Elegant and Efficient Algorithms for Real Time Implementation of Object Detection, Classification, Tracking and Counting using FPGA Zynq XC7Z020 for Automated Video Surveillance Ap
share
Video Surveillance is very important and essential task for public places and sensitive areas to monitor safety, security issues and prevent crime. Especially in the field of surveillance system it has gained greater significance than ever before due to the recent terror activities taking place all over the world. Computer vision has played a key role in developing object detection and tracking techniques for Surveillance system. Detection of objects precisely is vital for many applications such as person identification, abnormal activity detection, congestion analysis, military fields. Traditional surveillance was like capturing video from camera, storing the information in a database, required contents were accessed manually from the database and it requires human operator to constantly monitor suspicious or threatening activities. It may lead to loss of sensitive information in real time. In such cases automated video surveillance is very essential. Current or automated surveillance system uses a digital technology to capture, store and process an image or video. Smart and intelligent surveillance is required to minimize the role of human operator and automatically track, monitor the suspicious or threatening activities. It finds the required information, also informs to the administrator in real time. Most of the implementations currently employed are based on serial execution on general purpose processors. But the high cost and complexity of such implementations doesnt make it a viable option for real time surveillance system. Many efforts have been made to make the system automated in order to decrease the complexity and to increase the ease with which it can be implemented. The system proposed here is implemented on Field Programmable Gate Arrays (FPGA) Zynq XC7Z020 board using Modified Background Subtraction and adaptive background subtraction algorithm for real-time object detection, counting, tracking and classification. The presence of numerous configurable logic blocks, distributed memory and hard Digital Signal Processing (DSP) modules offers a great flexibility in achieving Temporal and Spatial parallelism. It uses Xilinx ISE software for implementation which is programmed in VHDL. The system works in real time with minimum time lag between the capture and display. Moreover the entire system is optimized in terms of speed; memory requirements as well as the number of logic elements used which makes it suitable for application in real-time surveillance system. Algorithms are successfully implemented based on different applications under different conditions.
READ FULL TEXT