Supercharge Your Innovation With Domain-Expert AI Agents!

Violence and terrorism picture safety detection system based on deep learning

A technology of security detection and deep learning, which is applied in the field of information security, can solve the problem that violent image detection products cannot be detected across domains, and achieve the effect of improving the recognition range and recognition accuracy, accurate performance, and reducing the false positive rate

Pending Publication Date: 2021-06-04
SHANGHAI JIAO TONG UNIV
View PDF4 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] Aiming at the above-mentioned deficiencies in the prior art, the present invention proposes a security detection system for violent terrorism pictures based on deep learning. Aiming at the shortcomings that the current violent terrorism picture detection products cannot be detected across domains, the violent terrorism picture detection and the analysis of the degree of violence and terrorism, The technical modules of violent terror scene classification and gun recognition, graphic analysis in natural scenes, violent terrorist political leader sample generation, recognition, and violent terrorist flag recognition are organically combined

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
  • Violence and terrorism picture safety detection system based on deep learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0015] like figure 1 As shown, this embodiment includes: violent terror picture detection and violent terror degree analysis module, violent terrorist scene classification and gun identification module, graphic analysis module under natural scene, violent terrorist political leader sample generation module, violent terrorist political leader identification Module and terrorist flag recognition module, in which: violent terror picture detection and violent terror degree analysis module, graphic analysis module in natural scenes, violent terrorist political leader sample generation module and violent terrorist flag recognition module are connected in parallel and receive picture information respectively, The violent terror scene classification and gun recognition module is connected in series with the violent terrorist picture detection and violent terror degree analysis module and receives the pictures judged as violent by the violent terrorist picture detection and violent terr...

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

A violence and terrorism picture safety detection system based on deep learning comprises a violence and terrorism picture detection and violence and terrorism degree analysis module, a violence and terrorism scene classification and gun identification module, an image-text analysis module in a natural scene, a violence and terrorism political head sample generation module, a violence and terrorism political head identification module and a violence and terrorism flag identification module, wherein the violence and terrorism picture detection and violence and terrorism degree analysis module, the image-text analysis module in the natural scene, the violence and terrorism political head sample generation module and the violence and terrorism flag identification module are connected in parallel and respectively receive picture information. According to the method, the pictures under various scenes serve as input, whether the pictures are violent and terrorism pictures or not can be judged in a short time, violent and terrorism elements existing in the pictures can be detected and identified, the recognition range and recognition accuracy of a picture safety detection system are remarkably improved, the recognition efficiency is high, flexibility is high, model updating is convenient, and each module model can be regularly trained in a strengthened manner according to use conditions, so that the system performance is improved.

Description

technical field [0001] The present invention relates to a technology in the field of information security, in particular to a security detection system for violent and terrorist pictures based on deep learning. Background technique [0002] Image detection systems for specific elements of violence and terror are increasingly used in various fields. It can not only be used for the prevention, emergency response, evidence collection and filing of public security incidents, but also provide a data basis for retrospective and reconstruction after the event. As an emerging research hotspot in the field of information security, terrorist image detection technology is one of the most challenging issues in the field of content security. It describes the process of determining whether a picture to be detected contains violent and terrorist elements, and identify and classify specific violent and terrorist elements. Due to the variety of elements involved in violent and terrorist pic...

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
IPC IPC(8): G06K9/00G06K9/20G06K9/62G06N3/04G06N3/08
CPCG06N3/04G06N3/08G06V40/168G06V40/172G06V10/22G06F18/2411G06F18/214
Inventor 郭捷陈欣然徐扬沈琪孙泽坤吴管浩邱卫东黄征
Owner SHANGHAI JIAO TONG UNIV
Features
  • R&D
  • Intellectual Property
  • Life Sciences
  • Materials
  • Tech Scout
Why Patsnap Eureka
  • Unparalleled Data Quality
  • Higher Quality Content
  • 60% Fewer Hallucinations
Social media
Patsnap Eureka Blog
Learn More