Motion detection using Nvidia Jetson platfom

• Main Author: Azri Ishak
• Other Authors: Chau Lap Pui
• Format: Final Year Project
• Language: English
• Published: 2018
• Subjects: DRNTU::Engineering
• Online Access: http://hdl.handle.net/10356/75210
• Institution: Nanyang Technological University

Motion Detection is a popular subject of research in the field of image processing. There have been many versions of this process produced to cater for the various applications that require the analysis of motion. The most common application that requires motion detection is security surveillance. A PIR sensor is commonly found in intruder alarm systems. Other applications include automation and human-machine interaction. Examples of these are the video consoles that make use of real-time video feedback to enable a function in the game. This can be found in systems like the Nintendo Wii and Microsoft Kinect. There are many techniques and methods created for this purpose of segmenting moving objects. Traditional approaches consist of using histograms analysis, gaussian models, estimation models and support vector models to create the background frame. This project will utilize a computer vision library to build a system that will allow the user to perform background subtraction in real-time. The system is built on an Nvidia Jetson TX2, which is a mini low-powered embedded device that is capable of offering real time responsiveness and low latency.