Industrial Attendance and Access Control System Using Face and

Biometric Recognition

S. Brindha, D. M. Bharath, S. R. Gokulprasath and M. Ravisrinivasan

Department of Electronics and Communication Engineering, Nandha Engineering College, Erode, Tamil Nadu, India

Keywords: Facial Recognition, Attendance Management, DeepFace, IoT, NodeMCU, Access Control, Industrial

Automation, Security.

Abstract: Tracking attendance and maintaining access control are of paramount importance in industrial setting today

not only for management purposes, but also for security. In this paper, an efficient and powerful Industrial

Attendance and Access Control System using facial recognition and biometric authentication is presented.

The platform includes state-of-the-art computer vision algorithms for real-time face detection and verification

in amalgamation with other biometric modality such as fingerprint recognition for correct identification of

staff. Combination of two factor in attendence management system [201520] In these methods the reliability

and diminishing spoofing of identity can be improved, but the system is still dependent on the user inputs and

also erases the user inputs which are prone to errors. Related data are saved and maintained in a centralized

server from which the monitoring and administrative tasks are carried out in real-time or are recorded and

stored for subsequent checks. The limitation for unauthorized entry helps to create a defense against general

industrial security violations in addition to resource management optimization. The system is highly accurate,

scalable and flexible, and can be used for a wide range of industrial applications where the safety and control

of the workforce is a key consideration.

1 INTRODUCTION

The necessity and trends towards secure, efficient,

and automated systems in involved industrial

environments, mandates a paradigm shift from

traditional methods of attendance management and

access control. Manual logging and RFID-based

approaches, traditional systems, are often

insufficient because they are susceptible to

inaccuracies, delays, and security breaches. The

proposed system is a revolutionary solution, The Face

Detection-Based Industrial Attendance Management

and Access Control System. A highly sophisticated

contactless attendance and access management

system that harnesses the power of advanced facial

recognition technology and connected IoT

capabilities. The Paper introduces this system, using

a combination of Python and the DeepFace library. It

provides the very accurate and a fastest process to

authenticate the identity of employees without having

the physical token and manual input to authenticate.

It also has the Internet of Things integrated as it uses

a NodeMCU microcontroller which is programmed

with MicroPython. That means there is no need to

train your bot with any data, therefore you can enjoy

a smooth flow of information with the system and

with the industrial machines, or any restricted access,

and then the bot can perform automated control once

the user is authenticated. The attendance logs are

saved in structured CSV files in the drive to ensure

that data management and reporting needs are as

simple as possible and the files are compatible with

existing analytical tools. This one system solution

always guarantees its utilization at areas, from small

industrial land up to gigantic and complex industrial

zones. Minimizing human involvement in the

procedure, along with utilizing contemporary

technology, makes the functioning of its system

more effective and assists in minimizing the risk of

security risks caused due to unauthorized access. It is

the last component needed when it comes to smarter,

safer, and more efficient industrial operations,

delivering transformative insight that can slot in

easily to existing workflows while delivering more

powerful scaling to demand.

408

Brindha, S., Bharath, D. M., Gokulprasath, S. R. and Ravisrinivasan, M.

Industrial Attendance and Access Control System Using Face and Biometric Recognition.

DOI: 10.5220/0013930600004919

Paper published under CC license (CC BY-NC-ND 4.0)

In Proceedings of the 1st Inter national Conference on Research and Development in Information, Communication, and Computing Technologies (ICRDICCT‘25 2025) - Volume 5, pages

408-412

ISBN: 978-989-758-777-1

2 RELATED WORKS

The field of biometric-based attendance control

techniques has been greatly developed in terms of

accuracy and security of their applications.

Fingerprint based solution has been used extensively

in this area. i. Access Control and Industrial

Workforce Management: Methods of Access Control

and Industrial Workforce Management Mittal et al.,

Shakil et al., illustrate the strong techniques. Thus,

these fingerprint-based systems work on the basis of

patterns of fingerprints and they need to depend

entirely on these patterns and also do not wish to have

redundancies in entries. Oloyede et al. expanded the

fingerprint biometrics that highlighted its

importance in enhancing staff attendance monitoring

with high accuracy. Due to hardware dependencies

and scalability say, issues for UNIX systems,

however, this has led to the search of alternative

approaches.

Three main systems are used to recognize faces,

and AI plays an important role in these face

recognition-based attendance systems, which are

gaining momentum because of AI and computer

vision improvements. Studies by Nadhan et al. 3 state

that Industry 4.0 seems to rely heavily on smart

technology, and face recognition enables smooth

attendance tracking. Further, Soniya et al. and Surve

et al. One of their major contributions was proposing

automated methods to use real-time facial

authentication and suggesting a non-intrusive and

user-friendly alternative. Al-Shebani et al. also

conducted a survey of embedded systems for door

access control, highlighting that face recognition can

be used in place of traditional mechanisms but Yang

et al. The problems of processing real-time videos

have been addressed in, with focus in optimizing the

algorithm in the dynamic environment.

Other research has explored combining multiple

biometric modalities for greater reliability. Hoo and

Ibrahim provided extensive literature review on

hardware necessities for biometric attendance system

in educational sector, with emphasis on multimodel

method for improved reliability. Face and fingerprint

recognition integrated systems, as discussed by

Singh et al., are more secure due to lower single-

point failures. Furthermore, Bavaskar describes face

recognition systems that highlights advancements in

the use of deep learning frameworks applied to

effectively detect faces in varying lighting conditions.

Vinod et al. have conducted an extensive analysis

of authentication and attendance systems tailored to

the needs of various sectors. Hidayat et al. for special

environments introduced a face recognition-based

surveillance framework for mining industries that

could work in harsh operating environments. Li et al.

Early studies of biometrics in attendance

management were presented and established a

technology platform for these developments in

system design and deployment.

New trends in the attendance and leave

management systems focus on a cloud-based and IoT-

enabled solution for attendance tracking in real-time.

Wahab et al. As it has been proposed with a complete

online system through face recognition while

enhancing the connectivity and remote management.

These advancements conform with the essentials of

Industry 4.0 in which the real-time data analysis and

implementation of decisions play a pivotal role.

Biometric attendance systems are evolving to meet

the challenges ofneed, scalability, and ecological fit,

paving the way for ubiquitous use across numerous

fronts.

3 PROPOSED SYSTEM

3.1 Overview

The "Face Detection-Based Industrial Attendance

Management and Access Control" system leverages

facial recognition to automate attendance and

machinery control with secure, remote operation via

Wi-Fi and API calls on the integration of Python,

DeepFace, and NodeMCU with MicroPython. Figure

1 Shows the Block diagram of the proposed system.

Figure 1: Block diagram of the proposed system.

3.2 Capturing the Employee Face

The system captures a high-quality image of the

employee’s face using a camera positioned at the

entry point or workstation. This camera functions as

the key input device for the facial recognition module,

ensuring providing and real-time collection of facial

data.

Workflow Diagram of Face Recognition-Based

Industrial Attendance System Shown in Figure 2.

Industrial Attendance and Access Control System Using Face and Biometric Recognition

409

Figure 2: Workflow diagram of face recognition-based

industrial attendance system.

3.3 Matching the Face with the

Database

The system captures the employee's facial image,

processes it, and compares it to the data that has been

stored in the employee database through advanced

facial recognition algorithms. An unambiguous

match serves to verify the employee's identity so that

access can be granted and attendance tracked

securely. In the event of a failure, access is denied and

the attempt logged for auditing purposes. Facial

feature extraction involves the usage of algorithms

such as DeepFace.

It ensures zero error-tamper record as the system

captures employee check-in and checkout

automatically and saves the correct date and time. It

acts as a guide to avoid input mistakes and improve

the efficiency of operations. Attendance data are

available in real-time, thereby allowing quick

decision- making regarding payroll, allocation of

resources and management of productivity. This

method is automatic, paperless, error-free, and more

secure.

3.3.1 Calling an API to Turn on Machines

Once attendance is recorded, it makes an API request

for the machine, which needs to be activated in

regards to the employee's work. Since the API

encompasses machine identification, parameters and

tasks configurations, this can solely be executed by

authorized individuals. Using machines for

attendance verification is a great way to add security,

streamline the process and reduce misuse. This then

improves operational continuity and productivity

with real-time automation of attendance and

equipment management.

3.3.2 API from NodeMCU and Enable

Respective Machine

A NodeMCU microcontroller processes the API

request sent and instructs the appropriate machine to

be turned on. NodeMCU has an inbuilt wi-fi which

makes communication from the system devices,

more secure. This integration ties machine

accessibility to real employee identities so that

equipment cannot be accessed by any unauthorized

person. The system certainly boosts security, reduces

human involvement, and cuts the operational cost by

automating attendance management and activating

the machine. Facial recognition which is powered by

IOT streamlines workforce management which

improves productivity and operational control.

4 RESULTS AND DISCUSSION

The above face detection based industrial attendance

management and access control system would give an

efficient and secure solution for managing attendance

of employees and access to machines in industrial

setup. This system uses face recognition technology

build upon a NodeMCU module to achieve the

wireless connectivity for the user.

4.1 System Performance

The face detection personality component uses

DeepFace (a deep learning library for Python) to

recognize individuals by comparing the captured

images. Please keep in mind that the accuracy of the

face recognition model is dependent on the input data

(which means, images used for training) and the

processing power of the system. Under controlled

conditions, the system has shown a very high

recognition accuracy.

Lighting conditions holds reliability for marking

attendance on real>time.

In this case, the NodeMCU runs MicroPython

and communicates to the central system through an

API. It makes sure that attendance data is tracked

live. If the login is successful after recognizing the

face, then NodeMCU makes a call to the associated

API to turn on the machinery thus making the access

control and attendance, more automated and secure.

ICRDICCT‘25 2025 - INTERNATIONAL CONFERENCE ON RESEARCH AND DEVELOPMENT IN INFORMATION,

COMMUNICATION, AND COMPUTING TECHNOLOGIES

410

4.2 System Efficiency

The chief benefit of this process is automatic

enrollment administration and access monitoring

without the need for human intervention, preventing

mistakes and saving time. The attendance CSV file

is stored, which also allows for easy integration with

other systems, such as payroll or HR management

tools, and also makes it easy to scale up. With the

ability of the system to manage multiple users in rapid

succession, the solution is well suited to situations of

high employee turnover.

4.3 Security and Privacy

Considerations

Also, being a biometric-based system, the face

recognition system is inherently secure because

biometric data is harder to counterfeit than ID cards

or passwords or similar systems. Nevertheless, some

privacy concerns around storage and use of

biometric data need attention. That includes making

sure that data is stored securely (preferably

encrypted) and complying with applicable data

protection law.

4.4 Scope and Challenges

This system is not without its limitations, however.

One of the main challenges of face recognition is

environmental factors such as lighting imperfections

or obstacles that may affect the precision of the

algorithm. Additionally, although the system

functions adequately during standard operating

conditions, the efficacy and precision can deteriorate

when faced with a significant database of users or

when the environment presents less-than-ideal real

time circumstances.

The NodeMCU suffers too from its limited

processing capabilities, which makes it hard for this

development board to work with tasks that involve

complex processing, or if the system needs to be scale

up in order to covers large installations.

Developments could be more effective code, more

essential equipment, or cloud- based computing.

5 CONCLUSIONS

In conclusion, the attendance management and access

control system with face detection automatically

tracks attendance and provides security measures

within industrial settings. By incorporating DeepFace

for facial recognition purposes, it receives real-time

accurate attendance, diminishes the possibility of

human error during operation, and minimizes human

intervention. Integration with NodeMCU,

additionally using MicroPython, allows remote

monitoring and control ensuring access to only the

authorized personnel. Despite its many advantages,

there are still challenges. There are issues related to

facial recognition under different conditions and the

need to protect individuals' privacy. Connectivity

issues, especially in remote control operations, might

also hamper real-time operations. Such challenges

can be overcome with better algorithms, improved

data protection, and optimization of.

Hardware. The scalability of the system allows for

the growth of operations and can be upgraded with

AI, ML, multi-factor authentication, cloud

integration, and mobile apps for ease of access.

Future development can include IoT devices, ERP

systems, and industry-specific applications to

improve efficiency. With sustainability and energy-

efficient hardware, the system will adapt to future

technological advancement and various industries.

REFERENCES

Ab Wahab, Kasmawahida, Lee Weng Yew, and Nor

Amizam Jusoh. "Online attendance system using face

recognition." Engineering, Agriculture, Science and

Technology Journal (EAST-J) 1.1 (2022).

Al-Shebani, Qasim, Prashan Premaratne, and Peter Vial.

"Embedded door access control systems based on face

recognition: A survey." 2013, 7th International

Conference on Signal Processing and Communication

Systems (ICSPCS). IEEE, 2013.

Bavaskar, Vaibhav. Face Recognition Attendance System.

Diss. Sant Gadge Baba Amravati University, Amravati,

2024.

Hidayat, Fadhil, et al. "Face Recognition-Based

Surveillance System in Mining Industry." Jurnal Sistem

Cerdas 7.2 (2024)

Hoo, Seng Chun, and Haidi Ibrahim. "Biometric ‐ Based

Attendance Tracking System for Education Sectors: A

Literature Survey on Hardware Requirements." Journal

of Sensors 2019.1 (2019).

Li, Tsai-Cheng, Huan-Wen Wu, and Tiz-Shiang Wu. "The

study of biometrics technology applied in attendance

management system." 2012 Third International

Conference on Digital Manufacturing & Automation.

IEEE, 2012.

Mittal, Yash, et al. "Fingerprint biometric based access

control and classroom attendance management

system." 2015 Annual IEEE India Conference

(INDICON). IEEE, 2015.

Industrial Attendance and Access Control System Using Face and Biometric Recognition

411

Nadhan, Archana S., et al. "Smart attendance monitoring

technology for industry 4.0." Journal of Nanomaterials

2022.1 (2022).

Oloyede Muhtahir, O., O. Adedoyin Adeyinka, and S.

Adewole Kayode. "Fingerprint biometric

authentication for enhancing staff attendance system."

system 5.3 (2013).

Shakil, Md, and Rabindra Nath Nandi. "Attendance

management system for industrial worker using finger

print scanner." Global Journal of Computer Science and

Technology 13.6 (2013).

Singh, Sahajpreet, et al. "Efficient face identification and

authentication tool for biometric attendance system."

2021 8th International Conference on Signal Processing

and Integrated Networks (SPIN). IEEE, 2021.

Soniya, V., et al. "Attendance automation using face

recognition biometric authentication." 2017

International conference on power and embedded drive

control (ICPEDC). IEEE, 2017.

Surve, Mayur, et al. "Automatic attendance system using

face recognition technique." International Journal of

Recent Technology and Engineering 9.1 (2020)

Vinod, Vibin Mammen, et al. "A comprehensive study on

academic and industry authentication and attendance

systems." International Journal of Scientific &

Technology Research 9.3 (2020).

Yang, Hao, and Xiaofeng Han. "Retracted: Face

recognition attendance system based on real-time video

processing." IEEE Access 8 (2020): 159143-159150.

ICRDICCT‘25 2025 - INTERNATIONAL CONFERENCE ON RESEARCH AND DEVELOPMENT IN INFORMATION,

COMMUNICATION, AND COMPUTING TECHNOLOGIES

412