Authors :
Asit Kumar Das; Vaanishree Kamthane; Umang Purwar; Deba Chandan Mohanty; Bharani Kumar Depuru
Volume/Issue :
Volume 9 - 2024, Issue 3 - March
Google Scholar :
https://tinyurl.com/27jkb4zs
Scribd :
https://tinyurl.com/mr49smb4
DOI :
https://doi.org/10.38124/ijisrt/IJISRT24MAR2142
Note : A published paper may take 4-5 working days from the publication date to appear in PlumX Metrics, Semantic Scholar, and ResearchGate.
Abstract :
This project aims to develop an innovative
system for accurately tracking employees' working hours
based on their presence within designated workspace
areas, particularly their work cabins. Leveraging
advanced technologies such as image annotation,
preprocessing, and augmentation, as well as robust object
detection models, this study addresses the need for
efficient employee surveillance and time management
solutions in contemporary workplaces.
The methodology involved detailed annotation and
enhancement of image data, enabling precise
representation of cabin areas and identification of
individual employees within images. Subsequently, a
state-of-the-art object detection model, YOLOv8, was
utilized to train using this annotated dataset, achieving an
impressive accuracy of more than 90% in recognizing and
tracking employee presence within the specified cabin
regions.
Through small incremental changes based on
previous insights and optimization, the project achieved
high levels of accuracy in inferring employees' working
hours based on their occupancy within the designated
workspace. By differentiating between time spent inside
cabins (considered as working time) and time spent
outside these areas (considered as non-working time), the
system offers an automated and objective approach to
time tracking, eliminating the need for manual input or
subjective assessments.
Future scopes for this research include exploring the
integration of additional sensors or data sources to
further enhance the accuracy and granularity of
employee activity tracking. Additionally, advancements
in machine learning algorithms and hardware may enable
real-time processing and analysis of surveillance data,
leading to more proactive management of employee
productivity and well-being. Moreover, the application of
this technology could extend beyond traditional office
settings to various industries, such as manufacturing,
retail, and healthcare, where precise monitoring of
employee activities is crucial for optimizing operations
and ensuring compliance with regulations.
This research underscores the potential of advanced
computer vision techniques, particularly YOLOv8, in
revolutionizing employee surveillance and time
management practices. By providing real-time
monitoring capabilities and ensuring compliance with
work regulations, this approach holds promise for
enhancing workplace productivity, transparency, and
accountability.
Keywords :
Human Resource Management, Employee Surveillance, Workplace Security, Real-Time Monitoring, Object Detection, Artificial Intelligence, YOLO.
This project aims to develop an innovative
system for accurately tracking employees' working hours
based on their presence within designated workspace
areas, particularly their work cabins. Leveraging
advanced technologies such as image annotation,
preprocessing, and augmentation, as well as robust object
detection models, this study addresses the need for
efficient employee surveillance and time management
solutions in contemporary workplaces.
The methodology involved detailed annotation and
enhancement of image data, enabling precise
representation of cabin areas and identification of
individual employees within images. Subsequently, a
state-of-the-art object detection model, YOLOv8, was
utilized to train using this annotated dataset, achieving an
impressive accuracy of more than 90% in recognizing and
tracking employee presence within the specified cabin
regions.
Through small incremental changes based on
previous insights and optimization, the project achieved
high levels of accuracy in inferring employees' working
hours based on their occupancy within the designated
workspace. By differentiating between time spent inside
cabins (considered as working time) and time spent
outside these areas (considered as non-working time), the
system offers an automated and objective approach to
time tracking, eliminating the need for manual input or
subjective assessments.
Future scopes for this research include exploring the
integration of additional sensors or data sources to
further enhance the accuracy and granularity of
employee activity tracking. Additionally, advancements
in machine learning algorithms and hardware may enable
real-time processing and analysis of surveillance data,
leading to more proactive management of employee
productivity and well-being. Moreover, the application of
this technology could extend beyond traditional office
settings to various industries, such as manufacturing,
retail, and healthcare, where precise monitoring of
employee activities is crucial for optimizing operations
and ensuring compliance with regulations.
This research underscores the potential of advanced
computer vision techniques, particularly YOLOv8, in
revolutionizing employee surveillance and time
management practices. By providing real-time
monitoring capabilities and ensuring compliance with
work regulations, this approach holds promise for
enhancing workplace productivity, transparency, and
accountability.
Keywords :
Human Resource Management, Employee Surveillance, Workplace Security, Real-Time Monitoring, Object Detection, Artificial Intelligence, YOLO.