REAL TIME OBJECT DETECTION: A PATH TOWARDS VIDEO SURVEILLANCE AS A SERVICE

Video surveillance is the largest segment in the security system. Object detection from the real time video under surveillance is challenging in terms of space and time to detect the object. The cloud storage service helps to fulfil the storage requirement of video under surveillance. This research paper aims to develop a solution for Real Time Object Detection: A Path Towards Video Surveillance as a Service is proposed to provide optimal solution for detecting objects in real time. The design approach of RTOD-VSaaS used for intruder detection that reinforces surveillance technology to provide essential security to the life and associated control and alert operations. This work contains various modules such as Admin, Video Capture, Preprocessing, Email Generating Module and Report Generating Module. Camera sensor captures video under surveillance with a speed of 20 frames per second. The Input video is streamed and stored in Cloud server using twitch and drop box respectively. Preprocessing techniques such as Frame generation and Sobel Edge Detection is considered in the experimentation. After the detection of the object using Kalman Filter an alert message is sent to the authorized person through mail. The cloud video streaming reveals, cloud-based video surveillance as a service.