Big data automatic matching method based on image recognition

A technology of image recognition and big data, applied in the field of big data, can solve the problems that it is difficult to avoid missing or wrong detection, relying on the language level of the checking personnel, and the huge difference in the language expression of the same topic, so as to avoid excessive consumption, The effect of reducing data complexity and high accuracy

Active Publication Date: 2022-04-05
深圳市对接平台科技发展有限公司
View PDF5 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0008] The above-mentioned existing technologies have the following problems: they all use text similarity to check plagiarism of scientific and technological projects, but the expressions of existing scientific and technological project achievements such as papers, monographs or patents are all written in various languages. The difference is huge, relying too much on the language level of the checker, it is difficult to avoid missing or wrong detection
However, there is no report on the use of image similarity as a means of plagiarism checking for scientific and technological items in the existing technology project plagiarism check

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
  • Big data automatic matching method based on image recognition
  • Big data automatic matching method based on image recognition
  • Big data automatic matching method based on image recognition

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0035] like figure 1 Shown, based on the result big data automatic matching method of image recognition, described method carries out the following steps:

[0036] Step 1: Obtain the historical achievement document, and extract the image information and text information in the historical achievement document;

[0037] Step 2: Carry out content recognition on the extracted text information to obtain a content recognition result;

[0038] Step 3: Based on the content identification results, perform content classification to obtain multiple content classification groups, generate multiple content tags based on the content classification results, and establish a content classification tree based on the content tags; the content classification tree is a tree structure database; each node in the content classification tree is a content label; each content label corresponds to a content classification group; the lower-level nodes of the content classification tree are a subset of th...

Embodiment 2

[0044] On the basis of the previous embodiment, in the step 2 and step 7: the method for performing content recognition on the extracted text information includes: locating the text area in the document to be matched or in the historical document; Classify; input the image information of the text area into the text recognition model, and obtain the text information output by the text recognition model; based on the classification result of the text area, classify the text information output by the text recognition model into the category to which the text area belongs.

[0045] specific,

Embodiment 3

[0047] On the basis of the previous embodiment, inputting the image information of the text area into the text recognition model includes: inputting the image information of the text area into the text recognition model corresponding to the category to which the text area belongs, wherein different categories correspond to different Text recognition model; said classifying the text area includes: inputting the image feature data of the text area into the first classification model to obtain the category information output by the first classification model; wherein, the first classification model uses a certain number of markers The image feature data of the category is obtained after training as sample data.

[0048] Specifically, image features mainly include image color features, texture features, shape features, and spatial relationship features.

[0049] The color feature is a global feature that describes the surface properties of the scene corresponding to the image or i...

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 present invention belongs to the field of big data technology, and in particular relates to an image recognition-based automatic matching method for achievement big data. The method performs the following steps: Step 1: Acquire historical achievement documents, and extract image information and text information in the historical achievement documents; Step 2 : Perform content recognition on the extracted text information to obtain the content recognition result. It establishes a content classification tree and a feature classification tree based on historical achievement documents, and then matches and queries the matching achievement documents to realize duplicate checking of the achievement documents; when performing matching queries, the present invention uses content matching first, and then feature matching This method can effectively improve system efficiency, because the speed of content matching is much higher than that of feature matching. If the content matching has been completed, there is no need to perform subsequent feature matching to avoid excessive consumption of system resources.

Description

technical field [0001] The invention belongs to the technical field of big data, and in particular relates to an automatic matching method of big data based on image recognition. Background technique [0002] According to statistics, the repetition rate of my country's scientific research projects is as high as 40%, and among the other 60%, the repetition rate with foreign countries accounts for more than 30%. Repeated project approval not only causes a large waste of scientific and technological resources, but also leads to the disorderly development of scientific research activities and a large number of low-level repetitions, which seriously damages the spirit of pioneering and innovative scientific research and hinders the pace of national scientific and technological development. [0003] At present, some scholars have studied the methods and mechanisms of some project plagiarism checks, and have made some progress. Commonly used methods for item duplication check main...

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): G06F16/33G06F16/35G06F16/532G06F16/55G06F16/583G06V10/75G06V10/764G06V10/40G06K9/62
CPCG06F16/3331G06F16/355G06F16/532G06F16/55G06F16/583G06F18/22G06F18/241
Inventor 张丰祥
Owner 深圳市对接平台科技发展有限公司
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap