object.

Images