A Visual Media Processing Method Based on Semantic Index Accuracy Enhancement

Abstract

The invention relates to a visual media processing method based on semantic index precision enhancement, and belongs to the technical field of visual media processing. The method comprises the following steps: firstly, carrying out semantic indexing on objects and scenes contained in the visual media, constructing a confidence coefficient matrix, judging and screening out a part of elements through a threshold value, and carrying out re-estimation on the screened matrix by applying a weighted non-negative matrix decomposition method with external correlation smoothness to complete global precision enhancement, and carrying out local precision enhancement out by using a similarity propagation method and using a similarity relationship between samples according to a result after global precision enhancement. According to the method, various semantic relations are utilized, the accuracy of visual media semantic indexing is enhanced, the external correlation of concepts is fully utilized,and the effect of semantic indexing precision enhancement is improved. According to the method, global precision enhancement and local precision enhancement based on external correlation smoothing areorganically combined, the calculation complexity is low, the expandability is high, and the method is suitable for practical industrial application.

Description

technical field [0001] The invention relates to a visual media processing method based on semantic index precision enhancement, and belongs to the technical field of visual media processing. Background technique [0002] Accuracy enhancement of semantic indexing of visual media is a key technology to further improve the indexing results by utilizing the correlation of concepts. The function of this index enhancement technology is to give the result of concept detection of visual media, adjust and improve the quantitative relationship of concepts applied to the result of concept detection, so as to achieve the purpose of effectively indexing a large number of visual media. The patent application titled "A Accuracy Enhancement Method for Visual Media Semantic Indexing" (Application No.: 201610108055.3) discloses a combination of global enhancement and local improve. However, in the process of enhancement, this method cannot effectively utilize the correlation between semanti...

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to propose a visual media processing method based on semantic index precision enhancement, and improve the existing visual media processing method to deal with the low and unbalanced concept detection accuracy caused by the current single-type concept detector. problem, which can not only deal with ordinary indexing accuracy enhancement tasks, but also make full use of the obtained concept correlation when external knowledge or labeled data is available, and can be used to enhance the accuracy of semantic indexing results in different application domains.

Images