Social distancing detection by using deep neural network
Natural Volatiles & Essential Oils
; 8(5):77-82, 2021.
Article
in English
| CAB Abstracts | ID: covidwho-1812253
ABSTRACT
This article helps in presenting a clear view of detecting social distance in order to evaluate the distance covered by the people from each other. It helps in providing alarming notification to the people in making safety during this pandemic period by the use of video feed. The frame of the video clipping is used as an input and is implemented based on the hybrid computer vision. The Deep neural network dependent algorithm named YOLOv3 has been used in alarming the detection of distance between the people. It is used along with the mapping technique known as Inverse Perspective Mapping (IPM) which is one of the tracking mechanisms that helps in monitoring the social distance. It is tested against MS COCO and image data sets that has been obtained from Google. The precision was found to be 97% that helps to design the outlook of the place where the public involvement is too high. It could help in controlling the violations of trespassers who does not obey the rules of social distancing and also servers to be precaution to control the disease prone zones.
Information and Documentation [CC300]; Policy and Planning [EE120]; Human Health and Biology (General) [VV000]; Techniques and Methodology [ZZ900]; detection; social distancing; disease control; health protection; health programmes; human health; computer techniques; machine vision; algorithms; optical tracking; imagery; public health; health policy; safety; man; Homo; Hominidae; primates; mammals; vertebrates; Chordata; animals; eukaryotes; health programs; computer applications; computer vision
Search on Google
Collection:
Databases of international organizations
Database:
CAB Abstracts
Language:
English
Journal:
Natural Volatiles & Essential Oils
Year:
2021
Document Type:
Article
Similar
MEDLINE
...
LILACS
LIS