Image retrieval method and system based on big data

Abstract

The invention relates to the technical field of image retrieval, and discloses an image retrieval method based on big data, which comprises the following steps: acquiring massive image data, and performing distributed storage on the massive image data; preprocessing of image graying and gray stretching is carried out on the stored massive image data; processing the preprocessed image data by usinga multi-label semantic extraction algorithm based on an auto-encoder to obtain multi-label semantic information of the image; establishing an image information connection graph according to the multi-label semantic information of the image; and storing the image data combined with the image information connection graph by using a deep hash-based data storage method, thereby taking the hash codedvalue as an image feature index, and performing image retrieval according to the image feature index. The invention further provides an image retrieval system based on the big data. According to the invention, image retrieval is realized.

Description

technical field [0001] The invention relates to the technical field of image retrieval, in particular to an image retrieval method and system based on big data. Background technique [0002] In the Internet age, a large number of instant messaging software, office software, shopping platforms, game platforms, etc. have greatly facilitated and enriched people's study, life and work, and at the same time produced a large amount of multi-category, heterogeneous, and unstructured data; The explosive growth of image data has brought great convenience to people's life because of its intuition and large amount of information. high demands. [0003] Most of the current search engines retrieve images based on text keywords, and the retrieval keywords often do not match the real semantics of the image, so the retrieval performance is affected; at the same time, the current image retrieval mainly uses the method of traversing the images sequentially, without much A good index mechani...

AI Technical Summary

Problems solved by technology

[0002] In the Internet age, a large number of instant messaging software, office software, shopping platforms, game platforms, etc. have greatly facilitated and enriched people's study, life and work, and at the same time produced a large amount of multi-category, heterogeneous, and unstructured data; The explosive growth of image data has brought great convenience to people's life because of its intuition and large amount of information. high demands

[0003] Most of the current search engines retrieve images based on text keywords, and the retrieval keywords often do not match the real semantics of the image, so the retrieval performance is affected; at the same time, the current image retrieval mainly uses the method of traversing the images sequentially, without much A good indexing mechanism for indexing also increases the load on the retrieval system, and most of the traditional image retrieval methods build image indexes offline at regular intervals based on existing data, and there is a problem of poor timeliness for newly added image retrieval

Images