Vision based Patient Fall Detection using Deep Learning in Smart Hospitals
Komal Singh1, Akshay Rajput2, Sachin Sharma3

1Komal Singh*, Department of Computer Science and Engineering, Graphic Era Deemed to be University, Dehradun, Uttarakhand, India.
2Akshay Rajput, Department of Computer Science and Engineering, Graphic Era Deemed to be University, Dehradun, Uttarakhand, India.
2Sachin Sharma, Department of Computer Science and Engineering, Graphic Era Deemed to be University, Dehradun, Uttarakhand, India.

Manuscript received on November 15, 2019. | Revised Manuscript received on 20 November, 2019. | Manuscript published on December 10, 2019. | PP: 4826-4832 | Volume-9 Issue-2, December 2019. | Retrieval Number: B7688129219/2019©BEIESP | DOI: 10.35940/ijitee.B7688.129219
Open Access | Ethics and Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)

Abstract: With the emergence of new concepts like smart hospitals, video surveillance cameras should be introduced in each room of the hospital for the purpose of safety and security. These surveillance cameras can also be used to provide assistance to patients and hospital staff. In particular, a real-time fall of a patient can be detected with the help of these cameras and accordingly, assistance can be provided to them. Different models have already been developed by researchers to detect a human fall using a camera. This paper proposes a vision based deep learning model to detect a human fall. Along with this model, two mathematical based models have also been proposed which uses pre-trained YOLO FCNN and Faster R-CNN architecture to detect the human fall. At the end of this paper, a comparison study has been done on these models to specify which method provides the most accurate results. 
Keywords: Deep learning, FCNN, R-CNN, Smart Hospitals
Scope of the Article: Deep learning