International Journal of Innovative Research in                 Electrical, Electronics, Instrumentation and Control Engineering

A monthly peer-reviewed journal

ISSN Online 2321-2004
ISSN Print 2321-5526

Since  2013

Abstract: The fall detection system has grown in importance within the homecare system in recent years. Among the elderly, accidental falls are a common source of damages, fatalities, and loss of control. Accident falls also significantly affect the costs of the national health system. Therefore, there is a need for in-depth research and the creation of fall detection technology to save elderly people. This article offers a thorough analysis of contemporary fall detection methods taking into account the most potent deep learning technique yolo algorithm which uses convolutional neural network (CNN) layers in recognizing persons and detects fall based on height and width dimension analysis of a person. This fall detection technique uses neural networks, open source human detection dataset, yolo v3 pre-trained model which make more reliable and accurate than the conventional fall detection methodology.

Keywords: Convolutional neural networks, bounding box, yolo, Gaussian blur.

PDF | DOI: 10.17148/IJIREEICE.2022.10739

Open chat