You are on page 1of 8

1

Automated Proper Personal Protective Equipment Inspection for Kitchen Staff using
Object Recognition with QR Code Clocking System

Loryvi Grace Leogo


Kyle Xavier A. Mendoza
Jhonella Jazmine R. Orquillo
Arishavelle Karl D. Villanueva

A Thesis
Presented to the Faculty of the
Computer Studies Department
College of Science
Technological University of the Philippines
Ayala Boulevard, Manila

In Partial Fulfillment of the


Requirements for the Degree
Bachelor of Science in Computer Science

October 2022
2

TABLE OF CONTENTS
Page
Title 1
Table of Contents 2
Chapter 1 THE PROBLEM AND ITS SETTINGS
Background of the study 3
Objectives of the Study 4
Scope and Limitations of the Study 6
Significance of the Study 7
REFERENCES 8
3

Chapter 1

THE PROBLEM AND ITS SETTINGS

Background of the study

Object recognition software grew in popularity as the internet developed and large

collections of images became accessible online for analysis. These expanding data sets have

made it possible for computers to recognize specific individuals in images and videos. Computer

vision is currently experiencing a revolution as a result of a number of factors involved:

computing power has become more accessible and affordable. The availability of hardware for

computer vision and analysis has increased. Convolutional neural networks are one such

algorithm that can benefit from the hardware and software capabilities. Additionally, Computer

vision is also relevant in digitalization of recognizing, interpreting and understanding. A person

has the advantage of lifetimes of context to train how to discern objects apart, how far away they

are, whether they are moving, and whether there is something wrong in an image, making human

vision superior to computer vision. Instead of using retinas, optic nerves, and a visual brain,

computer vision teaches robots to execute these tasks in considerably less time using cameras,

data, and algorithms. A system trained to inspect items or monitor a production asset can swiftly

outperform humans since it can examine thousands of products or processes per minute while

spotting imperceptible flaws or problems.

The purpose of object recognition, often referred to as object detection, is to identify and

locate the items in an image. Computer vision tasks like object recognition and image
4

recognition are different but related. The objective of image recognition is to assign a single label

to the entire image in order to categorize it. An object recognition model would, however, return

a bounding box and label. Face identification and recognition through images and videos is

referred to as face recognition. Additionally, object detection is taken into account in a number of

computer vision applications, including object retrieval, tracking, video surveillance, captioning,

image segmentation, and medical imaging (Ashish Patel, 2020). In general, the use of face

recognition can be compared to the use of both a face scanner and an object scanner in the same

system. It is advantageous to society, lessens the pressure on the food sector, and promotes

outstanding food service.

Automated Proper Personal Protective Equipment Inspection System aims to simplify

equipment inspection through the use of object identification, which could also act as a

system-wide attendance recorder. Additionally, the system's goal is not to replace human labor in

the food sector, but rather to offer a useful service that enables workers to execute their tasks

efficiently, prevent risks, and save time by avoiding manual inspection and processing

attendance.

Objectives of the Study

General Objective

The goal of this project is to create an application, a monitoring system, for

personal protective equipment that verifies the kitchen crew is using facemasks, hairnets,

and gloves appropriately.

Specific objective

The study's specific objectives are as follows:


5

1. Design and develop a web-based monitoring system with the following components:

a. Create an efficient detection system with the following features:

● Object recognition of Personal Protective Equipment such as

facemasks, hairnets / hair mask, and gloves.

● Level of Proper Usage of Personal Protective Equipment:

o Safe, NotSafe, NoMask, NoHat

● Alert System which include:

o Notification and record system of employees proper usage of

PPE (Personal Protective Equipment)

● Clocking System which include:

o Scanning of QR code designed for each employees for login

and logout

b. To assist the employees of the food industry in strictly complying with food

safety and avoid contamination during the processing of food products.

c. Develop the Web-based application using the following tools and

technologies:

● OpenCv for object and face recognition

● Yolov5 for Training Data and Creating a Model

● Python as the main programming language for building this program

● Django as the main framework of python programming language for

building a web application

2. Test and Implement the monitoring system

3. Evaluate and assess the acceptability component of the system using the assessment
6

criteria of the Monitoring and alert system on functionality, efficiency, usability,

compatibility, maintainability, reliability, and conveniency of the system.

Scope and Limitations of the Study

The Automated Proper PPE Inspection System will be designed, developed, and

implemented to ease the work and push to modernize the traditional inspection of safety kitchen

staff equipment.

The current system includes the design of the application that could detect proper use of

PPEs. The application contains equipment inspection, and a clocking system using QR code. The

system provides a label which tells if a person being checked is wearing the equipment properly

or not. The application can evaluate three monitored pieces of equipment that kitchen personnel

are required to wear inside the area at all times. The three equipment pieces that will be

recognized by the system are the facemask, hairnet / hair mask, and gloves. The system can only

process one personnel at a time. The alert function notifies the user when the scanner detects

whether the equipment is properly used or not. The current system does not have the function to

send the captured screen for verifying the penalties of the staff. Aligned with this, the application

can be operated with or without the clocking system.

The You Only Look Once (YOLO) algorithm will be used by the application.

Convolutional neural networks (CNN) are used by the YOLO method to recognize items

instantly. The approach just needs one forward propagation through a neural network to detect

objects, as the name would imply. This indicates that a single algorithm run is used to perform

prediction throughout the full image. Multiple class probabilities and bounding boxes are

simultaneously predicted using the CNN.


7

As every object has a different structure when improperly worn, the user of the

application can manually select which specific equipment to be inspected for more precise

detection. The application enables the user to instantly verify whether the PPE is being worn

properly. The application also has alert notifications that label the equipment being checked and

notify the user (proper, not proper, no equipment).

Ten purposively selected respondents would evaluate the system component consisting of

three information technology professionals, five kitchen staff, and two safety professionals.

The ISO 25010 Software Quality Model is the assessment tool that would be used to

gauge how acceptable the object recognition app is.

Significance of the Study

The traditional way of inspection and taking attendance has been difficult and time

consuming. It has been an inconvenience for the company if this will not do the task urgently.

The rising and emerging technology forced us to modernize and digitalize every system

including the improvement of Food Industry services.

The Automated Proper Personal Protective Equipment (PPE) Inspection System will

benefit the public and make local government officials' jobs easier. encouraging production and

efficiency, particularly in the food industry The system encourages work efficiency by increasing

openness of Project Developments, which will save time and manual inspection at the

workstation. For the client, feeling satisfied and receiving consistent food quality. For developers

and researchers, this system could serve as a guide and reference for future development of the

Automated Proper PPE Inspection.


8

REFERENCES

Ashish Patel. (2020, June 11). What is Object Detection ? (medium.com).


https://medium.com/ml-research-lab/what-is-object-detection-51f9d872ece7

You might also like