TN2022000316A1 - Intelligent and embedded system for video surveillance of high-risk sites based on visible and thermal bispectral detection. - Google Patents

Intelligent and embedded system for video surveillance of high-risk sites based on visible and thermal bispectral detection.

Info

Publication number
TN2022000316A1
TN2022000316A1 TNP/2022/000316A TN2022000316A TN2022000316A1 TN 2022000316 A1 TN2022000316 A1 TN 2022000316A1 TN 2022000316 A TN2022000316 A TN 2022000316A TN 2022000316 A1 TN2022000316 A1 TN 2022000316A1
Authority
TN
Tunisia
Prior art keywords
thermal
visible
proposed
intelligent
improve
Prior art date
Application number
TNP/2022/000316A
Other languages
French (fr)
Inventor
Ben Amara Najoua Essoukri
Hajer Fradi
Mohamed Amine Marnissi
Anis Sahbani
Original Assignee
Ecole Nat Dingenieurs De Sousse
Enova Robotics
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Ecole Nat Dingenieurs De Sousse, Enova Robotics filed Critical Ecole Nat Dingenieurs De Sousse
Priority to TNP/2022/000316A priority Critical patent/TN2022000316A1/en
Publication of TN2022000316A1 publication Critical patent/TN2022000316A1/en

Links

Landscapes

  • Image Analysis (AREA)

Abstract

La présente invention propose un système embarqué et intelligent de surveillance des sites à hauts risques qui détecte les cas d'intrusions dans les sites surveillés. Ce système est basé sur une technique d'analyse de vidéos capturées par des caméras bispectrales thermiques et visibles. Le système proposé permet, d'une part, d'améliorer la qualité visuelle des images thermiques et, d'autre part, de détecter les personnes dans les flux vidéo thermiques et visibles. De plus, il offre une surveillance intelligente et permanente en temps réel avec une prise de décision adéquate en cas de situations suspectes ou anormales. Le système proposé comprend deux modules : Le premier module portant sur l'amélioration de la qualité des images consiste à proposer une architecture de réseaux de neurones profonds pour améliorer la qualité des images thermiques. Cette architecture est constituée de deux étapes principales, la première est destinée à améliorer le contraste de l'image et la seconde étape vise à supprimer le bruit. Dans le second module, une technique de transfert des caractéristiques du domaine de l'imagerie visible au domaine thermique, est proposée. Ce choix est justifié par l'insuffisance de bases de données et de détecteurs pré-entraînés dans le spectre thermique. Dans ce module, un détecteur de personnes de type adaptatif est proposé. Il est fondé sur l'alignement des caractéristiques sur plusieurs niveaux pour bien couvrir le décalage entre les deux domaines.The present invention proposes an embedded and intelligent system for monitoring high-risk sites that detects cases of intrusions into the monitored sites. This system is based on a technique for analyzing videos captured by thermal and visible bispectral cameras. The proposed system makes it possible, on the one hand, to improve the visual quality of thermal images and, on the other hand, to detect people in thermal and visible video streams. In addition, it offers intelligent and permanent monitoring in real time with adequate decision-making in the event of suspicious or abnormal situations. The proposed system comprises two modules: The first module relating to the improvement of image quality consists of proposing a deep neural network architecture to improve the quality of thermal images. This architecture consists of two main steps, the first is intended to improve the contrast of the image and the second step aims to remove noise. In the second module, a technique for transferring characteristics from the visible imaging domain to the thermal domain is proposed. This choice is justified by the lack of databases and pre-trained detectors in the thermal spectrum. In this module, an adaptive type people detector is proposed. It is based on the alignment of features on several levels to cover the gap between the two domains.

TNP/2022/000316A 2022-11-25 2022-11-25 Intelligent and embedded system for video surveillance of high-risk sites based on visible and thermal bispectral detection. TN2022000316A1 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
TNP/2022/000316A TN2022000316A1 (en) 2022-11-25 2022-11-25 Intelligent and embedded system for video surveillance of high-risk sites based on visible and thermal bispectral detection.

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
TNP/2022/000316A TN2022000316A1 (en) 2022-11-25 2022-11-25 Intelligent and embedded system for video surveillance of high-risk sites based on visible and thermal bispectral detection.

Publications (1)

Publication Number Publication Date
TN2022000316A1 true TN2022000316A1 (en) 2024-07-04

Family

ID=93564189

Family Applications (1)

Application Number Title Priority Date Filing Date
TNP/2022/000316A TN2022000316A1 (en) 2022-11-25 2022-11-25 Intelligent and embedded system for video surveillance of high-risk sites based on visible and thermal bispectral detection.

Country Status (1)

Country Link
TN (1) TN2022000316A1 (en)

Similar Documents

Publication Publication Date Title
JP6469975B2 (en) Image monitoring apparatus, image monitoring system, and image monitoring method
Wang et al. Tracking anonymous peer-to-peer voip calls on the internet
US6081606A (en) Apparatus and a method for detecting motion within an image sequence
US6700487B2 (en) Method and apparatus to select the best video frame to transmit to a remote station for CCTV based residential security monitoring
JP3974038B2 (en) Intruder detection using trajectory analysis in surveillance and reconnaissance systems
US8447855B2 (en) Method, system and computer program product for preventing SIP attacks
JP4670943B2 (en) Monitoring device and disturbance detection method
CN107040544A (en) A kind of intrusion detection method based on flow, apparatus and system
TN2022000316A1 (en) Intelligent and embedded system for video surveillance of high-risk sites based on visible and thermal bispectral detection.
US20060114322A1 (en) Wide area surveillance system
EP1444831A4 (en) Security system and warning method using edge detection of image signal
KR20240009910A (en) Intelligent video surveillance system
Jainsky et al. Towards digital video steganalysis using asymptotic memoryless detection
WO2006109162A2 (en) Distributed smart video surveillance system
CN112449155A (en) Video monitoring method and system for protecting privacy of personnel
KR101929212B1 (en) Apparatus and method for masking moving object
KR100994418B1 (en) Image Processing System and Method for Intrusion Detection of Buildings
WO2009101325A1 (en) Secured method for event control
US11823538B2 (en) Systems and methods for incorporating passive wireless monitoring with video surveillance
Stadler et al. Real time video based smoke detection using double optical flow estimation
KR20020048574A (en) An Unmanned Security System
Lee et al. The method of P2P traffic detecting for P2P harmful contents prevention
Nagothu et al. Detecting Malicious False Frame Injection Attacks on Video Surveillance at the Edge Using Electrical Network Frequency Signals
JP4429232B2 (en) Device for detecting impulse noise in video signals
Velastin Special section on intelligent distributed surveillance systems