Video Surveillance System with Auto Informing Feature

EasyChair Preprint 2144

Abstract

The present document represents a thorough study of making of an efficient surveillance system along with a feature of automatically informing the owner about the suspicious movement. In this moving world normally people are suffering with the availability of time, so if any crime had happened at the site, it will take many days of searching for finding actual presence of criminal. And thus a good chance for those burglars to flee away in order to protect themselves. For making the task possible we chose Python as our weapon for this battle and used different efficient techniques like, COCO data set for getting labelled and annotated images, LabelImg for making the annotation set of images, TensorFlow, Object Detection API for object Detection and Faster RCNN for training as faster RCNN has shown the highest accuracy for the COCO data set so far. Owner can be informed in two ways : Either send a message to him via mail or phone or call at the time of suspicious image capturing. Here both of these cases are used : For mail we did it via SMTP and for phone calls we used Twilio which provides us registered phone no. and we can make both outbound and inbound calls. After using all the mentioned things and making the model in a way described above we found that faster RCNN is much accurate than the other conventional methods. The results have come very good as RCNN show 86.7% accuracy and a 100% have come out with the informing module as there simply the mail will be sent to the one whose mail is given in the code and the same is for Twilio calling.

Keyphrases: COCO, Preprocessing, RCNN, SMTP, TensorFlow, Twilio, annotation, labelImg

@booklet{EasyChair:2144,
  author    = {Lokesh Chouhan and Ekta Tyagi and Deeksha Rana and Shubham Poddar and Chandranshu Malhotra and Vikas Kumar Sahu and Jayash Verma},
  title     = {Video Surveillance System with Auto Informing Feature},
  howpublished = {EasyChair Preprint 2144},
  year      = {EasyChair, 2019}}