Unlock instant, AI-driven research and patent intelligence for your innovation.

On-line multi-quantization image retrieval method

An image retrieval and vector technology, applied in digital data information retrieval, instruments, calculations, etc., can solve problems such as low retrieval accuracy and inability to make good use of streaming data semantic correlation information, so as to improve search results and learning efficiency , the effect of reducing the burden

Active Publication Date: 2022-05-13
UNIV OF SCI & TECH OF CHINA
View PDF5 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] The existing online hashing method only needs to recalculate the binary code of the stream data after updating the hash function. Compared with the existing hashing method, although the online hashing method can improve the efficiency of stream data learning, the learning efficiency There is still a lot of room for improvement; in addition, because streaming data can only be used once in training, existing online hashing methods often cannot make good use of the semantic association information between streaming data, resulting in low retrieval accuracy

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • On-line multi-quantization image retrieval method
  • On-line multi-quantization image retrieval method
  • On-line multi-quantization image retrieval method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0015] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only some of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by persons of ordinary skill in the art without making creative efforts belong to the protection scope of the present invention.

[0016] An embodiment of the present invention provides an online multi-quantified image retrieval method, which can be applied to online search software to quickly and accurately retrieve query images. In terms of implementation, it can be installed on the user's work computer in the form of software to provide real-time retrieval; it can also be installed on the background server to provide large-scale background retrieval.

[0017] Such as figur...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses an online multi-quantization image retrieval method, which uses the multi-quantization technology based on stream data correlation learning for large-scale online image retrieval, effectively improves the search effect, and does not update the binary code during the learning process, only Updating the small-scale codebook can reduce the burden of recalculating the binary code of streaming data and improve learning efficiency.

Description

technical field [0001] The invention relates to the technical field of image retrieval, in particular to an online multi-quantified image retrieval method. Background technique [0002] In many practical applications, multimedia data often exists in the form of streams. For example, well-known search engine companies (Google, etc.) will have a large number of new web pages and images arriving at the data center every day. In such an environment, the query image must continuously respond to the user's retrieval based on the current total data flow. [0003] However, currently in the field of image retrieval, most existing hashing methods are based on batch learning. This also means that when new data arrives, existing hashing methods have to accumulate all available data and retrain a new hash function, which is obviously a less efficient way of learning from streaming data. [0004] The existing online hashing method only needs to recalculate the binary code of the stream...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Patents(China)
IPC IPC(8): G06F16/532G06V10/28G06V10/74G06K9/62
CPCG06F16/532G06F18/23213G06F18/22
Inventor 张勇东李攀登谢洪涛李岩
Owner UNIV OF SCI & TECH OF CHINA