Snap OCR

PROJECT REPORT

Submitted by
TISHA SHAH
2101031000251
BACHELOR OF TECHNOLOGY

in

Information Technology

College of Technology

Silver Oak College of Engineering & Technology

Silver Oak University, Ahmedabad

[April, 2025]
 Silver Oak College of Engineering & Technology
Opp. Bhagwat Vidhyapith, S.G. Highway, Ahmedabad-382481

CERTIFICATE

This is to certify that the project report submitted along with the project
entitled Snap OCR has been carried out by Tisha shah under my guidance
in partial fulfillment for the Bachelor of Technology in Information
Technology, 8th Semester of Silver Oak University, Ahmedabad during the
academic year 2024-25.

Ms. Purvi Patel

Internal Guide Head of the Department

s

Silver Oak College of Engineering & Technology

Opp. Bhagwat Vidhyapith, S.G. Highway, Ahmedabad-382481

DECLARATION

We hereby declare that the project report submitted along with the project entitled
Snap OCR submitted in partial fulfillment for the Bachelor of Technology in
Information Technology to Silver Oak University, Ahmedabad, is a Bonafide
record of original project work carried out by me at Silver Oak College of
Engineering And Technology under the supervision of Ms. Purvi Patel and that no
part of this report has been directly copied from any students’ reports or taken from
any other source, without providing due reference.

Name of the Student Sign of

Student Tisha shah
 ACKNOWLEDGEMENT

I would like to take this opportunity to express my sincere gratitude and appreciation to those who
have supported me throughout the journey of completing my internship. I am acutely aware that I did
not reach this point entirely on my own. First and foremost, I would like to thank my Internal guide,
Miss. Purvi Patel, for their invaluable guidance and feedback throughout the project's development.
Their expertise and mentorship were instrumental in steering me in the right direction and ensuring the
project's success.

I extend my heartfelt thanks to the faculty members and instructors at Silver Oak University for
providing a robust educational foundation and creating an environment conducive to independent
research and learning. Their dedication to academic excellence has been an inspiration.

Last but not the least, I would like to acknowledge the support and understanding of my friends and
family. Their encouragement, patience, and belief in my abilities motivated me to persevere through
the challenges and obstacles encountered during this internship.

Tisha shah
Information Technology
SILVER OAK UNIVERSITY
AHMEDABAD

i
 ABSTRACT

SnapOCR is an intuitive, AI-driven web application designed to streamline the process

of converting image-based content into editable digital text using Optical Character
Recognition (OCR) technology. Built with user convenience at its core, SnapOCR
enables users to upload image files, capture screenshots, or utilize sample demo files to
extract textual information in realtime.
The system integrates Tesseract OCR, an open-source OCR engine, alongside robust
image preprocessing techniques to ensure high-accuracy extraction across various
languages and font styles. To further enhance user experience, SnapOCR provides
confidence scores for each OCR result, offering transparency into the reliability of
extracted text. Users can easily copy, download, or edit the recognized content directly
from the interface, making the platform efficient and interactive.
This tool is particularly beneficial for students, educators, researchers, and
professionals who frequently work with scanned documents, handwritten notes, or
printed image materials. By eliminating the need for manual data entry, SnapOCR
significantly improves productivity and accessibility in handling digital conversions.

iii
 List of Figures

Fig 3.7.1 Gantt Chart.........................................................................11

Fig 5.1.1 Timeline Chart...................................................................18
Fig 5.1.2 Sequence Diagram.............................................................18
Fig 5.1.3 Activity Diagram..............................................................19
Fig 6.1 Snapshot 1...........................................................................21
Fig 6.2 Snapshot 2............................................................................21
Fig 6.3 Snapshot 3............................................................................22
Fig 6.4 Snapshot 4............................................................................22
Fig 8.3 Photograph with team lead…………………………………...23

iv
 List of Tables
Table 1.1 Technologies Used............................................................................................9
Table 5.1 Test Cases........................................................................................................24
Table 6.1 Evaluation Date Review 1……………..…………….………………………27
Table 6.1 Evaluation Date Review 2………………………...….……………………...27

v
 Table of Contents
Acknowledgment…………………………………………………………………………… I
Abstract……………………………………………………………………………………... II
List of Figures………………………………………………………………………………. III
List of Tables……………………………………………………………………………….. IV
Chapter 1: COMPANY OVERVIEW………………..……………………………………….. 1
1.1 …………………………………………………………………………… 2
1.2 Scope of work…………….……………………………………………………… 2
1.3 Capacity of plant…………………………………………………………………. 2
Chapter 2: OVERVIEW OF THE PROCESS BEING CARRIED OUT IN COMPANY 3

2.1 Details about the work…………………………………………………………………. 4

2.2 Technical Specification ………………………………………………………………… 4

2.3 Schematic layout………………………………………………………………………. 5

2.4 Stage of Production ……………………………………………………………………. 6

Chapter 3: INTRODUCTION OF PROJECT………………………………………….. 7
3.1 Project Summary…………………………………………………………………….. 8
3.2 Purpose………………………………………………………………………………. 8
3.3 Objective…………………………………………………………………………….. 8
3.4 Scope………………………………………………………………………………… 8
3.5 Technology and Literature Review………………………………………………..... 9
3.6 Project Planning……………………………………………………………………... 10
3.7 Project Scheduling (Gantt Chart) …………………………………………………… 11
Chapter 4: SYSTEM ANALYSIS…………………………………………………………….. 12
4.1 Study of the Current System………………………………………………………... 12
4.2 System Feasibility…………………………………………………………………... 12
4.3 Proposed System…………………………………………………………………..... 13
4.4 Features Of New System…………………………………………………………… 14
4.5 List Of Main Modules………………………………………………………………. 15
4.6 Selection of Hardware and Software……………………………………………...... 16
Chapter 5: SYSTEM DESIGN………………………………………………………………… 17
5.1 System Design & Methodology……………………………………………………... 18
5.1.1 Time-line Chart…………………………………………………………………… 18
5.1.2 Sequence Diagram………………………………………………………………... 18
 5.1.4 Activity Diagram…………………………………………………………………. 19
Chapter 6: IMPLEMENTATION…………………………………………………………….. 20
6.1 Snapshot 1……………………………………………………………………………… 21
6.2 Snapshot 2.…….…………………………...…………………………………………… 21
6.3 Snapshot 3.…….…………………………...…………………………………………… 22
6.4 Snapshot 4.…….…………………………...…………………………………………… 22
Chapter 7: TESTING………………………………………………………………………….. 23
7.1 Test Cases…………………………………………………………………………….. 24
Chapter 8: CONCLUSION AND DISCUSSION…………………………………………….. 26
8.1 Overall Analysis Of Project……………………………………………………............ 27
8.2 Dates of Continuous Evaluation i.e Internal Review 1 & Review 2………………….. 27
8.2 Problem Encountered and Possible Solutions…………………………………............ 28
8.3 Limitation and Future Enhancement…………………………………………………… 29
8.3 Conclusion…………………………………………………………………………….. 30
REFERENCES………………………………………………………………………………….. 31
2101031000251 Company Overview

CHAPTER – 1: COMPANY OVERVIEW

Silver Oak University 1 Silver Oak College of Engineering & Technology

 OVERVIEW OF OPERATION
2101031000251

1.1 ABOUT COMPANY

Technowire Data Science Limited (TDSL) is a public limited company incorporated on July 5,
2020, headquartered in Ahmedabad, Gujarat, India. The company specializes in delivering
robust financial and non-financial corporate data analytics services. Over the years, TDSL has
adeptly navigated technological advancements and market trends, evolving its business model,
expanding its verticals, and refining its product offerings to meet the dynamic regulatory
landscape of India.

1.2 DIFFERENT PRODUCTS AND SERVICES

TDSL is committed to providing comprehensive data analytics solutions by collecting data from
diverse sources, including the Ministry of Corporate Affairs, Goods and Service Department,
Employee Provident Fund Organisation, courts, credit bureaus, credit rating agencies, market
news, and other regulators. This extensive data collection enables informed decision-making,
improved customer experiences, and optimized business operations.

1.3 COMPANY VALUES AND PHILOSOPHY

TechnoWire is driven by a mission to bridge the gap between academia and industry,
empowering individuals to achieve their full potential. Their core values underpin this mission:

• Innovation in Technology: Technowire is dedicated to delivering sophisticated digital

workforce platforms that automate business processes. This commitment to innovation
reflects their focus on leveraging cutting-edge technology to enhance operational
efficiency. Scribd
• Empowerment of People: The company's vision emphasizes empowering
individuals by automating repetitive tasks, allowing employees to focus on creative and
strategic initiatives. This philosophy highlights their belief in human potential and the
importance of meaningful work. Scribd
• Adaptability to Market Dynamics: Technowire has demonstrated an ability to evolve
its business model and refine product offerings in response to technological
advancements and market trends. This adaptability showcases their commitment to
staying relevant and responsive in a dynamic regulatory landscape

Silver Oak University 2 Silver Oak College of Engineering & Technology

 OVERVIEW OF OPERATION
2101031000251

CHAPTER - 2: OVERVIEW OF OPERATIONS

2.1DESCRIPTION OF WORK IN EACH DEPARTMENT

1. Business Development
Engages with educational institutions and corporate clients to enhance Technowire’s
visibility and outreach. Builds and nurtures strategic partnerships, negotiates
collaborations, and identifies new business opportunities through market research and
trend analysis.

2. Project Management
Oversees the design, planning, and execution of academic-industry student projects.
Ensures alignment with academic standards and industry expectations. Collaborates with
mentors, students, and partner organizations for seamless project delivery.

3. Training & Development

Designs and delivers domain-specific training across AI, ML, IoT, and more. Works
alongside subject matter experts to ensure industry-relevant content. Conducts online
sessions, webinars, workshops, and manages certification programs.

4. Customer Success and Support

Provides onboarding assistance and technical support to students and institutions.
Troubleshoots issues and ensures user satisfaction by guiding users through platform
challenges. Collects feedback for ongoing service improvement.

Silver Oak University 2 Silver Oak College of Engineering & Technology

 OVERVIEW OF OPERATION
2101031000251
5. Product Development
Manages platform enhancement, UI/UX improvements, and feature rollouts. Works in
tandem with the tech team to ensure the system is user-friendly, scalable, and tailored to
stakeholder needs.

6. Marketing
Executes marketing campaigns to promote TechnoWire's offerings. Manages content,
social media, digital strategies, and organizes outreach events to boost brand engagement.

7. HR & Talent Acquisition

Recruits, trains, and supports TechnoWire’s employees and trainers. Focuses on team
onboarding, employee development, and fostering a positive company culture.

8. Finance
Handles budgeting, forecasting, billing, and financial audits. Manages payroll and ensures
compliance with financial norms while enabling efficient financial operations.

2.2 TECHNICAL SPECIFICATIONS

Product Development Department

• Cloud Infrastructure: AWS, Azure, or Google Cloud for scalable computing, storage, databases,
and container services (Docker, Kubernetes).
• Development Tools: IDEs like VS Code, JetBrains; version control via GitHub; task
management with JIRA.

Training & Development Department

• LMS Platforms: Moodle, Canvas, or custom-built systems for content delivery, tracking, and
management.

Project Management Department

• Project Tools: JIRA, Asana, Trello, or Monday for task tracking and project timelines.
• Collaboration Tools: Slack, Microsoft Teams, Google Workspace for team communication.

Marketing Department

• Design & Editing Tools: Adobe Suite (Photoshop, Illustrator, Premiere Pro), Canva.
• Marketing Analytics: Google Analytics, SEMrush, social media analytics platforms.
• Scheduling Tools: Hootsuite, Buffer for campaign scheduling and management.
Silver Oak University 2 Silver Oak College of Engineering & Technology
 OVERVIEW OF OPERATION
2101031000251
Customer Support Department

• Helpdesk Software: Zendesk, Freshdesk, Intercom for managing inquiries.

• CRM & Feedback Tools: Salesforce, HubSpot for support ticket tracking and customer
satisfaction.

Finance Department

• Accounting Software: QuickBooks, SAP, Oracle Financials for transaction and payroll
management.
• Data Visualization: Python, SQL, Tableau, Power BI for analysis and reporting.

2.3 SCHEMATIC LAYOUT OF THE PRODUCTION PROCESS

1. Curriculum and Content Development

• Research industry trends to identify in-demand skills and technologies.

• Develop a modular curriculum with defined learning outcomes.
• Create video lectures, presentations, coding tasks, and project frameworks.

2. Platform Development and Integration

• Implement or update features for LMS, code submission, and student interaction.
• Upload and configure all educational content and assessments on the platform.

3. Student Onboarding and Enrollment

• Promote offerings via digital channels and institutional partnerships.

• Support student registration, orientation, and onboarding processes.

4. Training Delivery and Support

• Conduct live mentorship sessions and webinars.

• Provide real-time technical support through various communication channels.

5. Project Execution and Evaluation

• Assign real-time industry-relevant projects.

• Evaluate based on detailed rubrics and provide constructive feedback.

6. Certification and Post-Training Engagement

Silver Oak University 2 Silver Oak College of Engineering & Technology

 OVERVIEW OF OPERATION
2101031000251
• Issue completion certificates.
• Offer career support services including resume and interview guidance.

7. Continuous Improvement

• Monitor metrics like completion rate, engagement, and feedback.

2.4 Stages of Production Explained

1. Requirements Gathering

Identify features and expectations through user input. Output: SRS document outlining
project scope and functionality.

2. System Analysis

Assess technical, operational, and economic feasibility. Helps identify constraints or risks
early in the process.

3. System Design

Create architecture and detailed system components. Focuses on security, scalability, and
maintainability.

4. Development

Write code using Python and Django. Develop core modules like billing, payment
integration, etc. Also includes initial debugging.

5. Testing

Execute test cases to validate features. Find and fix bugs to ensure system reliability and

Silver Oak University 2 Silver Oak College of Engineering & Technology

 OVERVIEW OF OPERATION
2101031000251
compliance with requirements.

6. Deployment

Deploy the system on production servers. Configure databases, ensure server uptime, and
monitor application health.

7. Maintenance & Support

Post-deployment support involves updates, user assistance, and system optimization to

maintain performance.

CHAPTER - 3: INTRODUCTION

Silver Oak University 2 Silver Oak College of Engineering & Technology

2201031030016 Introduction of Project

Silver Oak University 8 Silver Oak College of Engineering & Technology

2201031030016 Introduction of Project
3.1 Project Summary

This project focuses on building an AI/ML-based inventory forecasting system tailored

for vendors in e-commerce marketplaces. It leverages the ARIMA model to predict
product demand accurately by analyzing historical sales data. The system is designed to
identify patterns such as trends and seasonality, enabling vendors to make data-driven
inventory decisions. This minimizes the risk of overstocking and stockouts, ultimately
optimizing supply chain performance and improving financial outcomes. The solution is
scalable and adaptable across diverse product categories and is implemented using
Python and standard data science libraries.

3.2 PURPOSE
The primary purpose of SnapOCR is to provide users with a fast, reliable, and AI-driven solution
for extracting text from images. The app aims to enhance productivity by
converting visual content into editable and searchable digital text—efficiently and offline.
SnapOCR serves a wide range of users, including students, professionals, and travelers, offering a
simple yet powerful OCR experience.

Key Objectives:

4 Enable Multi-Source Text Extraction: Allow users to extract text from live screens, camera
input, and gallery images (PNG/JPG formats).

5 Ensure Offline Functionality: Operate without an internet connection, promoting accessibility

and preserving user privacy.

6 Promote User Accessibility: Provide an intuitive interface that makes OCR technology
accessible to both tech-savvy and casual users

3.3 OBJECTIVE
1. Convert Images into Editable Digital Text

▪ Use Optical Character Recognition (OCR) to extract text from screenshots, camera captures,
and gallery images .Support over 30 languages including English, Hindi, Chinese, and
Japanese

Silver Oak University 10 Silver Oak College of Engineering & Technology

2201031030016 Introduction of Project

2. Enhance Accessibility and Productivity

▪ Enable offline functionality to allow users to extract text anytime, anywhere

▪ Provide a lightweight solution (~5MB) with low battery consumption
▪ Facilitate quick copying, editing, and sharing of extracted text

3. Simplify the Text Extraction Process

▪ Use smart word segmentation and swipe selection for easier interaction
▪ Highlight extracted text using dynamic bounding boxes for better clarity
▪ Offer a user-friendly interface suitable for all age groups

4. Support Real-Time and On-Demand Use Cases

▪ Allow text capture from live screens and real-time camera input
▪ Process and display results instantly without server-side dependencies
▪ Cater to students, professionals, travelers, and visually impaired users

5. Build a Scalable and Future-Ready Platform

▪ Design a modular architecture for easy integration of future features like handwriting
recognition, cloud sync, and multi-format exports
▪ Stay adaptable for enhancements using AI and ML models
▪ Lay the foundation for advanced capabilities like real-time document scanning and
translation experts .

3.4 SCOPE
SnapOCR encompasses a wide range of functionalities designed to make Optical Character
Recognition (OCR) accessible, efficient, and multilingual. The app is developed for Android
platforms and focuses on offline processing, ensuring users can extract and edit text anytime,
anywhere. Below are the core areas that define the scope of SnapOCR:

Silver Oak University 10 Silver Oak College of Engineering & Technology

2201031030016 Introduction of Project
1. Image-Based Text Extraction

Extracts editable text from:

▪ Screenshots and saved images from the gallery (JPEG, PNG formats)

▪ Cropped images to focus on specific content areas

2. Multilingual Text Recognition

Supports text recognition in over 30 languages, including:

▪ English, Hindi, Chinese, Japanese, and more
▪ Regional languages supported by Tesseract OCR engine
▪ Options for users to choose preferred OCR language

3. Offline Functionality

Enables complete operation without internet access:

▪ Ensures user privacy with no need to upload data to cloud servers
▪ Fast and lightweight performance on most Android devices
▪ Ideal for use in remote or low-connectivity areas

4. User Experience and UI Tools

Offers intelligent and interactive tools such as:

▪ Swipe to select, copy, or edit extracted text instantly
▪ Dynamic bounding boxes to visualize recognized words
▪ Lightweight design with fast image processing and feedback

5. Limitations and Future Enhancements

While SnapOCR is powerful, it currently:

▪ Does not support handwriting recognition (planned future feature)
▪ Lacks cloud storage integration (coming in future releases)

3.5 TECHNOLOGY AND LITERATURE REVIEW

1. Core Technologies:

• FastAPI:
A high-performance Python web framework used for developing RESTful APIs.
FastAPI powers SnapOCR’s backend, enabling seamless communication between the
front end and OCR engine.

• Tesseract OCR:
An open-source Optical Character Recognition engine developed by Google. It supports
over 100 languages and performs highly accurate offline text extraction from images.
Silver Oak University 10 Silver Oak College of Engineering & Technology
 2201031030016 Introduction of Project

• OpenCV (Open Source Computer Vision Library):

A powerful computer vision library used for preprocessing images—converting them to
grayscale, resizing, denoising, and enhancing contrast—to improve OCR performance.

• Pillow (PIL):
Python Imaging Library used for basic image manipulations like cropping, rotating.

• PyTesseract:
A Python wrapper for Tesseract OCR that connects the backend with OCR processing
logic, making it easier to integrate image-to-text features.

2. Literature Review

The development of SnapOCR draws on extensive research in the fields of computer vision, image
preprocessing, and OCR technologies:

• OCR Algorithms and Techniques:

Studies on OCR engines such as Tesseract and its language training data guide
SnapOCR’s implementation, ensuring accurate extraction from noisy or complex images.

• Image Preprocessing in OCR:

Research highlights the importance of preprocessing (grayscale conversion, thresholding, resizing) to
boost OCR accuracy. SnapOCR applies these best practices using OpenCV.

• Mobile OCR Applications:

Previous mobile OCR projects inform the app’s offline capabilities and lightweight design, essential
for user accessibility in low-connectivity regions.

• Human-Computer Interaction (HCI):

Design principles derived from HCI literature shape SnapOCR’s simple UI/UX, making it intuitive
for students, travelers, and professionals alike.

Silver Oak University 10 Silver Oak College of Engineering & Technology

2201031030016 Introduction of Project

3.6 Project Planning

 Define Project: Set objectives, scope, and identify stakeholders.

 Gather Requirements: List functional, technical, and non-functional needs.

 Design & Architecture: Outline system structure, database schema, and API design.

 Plan Schedule: Break down tasks, set milestones, and establish a timeline.

 Development: Set up environments, code, and review progress.

 Testing: Conduct unit, integration, and user acceptance testing.

 Deploy & Launch: Deploy to Azure, launch the product.

 Maintenance: Monitor performance, gather feedback, and plan improvements.

Silver Oak University 10 Silver Oak College of Engineering & Technology

2201031030016 Introduction of Project

3.7 Project Scheduling (Gantt Chart)

figure 1.7.1 Project scheduling Gantt chart

Silver Oak University 11 Silver Oak College of Engineering & Technology

 System Analysis
2201031030016

CHAPTER - 4: SYSTEM ANALYSIS

4.1 STUDY OF CURRENT SYSTEM

In In this section, we explore the current landscape of OCR-based mobile applications, analyzing
existing solutions, methodologies, and tools used to understand their structure and limitations.

1. Delivery Models:
Most OCR applications available in the market are offered through mobile app stores, where users
install and use them independently. These apps range from lightweight OCR scanners to feature-rich
enterprise-grade tools. Many require an active internet connection
and depend on cloud-based OCR engines.
2. Curriculum and Content (Feature Set):
Existing apps typically include text extraction, image-to-text conversion, language support, and basic
file-saving functionality. However, few apps integrate real-time camera capture, multilingual offline
recognition, or personalized user storage.

3. Learning Methodology (Usage Flow):

Silver Oak University 12 Silver Oak College of Engineering & Technology
 System Analysis
2201031030016

Users usually follow a simple flow: capture an image → process the image using OCR → extract and
view text → copy/share/save the result. Advanced apps sometimes include preprocessing tools to
improve recognition accuracy.

4. Assessment and Evaluation (User Feedback):

User reviews and feedback indicate that many apps struggle with accuracy, especially with noisy or
unclear images. Offline functionality and multilingual support are also commonly requested features.

4.2 PROBLEM AND WEAKNESS OF CURRENT SYSTEM

1 .Limited Offline Functionality:

▪ Many OCR tools depend heavily on cloud-based services, which limits their usability in low-
connectivity environments.
▪ This creates accessibility issues for students, travelers, or field workers.

5 Poor Multilingual Support:

▪ Some apps lack robust support for non-English languages.

▪ Users working with diverse languages find recognition quality inconsistent.

6 Accuracy and Preprocessing Challenges:

▪ Image preprocessing is often minimal or absent.

▪ OCR accuracy drops significantly for skewed, low-light, or noisy images.

7 User Experience Limitations:

▪ Complex interfaces can overwhelm non-technical users.

▪ Some apps lack clear workflows, leading to confusion or errors during usage.

8 Limited Personalization and History Tracking:

▪ Users cannot save or revisit past scans.

▪ Absence of a history or log system reduces long-term usability.experience.

Silver Oak University 12 Silver Oak College of Engineering & Technology

 System Analysis
2201031030016

4.3 REQUIREMENT OF NEW SYSTEM

To address these gaps, SnapOCR is designed with the following essential requirements:

Core Requirements:

• Reliable Offline OCR:

The system should use an accurate, offline OCR engine (Tesseract) to ensure access anywhere without
an internet connection.
• Robust Image Preprocessing:
OpenCV-based image processing to enhance clarity, contrast, and alignment of images before text
extraction.
• Simple and Intuitive User Interface:
Kotlin-based Android interface with minimalistic design to help users navigate without confusion.
• Scan History and Local Storage:
Store extracted text and image history using SQLite, allowing users to access previously scanned
documents.
• Responsive System Architecture:
FastAPI backend should enable smooth integration of features and ensure scalable, fast processing
when needed.
• Performance Tracking and Feedback:
Incorporate optional analytics to measure recognition accuracy and user satisfaction for future
enhancements.
• Cross-Platform Compatibility:
The app should work on a wide range of Android devices and support image input from camera or
gallery.

4.4 SOFTWARE AND TECHNOLOGY USED

1. CRM OCR Technology:

• Tesseract OCR for robust, multilingual, and offline text extraction from images.

Silver Oak University 12 Silver Oak College of Engineering & Technology

 System Analysis
2201031030016
Image Preprocessing Technology:
• OpenCV and Pillow (PIL) for grayscale conversion, thresholding, resizing, noise reduction, and other
enhancements.

2. Backend Technology:
• FastAPI, a modern Python-based framework, used to implement the API and connect OCR logic with the
frontend.

3. Mobile Development Technology:

• Android SDK with Kotlin for building the mobile app interface, camera access, and user interaction.

4. Local Data Storage:

• SQLite for storing scan history and extracted data locally on the device.

5. Integration Libraries:

• PyTesseract to connect Python backend with Tesseract’s OCR functionalities.

UI/UX Design:

• XML layouts and Kotlin logic to provide an intuitive and responsive mobile interface.

4.5 List Of Main Modules

1. Data Preprocessing Module

 Cleans and structures historical sales data.

 Handles missing values, outliers, and time indexing.

2. Demand Forecasting Engine

Silver Oak University 12 Silver Oak College of Engineering & Technology

 System Analysis
2201031030016
 Trains and applies ARIMA models.

 Predicts short-term and long-term product demand.

1. Result Evaluation & Model Tuning

 Measures model accuracy (e.g., MAPE, RMSE).

 Performs parameter tuning and model selection.

2. Visualization and Reporting Module (Upcoming)

 Graphical insights and trend analysis (via matplotlib).

 Future plan: Streamlit-based dashboard.

3. Scalability & Adaptation Layer

 Supports batch processing of multiple products.

 Adapts to vendor-specific datasets.

4. Future Modules (Planned)

 Hybrid ARIMA + LSTM forecasting.

 Real-time data integration.

 User-friendly vendor dashboard

4.6 Selection of Hardware and Software

4.6.1 Requirements of Hardware

● Processor: Intel i5 or higher for efficient processing

● RAM: 8GB or more to handle data-intensive tasks

● Storage: At least 10GB of free space for datasets and project files

Silver Oak University 12 Silver Oak College of Engineering & Technology

 System Analysis
2201031030016
● GPU (Optional): For faster model training

● Platform: Personal computer or laptop suitable for running Python and machine learning
librarieswith the hotel management software.

4.6.2 Requirements of Software

● Operating System: Windows, macOS, or Linux

● Programming Language: Python 3.x

● IDE: Jupyter Notebook

● Libraries: pandas, NumPy, statsmodels & pmdarima, scikit-learn, matplotlib

● Interface Tool: Gradio for interactive sentiment feedback

Silver Oak University 12 Silver Oak College of Engineering & Technology

2201031030016 System Design

CHAPTER - 5: SYSTEM DESIGN

Silver Oak University 17 Silver Oak College of Engineering & Technology

2201031030016 System Design

5.1 GANTT CHART

figure 5.1.1 gantt chart

5.2 FLOWCHART

figure 5.1.2 working flowchart

Silver Oak University 18 Silver Oak College of Engineering & Technology

2201031030016 System Design

5.1.1 Activity Diagram

figure 5.1.3 Activity diagram

Silver Oak University 19 Silver Oak College of Engineering & Technology

2201031030016 Implementation

CHAPTER - 6: RESULTS

Silver Oak University 20 Silver Oak College of Engineering & Technology

2201031030016 Implementation

Fig. 6.1 Landing page

Fig. 6.2 Uploading page

Silver Oak University 21 Silver Oak College of Engineering & Technology

2201031030016 Implementation

Fig. 6.3 Extracted text

Fig. 6.4 Uploads screenshot

Silver Oak University 22 Silver Oak College of Engineering & Technology

2201031030016 Testing

Fig. 6.5 About us

Silver Oak University 23 Silver Oak College of Engineering & Technology

2201031030016 Testing

Fig. 6.6 More

Silver Oak University 23 Silver Oak College of Engineering & Technology

2201031030016 Testing

Fig. 6.7 Login/Register Page

Silver Oak University 23 Silver Oak College of Engineering & Technology

2201031030016 Testing
7.1Test Cases

Test Case Expected

Test ID Test Steps Actual Result Pass/Fail
Description Result
1. Open data
System
upload
Verify data should
module File uploaded
TC001 upload accept and Pass
2. Upload CSV successfully
functionality parse file
file with sales
successfully
data
1. Upload Missing
Validate
data with values
preprocessing Error thrown due
TC002 missing fields handled Fail
handles missing to NaNs
2. Trigger (imputed or
values
preprocessing dropped)
1. Load ARIMA
preprocessed model is
Validate ARIMA data trained Model trained
TC003 Pass
model training 2. Run without successfully
training errors and
module saved
1. Select 7-day
Test short-term product demand
Forecast graph
TC004 demand forecast forecast Pass
2. Generate displayed
generation generated
forecast
and plotted
1. Select
30-day
Test long-term product
forecast Graph and values
TC005 demand forecast 2. Generate Pass
results shown
generation 30-day generated
forecast
System
1. Run
shows
Verify inventory forecasting
restock Optimization
TC006 optimization Pass
2. Navigate to quantity or displayed
output
optimization overstock
suggestions alert
1. Open
dashboard Dashboard
Validate
loads with
dashboard 2. View Graph loaded as
TC007 correct Pass
displays forecasts product expected
prediction
correctly forecast graph, stats
visualization

Silver Oak University 24 Silver Oak College of Engineering & Technology

2201031030016 Testing
System
Verify system
1. Upload .txt shows error
handles invalid Error message
TC008 file instead message: Pass
file format displayed
of .csv "Invalid file
gracefully
format"

1. Upload new Forecast

sales data should
Test integration
2. Trigger reflect new Forecast updated
TC009 of updated real- Pass
update trends from correctly
time sales data
updated
3. Check
data
forecast
1. Upload
dataset with System
Test model error <10 data should
TC010 handling for points show Error displayed Pass
insufficient data "Insufficient
2. Run
data" error
training
Fig. 7.1 Test cases

Silver Oak University 24 Silver Oak College of Engineering & Technology

2201031030016 Conclusion and Discussion

CHAPTER - 8: CONCLUSION AND DISCUSSION

Silver Oak University 26 Silver Oak College of Engineering & Technology

 Conclusion and Discussion
2201031030016

8.1 OVERALL ANALYSIS OF INTERNSHIP VIABILITIES

Salesforce The SnapOCR internship provided a highly relevant and practical experience in the
domain of Optical Character Recognition (OCR) and application deployment. It involved
building a real-world utility that helps users extract text from images with ease, hosted via an
interactive web interface. This internship enabled skill development in Python, image processing
(using Tesseract OCR), and full-stack deployment with Streamlit. While the flexibility and
autonomy in the internship encouraged creativity, challenges such as OCR accuracy, noisy input
images, and limited team collaboration occasionally impacted efficiency. Future OCR-focused
internships can benefit from improving model performance, expanding use cases (e.g.,
handwriting detection), and incorporating user feedback loops to make apps more robust and
user-friendly.

8.2 Dates of Continuous Evaluation i.e Internal Review 1 & Review 2

1st Internal Review Record

Presentation From To
Time
Presentation Date
Date
Table 8.1 Evaluation Date Review 1

2st Internal Review Record

Presentation From To
Time
Presentation Date
Date
Table 8.2 Evaluation Date Review 2

Silver Oak University 27 Silver Oak College of Engineering & Technology

 Conclusion and Discussion
2201031030016

Photograph with team lead

8.4 SUMMARY OF INTERNSHIP

The SnapOCR internship revolved around designing and developing an intuitive webbased OCR
application. Interns explored key tools such as Python, Tesseract OCR, OpenCV, and Streamlit to
create a lightweight platform where users could upload images and extract embedded text
efficiently. The internship emphasized frontend-backend integration, usability testing, and
deployment of real-time image-processing solutions. Participants learned about error handling,
multilingual OCR, and optimizing image preprocessing for better accuracy. This experience
provided an excellent foundation for roles involving machine learning deployment, document
automation, and Python web development. It also helped interns sharpen their problem-solving
and debugging skills while focusing on delivering a minimal yet impactful tech solution.

8.5 LIMITATION AND FUTURE ENHANCEMENT:

8.5.1 Limitations
• OCR Accuracy Limitations: The base Tesseract model struggles with low-resolution
images, handwritten text, and noisy backgrounds.

• Limited Multilingual Support: While basic language packs were available, OCR
accuracy dropped in regional or non-standard fonts.

Silver Oak University 27 Silver Oak College of Engineering & Technology

 Conclusion and Discussion
2201031030016

• Design Constraints: Streamlit's layout limitations made custom UI/UX design more
restricted.

• No Database or Text History: Users couldn't view or download previously extracted text
from the app.

• Minimal Feedback Mechanism: The app lacked built-in options for user feedback or
rating the results.

8.5.2 Future Enhancements

• Model Upgrade & Preprocessing Improvements: Integrating advanced OCR models like
EasyOCR or PaddleOCR with preprocessing filters to improve accuracy.
• Multilingual & Handwriting Recognition: Expand support for multiple languages, regional
scripts, and basic handwritten OCR features.
• Enhanced UI/UX: Using custom components or migrating to React-based frontend for improved
user experience.
• User Text History & Downloads: Add functionality for users to save extracted text or export
results in various formats.
• Feedback Integration: Collect real-time feedback for continuous improvement of OCR output
and user satisfaction.
• API Deployment: Offer SnapOCR as a public/private API that other developers or platforms can
use.

• By addressing these areas, the SnapOCR internship experience can evolve into a more scalable
and professional-grade tool with broader impact in document digitization, accessibility tools,
and AI-driven automation.

Silver Oak University 27 Silver Oak College of Engineering & Technology

 Conclusion and Discussion
2201031030016

8.6 CONCLUSION

SnapOCR is a powerful and user-friendly AI-driven web application that simplifies the conversion
of image-based content into editable digital text through advanced OCR technology. By
integrating Tesseract OCR and effective image preprocessing, it delivers high-accuracy results
with multilingual support. Its real-time processing, confidence scoring, and interactive features
make it a valuable tool for users across education, research, and professional fields, ultimately
boosting productivity and eliminating manual data entry efforts.

Silver Oak University 27 Silver Oak College of Engineering & Technology

 Conclusion and Discussion
2201031030016

REFERENCES

1. Tesseract OCR Documentation. https://tesseract-ocr.github.io/

2. Streamlit Docs – Deploying Machine Learning Apps.
https://docs.streamlit.io/
3. EasyOCR GitHub Repository. https://github.com/JaidedAI/EasyOCR
4. OpenCV Tutorials – Image Preprocessing for OCR.
https://docs.opencv.org/
5. Medium (2023). Building a Simple OCR App using Python and Streamlit.

Silver Oak University 27 Silver Oak College of Engineering & Technology