LOST AND FOUND MANAGEMENT

SYSTEM
21CSC205P/DATABASE MANAGEMENT
SYSTEMSPROJECT REPORT

Submitted by

METHUNRAJ A RA2311003050180

HARI RAMAN R RA2311003050215

THARANEESVAR A RA2311003050238
Under the guidance of

Dr. RAHMATH NISHA

(Assistant Professor, School of Computing)

in partial fulfilment for the award of the degree

of

BACHELOR OF TECHNOLOGY
in

COMPUTER SCIENCE AND ENGINEERING

Of

FACULTY OF ENGINEERING AND TECHNOLOGY

SRM INSTITUTE OF SCIENCE AND

TECHNOLOGY TIRUCHIRAPPALLI

1
 SRMINSTITUTEOFSCIENCEANDTECHNOLOGY
(Deemed to be University U/S3 of UGC Act,1956)

BONAFIDECERTIFICATE
Certified that this project report titled LOST AND FOUND MANAGEMENT
SYSTEM is the Bonafide work of METHUNRAJ A (RA2311003050180),
HARI RAMAN R (RA2311003050215), THARANEESVAR A
(RA2311003050238) who carried out the project work under my supervision.
Certified further, that to the best of my knowledge the work reported herein does
not form any other project report or dissertation on the basis of which a degree or
award was conferred on an occasion on this or any other candidate.

SIGNATURE SIGNATURE

Dr. M.Tech.,Ph.D., Dr.KANAGA SUBA RAJA.S, M.E.,Ph.D.,

Assistant Professor
Head of the Department
School of Computing,
Department of COMPUTER SCIENCE AND
SRM Institute of Science and Technology,
ENGINEERING,
Tiruchirappalli.
SRM Institute of Science and Technology,
Tiruchirappalli.

Submitted for the project viva-voce held on___________ at SRM Institute of Science and
Technology, Tiruchirappalli.

2
 SRM INSTITUTE OF SCIENCE AND
TECHNOLOGY

TIRUCHIRAPPALLI

DECLARATION

We hereby declare that the entire work contained in this project report titled
LOST AND FOUND MANAGEMENT SYSTEM has been carried out by
METHUNRAJ A(RA2311003050180),HARIRAMAN
R(RA231100305021,THARANEESVAR A(RA2311003050238) at SRM
Institute of Science and Technology, Tiruchirappalli, under the guidance of
Dr.RAHMATH NISHA, Assistant Professor, School of Computing.

Place: Tiruchirappalli XXXXX

Date:XXXXX
XXXXX

3
 ABSTRACT

The Lost and Found Management System is a full-stack web application

designed to streamline the reporting and recovery of lost and found items. Built

using React (TypeScript, TailwindCSS) for the frontend and Flask for the

backend, it enables users to register, report items, upload images, and search

records efficiently. A Siamese Neural Network is integrated for AI-based

image matching, improving accuracy beyond traditional text search.

The backend uses MySQL for data storage and provides secure RESTful APIs

for authentication, data handling, and image serving. File uploads are managed

through scalable custom endpoints. The modular architecture supports future

enhancements like notifications, chat, or cloud-based AI, making it suitable for

deployment in schools, public areas, or offices.

4
 TABLE OF CONTENTS
Page .no
ABSTARCT 4
LIST OF FIGURES 6
LIST OF ACRONYMS AND ABBREVIATIONS 7
1 Introduction 8

1.1 Introduction 8

1.2Problem Statement 8

1.3 Objectives 9

1.4 Scope and Motivation 9

2 Existing System 10

3 Design (ER Diagram) and Proposed Methodology 11

4 Implementation 16

5 Result and Discussion 17

6 Conclusion 19

7 References 20

5
 LIST OF FIGURES

3.1 Block Diagram 11

3.2 Flowchart 14
3.3 ER Diagram 15
5.1 Result 18
5.2 Output when the user logs in 18
5.3 Output when the user asks access 19
5.3 Output when the admin gives access 19

6
 LIST OF ACRONYMS AND ABBREVIATIONS

Acronym Abbreviation Full Form

DB Database

ER Diagram Entity-Relationship Diagram

RBAC Role-Based Access Control

MAC Mandatory Access Control

DAC Discretionary Access Control

Amazon Web Services Identity

AWS IAM
and Access Management

National Institute of Standards

NIST
and Technology

Institute of Electrical and

IEEE
Electronics Engineers

International Business
IBM Machines Corporation

7
 CHAPTER 1
INTRODUCTION
1.1INTRODUCTION
In today’s digital world, access control systems play a crucial role in ensuring
that resources such as data, files, or hardware are used appropriately by
authorized users. As the number of users increases in a system, so does the
complexity of managing access requests fairly and efficiently. Traditional access
control models often rely on sophisticated backend architectures involving
databases, authentication servers, and real-time conflict resolution mechanisms.
However, for educational purposes and lightweight applications, building a full-
scale backend system might not always be practical. This project proposes a
simplified, front-end based Conflict-Driven Access Control System built
entirely with HTML, CSS, and JavaScript, focusing on the core logic of conflict
management without the overhead of server-side complexities.
The proposed system introduces the concept of conflict resolution using an
algorithm inspired by Dijkstra’s shortest path principle — selecting the optimal
user for resource allocation based on priority factors like request time or role
weight. Instead of resolving conflicts randomly or simply giving priority to the
earliest request, the system intelligently computes the most appropriate user to
grant access. Users can register, log in, view available resources, and request
access. An admin panel allows administrators to monitor all conflicts and
manually intervene if needed. The project provides a foundation for
understanding access control strategies combined with algorithmic decision-
making at the front-end level.

1.2 PROBLEM STATEMENT:

In high-traffic public spaces such as schools, offices, and transportation hubs,

managing lost and found items is often inefficient and unreliable. Traditional
systems rely on manual processes or basic text logs, which are prone to human
error, slow matching, and low item recovery rates. These systems lack
intelligent mechanisms to identify similarities between lost and found items,
especially when users provide vague or incomplete descriptions.
There is a need for a lightweight, intelligent Lost and Found Management
System that automates item reporting, searching, and matching. By integrating

8
AI-powered image matching model (Siamese Network), the system can
compare visual features of reported items and suggest accurate matches—even
when textual data is insufficient. This approach reduces manual effort, increases
recovery success, and provides a more efficient and user-friendly experience for
both item reporters and seekers.

1.3 Objectives:
The main objective of this project is to design and implement a web-based Lost
and Found Management System that streamlines the process of reporting,
searching, and recovering lost or found items using modern full-stack
technologies. The system aims to provide a responsive frontend built with React
(TypeScript, TailwindCSS) and a secure Flask-based backend, enabling users to
register, log in, report items, and perform intelligent searches.
A core feature of the system is the integration of an AI-powered image
matching algorithm based on a Siamese Neural Network, which enhances the
accuracy of matching items beyond traditional text-based search. Instead of
relying solely on keywords, the system uses visual similarity to identify
potential matches, helping users who cannot describe their items precisely.
Additionally, the platform includes efficient file upload handling,
structured MySQL database management, and modular APIs to support
scalability. The system also lays the groundwork for future extensions such as
real-time notifications, chat support, and cloud-based AI services. By combining
deep learning with web development, the project demonstrates a practical and
intelligent solution to common problems in lost and found operations.

1.4 SCOPE AND MOTIVATION:

This project aims to build a web-based Lost and Found Management
System using React for the frontend, Flask for the backend, and MySQL for
data storage. The scope includes user registration, item reporting, image
uploads, intelligent search, and AI-based image matching using a Siamese
Neural Network. The system also handles secure file management and scalable
API communication.
The motivation is to modernize the outdated manual lost and found
process by integrating AI and full-stack web development. It improves item
recovery accuracy and user experience while serving as a valuable learning tool

9
for students and developers exploring real-world applications of AI, databases,
and responsive design.

CHAPTER 2
In most traditional environments, Lost and Found management is still
handled through manual or semi-digital methods, such as physical logbooks,
spreadsheets, or basic reporting tools. When someone loses or finds an item,
they must approach a designated office or staff member, fill out a form, and wait
for follow-up actions. This process is time-consuming, inconsistent, and often
leads to low item recovery rates.
There is typically no centralized database, and search functions are
rudimentary or completely absent. Users rely heavily on memory, vague textual
descriptions, and manual browsing through item logs. This significantly limits
the system's effectiveness—especially when there are many similar-looking
items or users cannot provide accurate details.
Another critical limitation is the lack of intelligent matching mechanisms.
Traditional systems do not use image recognition or AI to assist in identifying or
comparing lost and found items. As a result, matching is often based solely on
text, which can be ambiguous and subjective.
Additionally, file handling and image uploads are either unsupported or
poorly managed. Users may not be able to submit images easily, and even when
they do, there is no automated way to compare those images against existing
records.
Moreover, security and scalability are major concerns. Without proper
user authentication or structured backend support, the systems are vulnerable to
errors, misuse, and data loss. Manual entries are prone to mistakes, and tracking
item status becomes difficult as volume increases.

10
 CHAPTER 3
ER DIAGRAM AND PROPOSED METHODOLOGY
3.1 BLOCK DIAGRAM

Fig.3.1 block diagram

11
1. Problem Definition
 Identify the issue: Traditional lost and found processes are
manual, slow, and inaccurate.
 Define system goals: Enable users to report, search, and retrieve
lost/found items efficiently.
 Integrate AI-based image matching to improve item
identification accuracy.

2. Environment Modeling
 Model the system as a web-based platform with:
 Users (reporters, searchers)
 Items (lost/found reports)
 Database (storage for users, items, images, statuses)
 Define data flow:
 User → Form Submission → Backend → Database → AI
Matching → Output

3. User Interaction and Data Input

 Users register/login using secure authentication.
 Users report a lost/found item with:
 Title, description, category
 Location and time
 Uploaded image

12
4. Backend Processing
 Flask API receives and validates requests.
 Handles:
 Data routing
 File (image) storage
 Status tracking
 Saves item data into the MySQL database.

5. AI-Based Image Matching

 Uploaded image is passed to a Siamese Network / CNN model.
 Model compares new images with existing reports in the
database.
 Calculates similarity score based on visual features.

6. Matching Decision
 If similarity threshold is met:
 Notify the user about a potential match.
 Display matched items for user confirmation.
 If not:
 Store the item for future matching.

7. Search and Retrieval

 Users can manually search:
 By item name, description, category, or location.
 System provides filtered and sorted results.
13
  Image preview helps with visual identification.

8. File Handling
 Image uploads are handled through a custom endpoint
(/uploads/<filename>).
 Images are stored and served efficiently.
 Scalable design for handling many files without performance
loss.

9. Conflict Handling (Duplicates / False Matches)

 Track duplicate reports using:
 Time stamps, image hashes, and location proximity.
 Allow users to confirm/reject automated matches.
 System learns from user responses to improve accuracy over
time.

10. Feedback & Future Optimization

 Collect user feedback after item recovery.
 Analyze matching success rate and user satisfaction.
 Optimize:
 AI model with more training data
 Add real-time notifications
 Integrate chat support or smart filters
 Consider cloud deployment for scalability
3.2 FLOWCHART

14
Fig 3.2 Flowchart

15
3.3 ENTITY RELATIONSHIP DIAGRAM

Fig 3.3 ER Diagram

16
 CHAPTER 4
IMPLEMENTATION
4.1 IMPLEMENTATION

17
 CHAPTER 5
RESULTS AND DISCUSSION
5.1 RESULT:
Output of the Program:

Fig 5.1 Result

Output when user logs in:

Fig 5.2 Output when the user logs in

18
Output when the user asks access:

Fig 5.3 Output when the user asks access

Output when the admin gives access:

Fig 5.3 Output when the admin gives access

19
 CHAPTER 6
CONCLUSION
In conclusion, the Lost and Found Management System addresses the
inefficiencies and limitations of traditional, manual lost item tracking
methods. Conventional approaches often suffer from low item recovery
rates, lack of automation, and poor user experience due to their reliance
on handwritten records or static databases. Our proposed system
revolutionizes this process by integrating AI-powered image matching, a
modern web interface, and a secure backend to streamline reporting,
matching, and retrieving lost and found items.
By automating the item matching process using a Siamese Neural
Network, the system greatly improves accuracy and reduces the
dependency on textual descriptions. The combination of a responsive
frontend (developed using React) and a Flask-based backend with a
MySQL database ensures data is processed, stored, and retrieved
efficiently. File uploads and image management are handled through
scalable API endpoints, supporting system growth over time.
Furthermore, the intuitive user interface simplifies reporting and
searching for users, while future-ready features such as notifications and
real-time filtering improve engagement and responsiveness. The
architecture is designed with extensibility in mind, allowing future
integration with cloud-based AI services or advanced user communication
tools.
Ultimately, this Lost and Found Management System delivers a
comprehensive, intelligent, and user-centric solution. It not only enhances
item recovery rates but also offers a scalable foundation for deployment
in environments such as schools, transportation hubs, workplaces, or
public venues. Through this innovation, we aim to create a more reliable,
efficient, and accessible system that benefits both users and
administrators.

20
 CHAPTER 7
REFERENCES
1. Bromley, J., Guyon, I., LeCun, Y., Säckinger, E., & Shah, R. (1993).
Signature Verification using a "Siamese" Time Delay Neural Network.
Advances in Neural Information Processing Systems (NeurIPS).
https://papers.nips.cc/paper_files/paper/1993/hash/afbfc3c3280072984e8
00db3cfdef498-Abstract.html
2. Krizhevsky, A., Sutskever, I., & Hinton, G. E. (2012).
ImageNet Classification with Deep Convolutional Neural Networks.
Advances in Neural Information Processing Systems.
https://papers.nips.cc/paper_files/paper/2012/file/c399862d3b9d6b76c84
36e924a68c45b-Paper.pdf
3. Zhang, D., Islam, M. M., & Lu, G. (2011).
A Review on Automatic Image Annotation Techniques.
Pattern Recognition Journal.
https://www.sciencedirect.com/science/article/abs/pii/S003132031100215
6
4. React Documentation — Building User Interfaces with ReactJS.
https://reactjs.org/
5. Flask Documentation — Flask Web Framework for Python.
https://flask.palletsprojects.com/
6. MySQL Documentation — Database Design and Management.
https://dev.mysql.com/doc/
7. Google Cloud AI Services — Vision API for Image Analysis.
https://cloud.google.com/vision

21