Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Horrible image identification method and system based on visual significance analyses

An image recognition and analysis algorithm technology, applied in pattern recognition, computer network content security, can solve the problem of not having too many scary images, and achieve the effect of wide application prospects

Inactive Publication Date: 2013-12-11
INST OF AUTOMATION CHINESE ACAD OF SCI
View PDF1 Cites 12 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Compared with the filtering of pornographic images, there is not much research on the filtering of horror images in the Internet.

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
  • Horrible image identification method and system based on visual significance analyses
  • Horrible image identification method and system based on visual significance analyses
  • Horrible image identification method and system based on visual significance analyses

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0025] In order to make the object, technical solution and advantages of the present invention clearer, the present invention will be further described in detail below in conjunction with specific embodiments and with reference to the accompanying drawings.

[0026] figure 1 A flow chart of a horror image recognition method based on visual saliency analysis disclosed in the present invention is shown. like figure 1 As shown, the method includes the following steps:

[0027] Step 1: First collect as many horror and non-horror image samples as possible to form a training set; and let 7 users vote "horror, a little bit scary, not scary" for each horror image. If a horror image gets less than 4 votes "horror", it is removed from the training set to obtain a typical training set. Collections of horrific images are available from a large number of image sharing sites. Non-horror imagery needs to be as diverse as possible and can include: people, animals, landscapes, cartoons, et...

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 horrible image identification method and system based on visual significance analyses. The method comprises the steps that a training set containing horrible image samples and non-horrible image samples is built; visual significance analyses are conducted on each training sample image through an image significance analysis algorithm to acquire a visual significance image of each training sample image; each training sample image is divided into small W*W image blocks, virtual and affective characteristics are conducted on each image block, and virtual word histogram representation of each training sample image is acquired through a packet representation model based on visual significance; a classification model of horrible images is acquired through an acquired virtual word histogram of each training sample image and a label training support vector machine corresponding to the acquired virtual word histogram, and new testing image is identified through the classification model to judge whether the new testing image is a horrible image or not. The horrible image identification method and system based on visual significance analyses can be applied to the fields like horrible web image filtration and affective and semantic image recognition, and has wide application prospects.

Description

technical field [0001] The invention relates to the technical fields of pattern recognition, computer network content security, and image emotion understanding, in particular to a horror image recognition method and system based on visual salience analysis. Background technique [0002] In the past few decades, the explosion of information and resources on the Internet has made it very convenient for us to share multimedia information such as text, images and videos across regions. However, the lack of Internet information supervision has led to more and more harmful and illegal content, such as pornography, violence, terror, terrorism, etc., flooding the entire Internet. Therefore, an effective bad web page filtering system is of great significance to prevent the illegal dissemination of bad information and protect the physical and mental health of young people. [0003] Recently, scary texts, images, and videos widely distributed on the Internet are invading the daily liv...

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): G06K9/62
Inventor 李兵胡卫明吴偶
Owner INST OF AUTOMATION CHINESE ACAD OF SCI
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com