Automatic image annotation method based on deep learning and canonical correlation analysis

Abstract

The invention discloses an automatic image annotation method based on deep learning and canonical correlation analysis. The method includes: using a depth Boltzmann machine to extract the high-level feature vectors of images and annotation words, selecting multiple Bernoulli distribution to fit annotation word samples, and selecting Gaussian distribution to fit image features; performing canonical correlation analysis on the high-level features of the images and the annotation words; calculating the Mahalanobis distance between to-be-annotated images and training set images in canonical variable space, and performing weighted calculation according to the distance to obtain high-level annotation word features; generating image annotation words through mean field estimation. The depth Boltzmann machine comprises I-DBM and T-DBM which are respectively used for extracting the high-level feature vectors of the images and the annotation words. Each of the I-DBM and the T-DBM sequentially comprises a visible layer, a first hidden unit layer and a second hidden unit layer from bottom to top. By the method, the problem of 'semantic gap' during image semantic annotation can be solved effectively, and annotation accuracy is increased.

Description

technical field [0001] The invention relates to image automatic labeling and retrieval technology, in particular to an image automatic labeling method based on deep learning and canonical correlation analysis. Background technique [0002] With the growth of image data showing a geometric progression, how to effectively manage and retrieve these image data has become a research hotspot in the construction of information technology. Although the current content-based image retrieval technology has made great progress, and there are also a variety of prototypes, technologies and retrieval products for civilian use, but due to the main problem-the "semantic gap" has not made a fundamental breakthrough, resulting in its retrieval effect and method. Not ideal. To overcome these problems, the best solution is to add text semantic information related to the image content to the image, that is, image annotation. In view of the problems of strong subjectivity and low labeling effic...

AI Technical Summary

Problems solved by technology

From the experimental results, compared with other deep learning models, the effect of this model is better, but there is still a gap compared with the classic automatic image labeling algorithm, because the vocabulary model and the top-level feature fusion mechanism are not suitable for automatic image labeling tasks