Publication: Human Fall Detection with Computer Vision and Deep Learning
Loading...
Date
2020-09
Authors
Muhammad Abid bin Amer
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
This thesis reports on the fall detector using computer vision with deep learning. The main objective of this project is to help reduce the number of a fatal accident due to a fall in an elderly house and hospital. The conventional fall detector usually is not user friendly as it requires the user to wear it all the time. Hence, the fall detector using input from Closed-Circuit Television (CCTV) footage is used to detect fall as it is user friendly and able to be ready to detect fall all the time. A deep learning approach is used in this project to detect a human to reduce false alarm in detection. Thus, making the system to be more reliable and accurate in detecting fall and eventually will reduce the number of a fall accident. Visual studio code and other IDE were used to design and test the system. The accuracy of the system then determined by using fall dataset videos
Description
Interim Semester 2020/2021
Keywords
Fall Detection , Computer Vision , Deep Learning