Knowledge base construction method based on image recognition

A construction method and image recognition technology, applied in the field of knowledge base construction based on image recognition, can solve problems such as low knowledge reusability, low efficiency, and unrealizable accumulation of knowledge, so as to expand the construction method, realize accumulation and reliability The effect of reusability

Active Publication Date: 2020-04-21
UNIV OF ELECTRONICS SCI & TECH OF CHINA
View PDF6 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

But at present, in order to obtain information and knowledge in images, people basically can only identify useful information through long-term viewing, and artificially integrate information to make judgments
This way of acquiring knowledge is not only inefficient, but also has a low degree of reusability of knowledge, and the accumulation of knowledge cannot be realized.

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
  • Knowledge base construction method based on image recognition
  • Knowledge base construction method based on image recognition
  • Knowledge base construction method based on image recognition

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0033] figure 1 It is a flowchart of a method for building a knowledge base based on image recognition in the present invention.

[0034] In this example, if figure 1 Shown, a kind of knowledge base construction method based on image recognition of the present invention comprises the following steps:

[0035] S1. Obtain the target image

[0036] Obtain images containing multiple scenes and entities related to the knowledge base to be constructed, and each image contains multiple entities E 1 ,E 2 ,...,E i , ..., and each entity has multiple attribute values ​​A 1,1 、A 1,2 ,...,A 2,1 、A 2,2 ,...,A i,j ,...;

[0037] In this embodiment, each image contains a scene such as the sky, and multiple entities such as people, objects, animals, etc., and each image needs to meet the basic definition, that is, the definition sufficient for human eyes to recognize.

[0038] S2. Target image preprocessing

[0039] First convert each image into a grayscale image, then smooth and ...

example

[0059] This embodiment builds a personnel information knowledge base based on monitoring images, which is used to assist the public security organs to quickly identify suspects, describe the suspect's movement trajectory, and help them quickly solve cases.

[0060] First, the image set collected from the case-related monitoring equipment is preprocessed according to the above method.

[0061] Second, the image is recognized using a neural network model. In this embodiment, the main task used for entity recognition is character recognition, and the attributes involved include: human body shape (fat and thin degree), clothing color, height size, which can be subdivided into: human body shape (thin, normal) , fat), clothing color (red, orange, yellow, green, blue, blue, purple), height size (less than 1.5 meters, 1.5-1.6 meters, 1.6-1.7 meters, 1.7-1.8 meters, 1.8-1.9 meters, 1.9 meters meters or more).

[0062] Finally, the coincidence degree of the pixel range of the entity a...

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 a knowledge base construction method based on image recognition. The method includes: obtaining a series of images related to a target knowledge base and perform image preprocessing; identifying the input image by using three neural network groups to obtain scene, entity and attribute information of image content, matching the output entity and attribute according to the overlap ratio of the pixel range occupied by the entity and the pixel range occupied by the attribute, and completing the construction of a knowledge base based on a scene, entity and attribute triple.Therefore, by taking the image as a knowledge base construction source, the construction method of the knowledge base is expanded, the redundancy of the knowledge base can be reduced, the query efficiency can be improved, and the accumulation and reusability of image knowledge can be realized.

Description

technical field [0001] The invention belongs to the technical field of machine learning and information processing, and more specifically relates to a method for constructing a knowledge base based on image recognition. Background technique [0002] In the context of the information age, Internet technology has not only shortened the communication distances around the world step by step, enabling information to be disseminated and communicated quickly, but also accompanied by the rapid popularization of cheap electronic terminals resulting from the rapid iteration of industrial technology, each terminal Users not only become recipients of information, but also naturally become producers of information. One of the great changes that followed was the explosive export of knowledge. On the one hand, blowout knowledge output can be used as the data cornerstone of new technologies. For example, the deep learning technology that is most concerned by the society today is a represen...

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 Applications(China)
IPC IPC(8): G06F16/583G06K9/62G06N3/04G06N3/08
CPCG06F16/583G06N3/08G06V10/751G06N3/045
Inventor 郑文锋杨波陈小兵刘珊曾庆川
Owner UNIV OF ELECTRONICS SCI & TECH OF CHINA
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