Skip to content

The project is investigating methods to extract human-marked data from document forms such as surveys and tests. They can read questions, multiple-choice exam papers, and grade.

HarryxDD/OpticalMarkRecognition

Repository files navigation


Project to check multiple choice test with opencv and python

This project help teacher to check the students' answer faster and more exactly.

Report Bug · Request Feature


Table of contents:

  1. About the project.

  2. Built with.

  3. Getting started.

    i. Installation.

    ii. Data.

  4. Usage.

  5. Instructors.

  6. Contact.

About the project:

Nowadays, teachers face a major challenge in ensuring that their students' answers are correct and on time. With this project, teachers will spend nearly one minute checking all of their students' answers with a high correct rate of over 99.9 percent, and this project will analyze and summarize all important information about wrong or right answers, which difficult questions have a high error rate.

Built with:

  • Python
  • Visual studio
  • Github

Getting started:

Installation:

Opencv:

   pip install opencv-python
   conda install opencv (if you use conda)

Numpy:

    pip install numpy
    conda install numpy (if you use conda)

Pandas:

    pip install pandas
    conda install pandas (if you use conda)

Regex:

    pip install regex
    conda install regex (if you use conda)

Data:

In this project, we used 50 photos-this is the student result in the answer sheet. You can find it in folder "Student".

Example data:

With these datas, we compares the students' answers with the our answer to find the students' result. Our anwser in folder "answer".

Usage:

1/ Move to folder: "OpticalMarkRecognition" (cd OpticalMarkRecognition)

2/ Start with file :"Student_info.py" to  handle the infomation of students'answer.

3/ Continue with file:"answer.py" to analyze the root answer.

4/ Finally, we use file:"student.py" to check the right answer, wrong answer, and the final score of each student. All of the information will be display in the terminal.

5/ We can read file csv in folder:"csv_files".

Instructors:

Thank you Mr. Vu and Mr. Thien for the support and give us a wonderful challenges to improve our coding skill and use the python library better. Thank you very much and we hope we can get more challenge to improve our coding more and more.

Contact:

1/ Vu Truong:

Email: vu.truong200201@vnuk.edu.vn
Github: https://github.com/HarryxDD

2/ Bao Ngo:

Email: bao.ngo200202@vnuk.edu.vn
Github: https://github.com/ngobao2002

About

The project is investigating methods to extract human-marked data from document forms such as surveys and tests. They can read questions, multiple-choice exam papers, and grade.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages