Abnormal behavior identification method and system combined with laboratory video data analysis

By dividing image sequences and identifying the laboratory videos, combined with convolutional neural networks and twin networks, the problem of slow response and low accuracy of abnormal behavior recognition in laboratory environments in traditional methods is solved, and efficient and accurate abnormal behavior recognition and security prevention are achieved.

CN120014515AActive Publication Date: 2025-05-16EASTERN LIAONING UNIV
View PDF 8 Cites 1 Cited by

Patent Information

Application Number
CN202510097630.3
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-01-22
Publication Date
2025-05-16
Estimated Expiration
2045-01-22

Smart Images

  • Figure CN120014515A_ABST
    Figure CN120014515A_ABST
Patent Text Reader

Abstract

The invention relates to an abnormal behavior recognition method and system in combination with laboratory video data analysis, and relates to the field of video recognition, comprising: performing key image extraction on a plurality of part monitoring image sequences to obtain a plurality of key part monitoring image sequences, performing key timestamp extraction on a monitoring timestamp sequence to obtain a plurality of key part monitoring image sequences; obtaining a plurality of key monitoring timestamp sequences; performing action recognition according to the plurality of key part monitoring image sequences to obtain a plurality of behavior recognition results, and calculating according to the plurality of key monitoring timestamp sequences to obtain a plurality of change speeds; according to the multiple behavior recognition results and the multiple change speeds, the abnormal behavior probability is obtained through calculation, and a recognition result is obtained through judgment. Through the method, the problems that potential dangerous behaviors cannot be quickly and accurately identified in a dynamic and complex laboratory environment, response is slow and identification accuracy is low in a traditional method can be solved; the efficiency and accuracy of abnormal behavior recognition can be remarkably improved, and potential safety hazards are effectively prevented.
Need to check novelty before this filing date? Find Prior Art

Citation Information

Patent Citations

  • Suspicious behavior detection method and suspicious behavior detector

    JP2011035571A

  • Volatility-based Classifier for Security Solutions

    US20160021124A1

  • Explaining Machine Learning Models by Tracked Behavioral Latent Features

    US20190156196A1

  • Method and apparatus for recognizing body movement

    US20200042776A1

  • Spatial regime change detection for video analytics

    US20240135546A1

Cited By

  • LIMS laboratory process maintenance method based on adaptive learning model

    CN121482712A