0% found this document useful (0 votes)
15 views12 pages

Automated Handwritten Marks Recognition

ppt for capstone

Uploaded by

shreypachauri3
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
15 views12 pages

Automated Handwritten Marks Recognition

ppt for capstone

Uploaded by

shreypachauri3
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd

Smart Scan

Shrey Pachauri (102103354)


Chirag Gupta(102103278)
Geetika Arora(102153027)
Ishaan Gaba(102103281)
Abhinav Maheshwari(102103284)

Mentor : Dr. Smita Agrawal


Problem Definition, Project Scope,
and Objectives
1 Problem Definition
This project aims to automate the capture and recognition of handwritten marks
from student answer sheets to reduce errors and improve efficiency.

2 Project Scope
• Automates capture and recognition of handwritten marks.
• Reduces errors in grading and data entry.
• Enhances efficiency and saves time.
• Streamlines the examination process.

3 Objectives
• Enhance Efficiency: Speed up and streamline the capture of key details
from answer sheets.
• Improve Accuracy: Minimize errors in data extraction with advanced algorithms.
• Save Time: Automate processes to reduce administrative workload.
• Simplify Access: Ensure easy retrieval and use of captured data.
Detailed Design
Tools/Platforms Used Data Design
• React: Front-end interface development. • Answer Sheets: Images captured
• Firebase: Backend data storage and from student answer sheets.
management. • Marks Tables: Structured data
• Python & YOLO: Mark recognition representing marks extracted from the
model development and training. • sheets.
Recognition Outputs: Digitized
• Roboflow: Dataset preprocessing and marks stored in Firebase.
augmentation for improved accuracy. • Metadata: Associated information like
student ID, exam date, and sheet
identifiers.
Swimlane Diagram
DFD (LEVEL-1)
User Interface Design

Login Functionality Mark Verification Screen Teacher's Dashboard


The user interface provides a The system displays After authentication , the user
simple Login form to securely recognized marks on a is directed to dashboard to see
authenticate the user and dedicated screen, allowing all the students whose marks
redirect the user to the teachers to verify the accuracy are successfully scanned.
dashboard. of the recognition process.
Cost Analysis
Category Cost(in Rs)

ESP32 Camera 2500

Wires and Batteries 500

Structure/Framework/ 3000
Materials

Miscellaneous Costs 1000


Project Outcomes
1 Final Product/Service 2 Working Prototype
• The final automated marks • The working prototype showcases
recognition system offers a user- the system's capability to capture an
friendly and efficient solution for image, process it, and display the
• educators.
It captures and recognizes marks recognized marks.
from answer sheets automatically. • The prototype has undergone
• This system significantly reduces the testing and validation to ensure
time and effort needed for manual accuracy and reliability.
grading.
Individual Contributions

Shrey Pachauri Ishaan Gaba Chirag Gupta


Managed the frontend logic for Implemented the backend logic Implemented database and
the application and created a using firebase along with the data design logic along with
basic layout for the webapp implementation of ML Model. integration of ESP32 Cam
along with the implementation integration along with
of the ML Model. integration of ML Model.

Geetika Arora Abhinav Maheshwari


Managed the front end part of Designed the entire UI/UX
the webapp and helped in Logics for a seamless user
successful hosting and experience along with the
deployment of app along with implementation of ML Model.
the implementation of ML
Model.
Current Progress
1 Completion of the image preprocessing module
The image preprocessing module has been successfully
developed and integrated into the system.

2 Successful training and deployment of the YOLO


model
The YOLO model has been trained and deployed,
demonstrating accurate recognition of handwritten marks.

3 Development of the basic user interface


The basic user interface has been developed, providing a
foundation for further enhancements.
Future Work Plan
1 Automated Paper Handling Integration: Fully automate
document digitization using a roller-based paper feeding
mechanism synchronized with OCR for error-free processing.

2 Enhanced OCR & Data Extraction: Implement AI-driven


OCR to accurately digitize handwritten text, reducing manual
effort and errors.

3 Extended Analytics: Expand admin dashboard with


advanced visualizations and predictive analytics for deeper
academic insights.

4 Mobile App Development: Create a mobile-friendly app with


real-time updates and offline access, enhancing accessibility
for all users.
Thank You

You might also like