Semi-supervision sequencing study method for image searching based on manifold regularization

Abstract

The invention discloses a semi-supervision sequencing study method for image searching based on manifold regularization. The semi-supervision sequencing study method comprises the following steps of: extracting visual characteristics from a network searching result from a database or an initial text-based network to form an image sample set; dividing the image sample set into three grades including 2, 1, and 0 according to degrees for inquiring subject coherence, wherein 2 represents that the image sample set is very coherent with the inquiry, 1 represents that the image sample set is commonly coherent with the inquiry and 0 represents that the image sample set is incoherent with the inquiry; calculating spurious correlation grade information yi of an unmarked image sample; calculating a distance between two image samples; constructing a Laplace manifold regularization item according to the distance between the two image samples; constructing a target function through the Laplace manifold regularization item; and solving a sequencing score for obtaining each image sample by the target function and feeding a sequenced result back to a user. According to the semi-supervision sequencing study method disclosed by the invention, the searching and sequencing performances are improved, marking information is sufficiently utilized and the searching precision is improved; and less supervision information is effectively utilized to improve the sequencing performance.

Description

technical field [0001] The invention relates to the field of multimedia information retrieval, in particular to a manifold regularization-based semi-supervised ranking learning method for image retrieval. Background technique [0002] With the rapid development of information technology, multimedia resources such as images and videos have grown rapidly. Images and videos have become one of the important ways for people to obtain information because they contain rich, intuitive and interesting information. How to quickly and accurately obtain the information required by users from massive amounts of data is a challenging task. Existing commercial search engines retrieve images or videos mainly by retrieving the text information of the webpage where the images or videos are located. However, limited textual information cannot fully describe the rich content of an image or video, nor can it accurately describe how relevant an image or video is to a query. Furthermore, there m...

AI Technical Summary

Problems solved by technology

[0007] 1) The existing ranking learning is usually supervised, that is, only the labeled samples are used to train the ranking model, and the widely existing unlabeled samples are not used, which is not conducive to the further improvement of retrieval and sorting performance; 2) Semi-supervised ranking learning method , seldom introduce unlabeled samples directly through the method of manifold learning or manifold learning is based on traditional similarity measure calculation, and does not introduce sample label information in the process of similarity calculation, which is not conducive to the full use of label information , so that the retrieval accuracy is not high

Images