roc198 / Grade-Auto-Filler

This is the course project of Image Processing course which we implement a tool to scan grade sheets or answer bubble sheets, then, make another grade sheet in digital form.

Grades Auto Filler

📝 Table of Contents

• About
• Technology
• Getting Started
• Grade Auto Filler Model
  • Overview
  • Flow Diagram
  • Results
• Bubble Sheet Correction Model
  • Overview
  • Flow Diagram
  • Results
• Contributors
• License

📙 About

• The main idea of (Grade Auto Filler) is to give an image of a table with some data to the program and get an output of excel sheet containing the data that was in that image after mapping the symbols to the wanted grades.

• The main idea of (Bubble sheet correction) islocalize the filled circles referenced to the number of columns and number of rowsand compare them to a given model answer and provide sheet of students’ grades.

💻 Technology

• Python
• Jupyter Notebook
• OpenCV

🏁 Getting Started

• Clone the repository
  

  git clone https://github.com/AhmedAlaa2024/Grade-Auto-Filler.git
  

• Install Packages
  

  pip install -r requirements.txt
  

• To run the Grade Auto Filler
  

  Set the path of the table image in the config file [configGrade.conf]
  
  Set any other needed features in the config file [get names, wanted method to detect symbols and numbers, etc..]
  
  Use the command "python gradesSheet.py"
  

• To run the Bubble Sheet Correction
  

  Set the path of the directory of the samples in the config file [configBubble.conf]
  
  If there is no id in the image, set the ids in the id list file
  
  Set all the other important parameters in the config file [number of students, number of questions, number of choices, id length if exist, number of coulmns and rows, etc...]
  
  Use the command "python bubbleSheet.py"
  

---

Grade Auto Filler Model

📷 OverView

• It allows you to turn an image into a digital form (excel sheet)
• It handles Skewing and orientation
• Printed Student ID is detected using OCR or Feature and execration
• Written Symbols like ✓ & ? are detect using HOG feature extractor and predicted using SVM or with normal image processing techniques
• Handwritten numeric values are detected using OCR and Feature and execration

---

🖍 Flow Diagram

---

📚 Results

(1)

(2)

---

Bubble Sheet Correction Model

📷 OverView

• It handles Skewing and orientation
• It handles different ink colors
• It allows different formats for the sheet ( but bubbles must be vertically aligned in all formats )
• Differnet number of questions and choices

---

🖍 Flow Diagram

---

📚 Results

(1)

(2)

---

Contributors

Ahmed Alaa

Beshoy Morad

Waleed Hesham

Zeyad Tarek

License

This software is licensed under MIT License, See License for more information.