Implementation of a Remote Real-time Surveillance Security System for Intruder Detection

Abstract

Family security is one of the key goals of a digital home system. At present, webcams are widely used in home remote monitoring systems. This is due to the fact that its passive monitoring nature leaves a potential security problem and cannot be used for providing proactive security. To solve this problem, this paper proposes the design of a remote embedded intelligent security monitoring system based on the background modeling algorithm in computer vision which can proactively detect intruders. The system uses the ViBe algorithm to model the background image which is acquired from the camera, and then carries out the object detection in the monitored environment. If a moving object (including people) is detected, the system will automatically sound an alarm, while sending a message or calling the user to take preventive measures. Users can log in to the server through the mobile application to detect the intruder from the intercepted picture, and to get a better understanding of the situation. After testing, the system can accurately determine whether there is something intruding the monitoring area or not. It is robust, real-time, and with a good practical application value.

Existing System

Mobile Phone Video Surveillance System for Home Security.

Proposed System

The purpose of this paper is to design an economical and practical remote intelligent monitoring system based on the background modeling algorithm. When the user is out, the task of the whole system is to decide whether there is a break in. In cases where such events are detected, the system will record and save the key images for the user to download and view, to understand what happened in their house. This method avoids the problem of network security, for the embedded ARM development board, which not only solves the problem of the bandwidth of data transmission, but can also reasonably utilize the storage space. We research the application of background modeling algorithm for remote intelligent monitoring. We implement an intelligent video analysis technology and embedded technology to implement safety and security at homes and other environments. This system has the advantage that the traditional monitoring system do not have with respect to its being intelligent in detecting intruders, and its ability to integrate with other systems. In the ARM development board, the ViBe algorithm processes the real-time video image, and inspects the monitoring area intelligently. If a moving object is detected, the development board will sound an alarm, which can play a deterrent role to the intruder. At the same time, the system will automatically save the pictures to the server which has been transplanted to the development board. If the house is intruded, users can choose the system to inform them by phone or SMS, and then users can download the pictures through the mobile phone to know more clearly the situation of house. However, in the case of light changes drastically, this system is prone to making errors in detection. When the system has just started to run, the background modeling algorithm cannot handle ghost regions quickly. Moreover, this system is not combined with other sensor technologies to realize an integrated security system.

Architecture

The structure design of system

The structure of hardware design