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

System and method for detecting violent events in image based on anchor node

An event detection and anchor node technology, applied in computer parts, instruments, characters and pattern recognition, etc., can solve the problems of being easily affected by noise, unable to achieve real-time effects, etc., and achieve the effect of improving efficiency and accuracy

Active Publication Date: 2019-01-01
四川电科维云信息技术有限公司
View PDF9 Cites 5 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

At present, the traditional violent behavior detection mainly adopts the method based on artificially designed features. Although the recognition accuracy is high, it also has some defects, such as: it cannot achieve real-time effects, and it is easily affected by noise.

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
  • System and method for detecting violent events in image based on anchor node
  • System and method for detecting violent events in image based on anchor node
  • System and method for detecting violent events in image based on anchor node

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0068] The scene described in this embodiment is: a distributed application scene that does not need real-time processing, such as after the cameras of several streets capture images, the images are transmitted to a host to store image files, and the host processes the images of a small number of cameras corresponding to it. image and obtain the prediction results, and temporarily store the results locally, and upload the temporarily stored prediction results to their respective databases at regular intervals of several hours, and provide query functions for clients through the reverse proxy server. The client computer does not feel the existence of multiple databases, and can directly access the data of a large number of cameras, such as figure 2 As shown, the specific steps are as follows:

[0069] Before the deployment of the system, collect a large number of image data sets used for training on the network or other channels, organize the data sets, accurately classify all...

Embodiment 2

[0074] The scene described in this embodiment is: some special places that require real-time centralized processing of images. Since the area is generally small and there are fewer cameras overall, the demand for real-time monitoring of violent events is high. Therefore, centralized real-time prediction of these images is required. Each video streaming server is only connected to a small number of cameras. These servers are only responsible for image storage and a small amount of processing to obtain hash codes. These hash codes will be marked and sent to the same detection server. The server quickly detects and simultaneously Stored in the database, the client will poll the database and issue a warning when a violent event is found. refer to image 3 ,Specific steps are as follows:

[0075] Before the deployment of the system, collect a large number of image data sets used for training on the network or other channels, organize the data sets, accurately classify all image da...

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 relates to a video image processing technology, which discloses an image violence event detection method based on an anchor node, which automatically analyzes and detects the violence behavior in a video stream, improves the real-time performance and accuracy of detection, and can be adapted to a plurality of different monitoring environments. The method comprises the following steps: a. obtaining sample features according to all feature vectors of an image data set and soliciting anchor nodes according to the sample features; B. calculating a hash code corresponding to the sample in the data set according to the similarity between the anchor node and the sample in the image data set; C. using the hash codes corresponding to each sample of the image data set as inputs, and using the trained SVM model to predict the image, so as to judge whether there is a violent event in the image. In addition, the invention also discloses an image violence event detection system based on an anchor node, which is suitable for a plurality of application scenes.

Description

technical field [0001] The invention relates to video image processing technology, in particular to a system and method for detecting violent events in images based on anchor nodes. Background technique [0002] With the extensive use of surveillance systems, video data has exploded. The role of the monitoring system is to perform target detection and abnormal behavior detection. With the rapid growth of data, the traditional method of relying on manual monitoring has become increasingly difficult and inefficient. Therefore, research on surveillance systems relying on artificial intelligence has become a hotspot. Among them, the detection of human violence is a very important research direction. [0003] Since the action of violent behavior is much more complicated than simple running and jumping behaviors, how to detect violent behavior is also a difficult point in related research. At present, the traditional violent behavior detection mainly adopts the method based on...

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): G06K9/00G06K9/62
CPCG06V20/44G06V20/41G06F18/23G06F18/2411
Inventor 殷光强田玲张栗粽包益全
Owner 四川电科维云信息技术有限公司
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