A Smart Watchdog - Intruder Detection System

 
 
 
  • Abstract
  • Keywords
  • References
  • PDF
  • Abstract


    In todays world, Security is a matter of great concern. Security controls play a vital role in protecting resources from espionage, sabotage, damage and theft. Our proposed system is to develop a security system with improved facilities, which tries to eliminate the limitations posed by the existing security systems. The current manual security system depends mostly on human involvement, which is prone to error, and the security is concentrated only at the front door which requires subjects cooperation. To solve these issues we have proposed a Smart Watchdog System. The system watches the environment, and if there is a human activity, the system captures it. The system automatically detects faces of the individual from the activity using firmware. We have planned to maintain the database of authorised inmates and workers of a place and verifies of every individual arriver. This feature enables the system to automatically recognises the unauthorised users and gives an alert when it encounters entry of unauthorised users even without the human assistance. The system also detects the unauthorised entry in the mass. The entire system is planned to be ported to Raspberry-Pi based Embedded System supported with DC power back up. This method can be employed in ladies hostels as well as to the secured places like the data centre, atomic research centre and military where the unauthorised entry is restricted.


  • Keywords


    Security system, Biometrics, Face Recognition, Machine learning, Image processing, Deep neural network

  • References


      [1] K. Simonyan and A. Zisserman. Very deep convolutional networks for large-scale image recognition In International Conference on Learning Representations, 2015

      [2] A. Krizhevsky, I. Sutskever, and G. E. Hinton. ImageNet classification with deep convolutional neural networks. In NIPS, pages 1106-1114, 2012

      [3] Sujata P.Salunkhe and Bhausaheb S. Sonawane and Geeta N. Sonawane A Review on Various Techniques of women security.


 

View

Download

Article ID: 18971
 
DOI: 10.14419/ijet.v7i3.34.18971




Copyright © 2012-2015 Science Publishing Corporation Inc. All rights reserved.