Abnormal behavior detection method based on generative adversarial network

A detection method and generative technology, applied in the field of video analysis, can solve the problem of unsatisfactory video frame prediction

Inactive Publication Date: 2020-01-17
NANJING UNIV OF POSTS & TELECOMM
View PDF0 Cites 21 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Current video frame prediction may be far from satisfactory

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
  • Abnormal behavior detection method based on generative adversarial network
  • Abnormal behavior detection method based on generative adversarial network
  • Abnormal behavior detection method based on generative adversarial network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0056] Mathematically, given a continuous t frame I 1 , I 2 ,..., I t , we stack all these frames in sequence and use them to predict future frames I t+1 ,we use represents our prediction. Let close to I t+1 , we minimize the distance between them in terms of intensity and gradient. To preserve the temporal coherence between adjacent frames, we strengthen the I t+1 with I t The optical flow between to I t near. Finally, the difference between the prediction of the future frame and itself determines whether the future frame is normal or abnormal. The structure of our network framework is as follows figure 1 shown. Next, we will describe all the components of the framework in detail.

[0057] 1. Build a generator network

[0058] The frame generation network or image generation network commonly used in existing work usually contains two modules: i) an encoder that extracts features by gradually reducing the spatial resolution; ii) a decoder that gradually resto...

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 an abnormal behavior detection method based on a generative adversarial network (GAN). The method comprises the following steps that firstly, a U-Net network is built and serves as a generator module in a GAN, and not only is appearance (space) constraint used, but also motion (time) constraint is introduced; secondly, a patch discriminator (patch discriminator) is adoptedas a discriminator module in the GAN; and then, alternately performing adversarial training on the generator and the discriminator until the discriminator cannot distinguish the generated frame and the real frame. And finally, carrying out an abnormal behavior detection experiment through the trained GAN model. Experimental results on three public and available anomaly detection data sets show that the method provided by the invention effectively improves the accuracy of anomaly behavior detection.

Description

technical field [0001] The present invention relates to the technical field of video analysis, relates to a method for detecting abnormal behavior in a surveillance video scene, and in particular to a method for detecting abnormal behavior based on a generative confrontation network. Background technique [0002] Traditional video surveillance mainly relies on artificial monitoring of abnormal behaviors in the scene, which not only requires extremely high labor costs, but also easily causes visual fatigue, and even causes some abnormal behaviors to not be observed in time. Abnormal behavior detection and analysis aims to automatically detect abnormal behaviors in surveillance scenes through algorithms such as video signal processing and machine learning, so as to help people take corresponding measures in a timely manner. Therefore, it is of great significance and value to develop an abnormal behavior detection algorithm for surveillance video. [0003] Anomaly detection in...

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
CPCG06V20/42G06F18/214
Inventor 卢博文郭文波朱松豪
Owner NANJING UNIV OF POSTS & TELECOMM
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