Activity recognition using C++ OpenCV

• Main Author: Ng, Alex Yong Ming
• Other Authors: Chua Chin Seng
• Format: Final Year Project
• Language: English
• Published: 2018
• Subjects: DRNTU::Engineering
• Online Access: http://hdl.handle.net/10356/75480
• Institution: Nanyang Technological University

With the rising threat around the globe, surveillance is a key factor. Human Activity Recognition techniques integrated in surveillance camera can be employ to recognize human activity in present world context. Activity recognition technology in surveillance cameras will definitely aid in the detection of violent actions and suspicious behaviours. The aim of activity recognition is to recognize and classify human activities using cameras and computer vision algorithms. Presently, to detect complex action and activity can be highly challenging. One of the challenges is day and night detection. As day and night detection might need to use different algorithm and techniques to achieve detection. Another reason is, complicated actions such as a normal kick for sports and a real violent kick makes it hard to differentiate for the system. Nevertheless, the aim of this project is to develop a system using Microsoft Visual Studio C++ with external library such as OPENCV to create an activity recognition application. OPENCV is an open source library created by the computer vision community to aid in the function of image processing and computer vision algorithms. The techniques used to construct this application will be, Motion History Image and Template Matching Function.