rbaron / omr Goto Github PK

Optical Mark Recognition in Python

License: MIT License

Python 100.00%

omr's Introduction

omr

omr is a small Python 3 optical mark recognition script. It takes as input an image of an answered answer sheet and outputs which alternatives were marked. The project is generally optimized for conciseness and teachability, and the goal is to provide a reasonable starting point for learning and hopefully building more powerful applications.

Usage

$ python omr.py --help
usage: omr.py [-h] --input INPUT [--output OUTPUT] [--show]

optional arguments:
  -h, --help       show this help message and exit
  --input INPUT    Input image filename
  --output OUTPUT  Output annotated image filename
  --show           Displays annotated image

Example

$ python omr.py --input img/answered-sheet-photo.jpg  --output /tmp/results.png --show

Q1: A
Q2: C
Q3: C
Q4: E
Q5: N/A
Q6: N/A
Q7: A
Q8: N/A
Q9: N/A
Q10: N/A

In this case, we used the following image as input:

And got the following output:

Installation

Using pipenv with pyenv

Install pyenv

$ git clone https://github.com/rbaron/omr
$ cd omr/
$ pyenv install 3.7.5
$ pyenv local 3.7.5
$ pipenv --python 3.7.5
$ pipenv install
$ pipenv shell
$ py.test

Using virtualenv

$ git clone https://github.com/rbaron/omr
$ cd omr/
$ virtualenv --python=`which python3` venv
$ source venv/bin/activate
(venv) $ pip install -r requirements.txt
(venv) $ python omr.py --help

Anwer Sheet

The answer sheet is available in the sheet/ directory.

Development

Run unit tests

test_omr.py contains unit tests that can be run using:

$ py.test
=========================================================== test session starts ===========================================================
platform darwin -- Python 3.7.5, pytest-5.2.4, py-1.8.0, pluggy-0.13.0
collected 1 item

test_omr.py .                                                                                                                       [100%]

============================================================ 1 passed in 0.31s ============================================================

omr's People

Contributors

Stargazers

Watchers

Forkers

jsrj78 iderbukh kingmalza arthurj58 kiboom bellafontes tejastank drewyoungren lonicera pustar jcarhuazv tak-goto yaki-lk nursinghoming gursimar020 vedraiyani fullstackenviormentss raina99 kizzna scls19fr chapatiyaq ajscb007 ozymandiiaz okican82 hotahitz javid020 wutijat maewteam ccszwg kubilaycagri shyama95 tpponmat ashish-cloned-forked-repo februa22 phil1618 bahrulhikmi ranfu xettri doctral wanghyungjun raihanba13 tamutamu-archive mwta kena421 manikant92 kubantjan mertdogan aqeelikh artitsamchooosri marufhasan1 charmsoya manzoorhusain sajidmeo soap kiran-taylor nzasif zekkontro kiranbalamalynur kuzz84 yemenpython zadok7 ai-motive icodein edwardceballos kyusungpark ridaayed truncgil bolllovely piyush-multiplexer ariannexux rodanmuro linback acdwisu elshadaghazade bmcbride22 yakiyo egemenakan visncoder falsecometc76 ganesh20101

omr's Issues

How Other Then sample sheet be evaluated

Dear Sir I am facing problem when I give some sheet other then sample give in project How I fix this kindly guide me. Suppose I want to evaluate 100 number of question sheet. And It Produce correct result only when diagonal sample page given to it he I fix it?

Python 3

Hello,

It will be great to have a Python 3 version of this script.
#7 is an unsuccessful attempt to convert from Python 2 to Python 3.

Kind regards

calculating the corner feat

Could you explain how to get corner feats

Column issue

H rbaron,
I'm having some trouble when trying to make some more Column in your code.
Please help!

Enable continuous integration

Hello,

Continuous integration should be enabled (Travis-CI for example)

CI could run unit tests but could also check (using flake8 for example) that code respect PEP8.

Best regards

error while compiling in Windows

Traceback (most recent call last):
File "omr.py", line 251, in
main()
File "omr.py", line 234, in main
answers, im = get_answers(args.input)
File "omr.py", line 189, in get_answers
contours = get_contours(im)
File "omr.py", line 27, in get_contours
image_gray, cv2.RETR_TREE, cv2.CHAIN_APPROX_SIMPLE)
ValueError: need more than 2 values to unpack

Sample data should be used as unit tests

Hello,

https://github.com/rbaron/omr/tree/master/img provide sample data
maybe this kind of data could be used as unit tests.

Best regards

90% fail rate on scanned sheet.pdf

Nice work, thank you. Maybe you can see something I missed while testing with the scanned sheet.pdf

$ python omr.py --input ./scan_2022-01-07-162351_001.jpg  --output /tmp/results.png --show

Q1: A
Q2: C
Q3: C
Q4: E
Q5: N/A
Q6: N/A
Q7: A
Q8: N/A
Q9: N/A
Q10: N/A

scan_2022-01-07-162351_001.jpg (using sheet.pdf):

results.png

perspective transform is not done correctly

Hi,
Instead of detecting the corners as in the picture here, the calculation is made over the corners of the largest rectangle containing the contours (corner.png). How can we do this so that it detects the corners of each corner.png.
I think we should develop this project. this project is one of the nice omr (made with python) projects on github.
https://camo.githubusercontent.com/b3272c76cfe1dbd447255e330e5e7f8d4425621b26853b2724455b6b30502554/687474703a2f2f692e696d6775722e636f6d2f346e39664b46462e706e67

wrong project

when i want to use this project the python said

SyntaxError: invalid syntax

please help me

getting multiple marks from the same question

hi @rbaron. , Question no-6 two options are marked Option-B and Option-D because it is marked with two options program bypasses it. is it possible to get multiple marks from the same question?

Different bubblesheet

It is simple and good code.
If we want to create different type of bubblesheets, What changes need to done in the code. Can you add documents or some tutorial for the same.

2 marked answers for the same question

Here is an example image with 2 marked answers for Q1: A and C

Unfortunately, the current OMR code doesn't detect them:

Q1: N/A
Q2: C
Q3: C
...

How would you modify your code to allow multiple answers per question @rbaron?

Image download link

How to get a new corner features,such as a black square

CORNER_FEATS = (
0.322965313273202,
0.19188334690998524,
1.1514327482234812,
0.998754685666376,
)

Hi rbaron different sheet

Hello dear Baron. I want to use a different sheet, but i can't do this. I changed fixed dimension (height of each question is 40) because i used 20 questions. What changes need to done in the code?

Omr

No matching distribution found for opencv-python==4.2.0.32

[ra@x omr]$ virtualenv --python=`which python3` venv         

created virtual environment CPython3.10.1.final.0-64 in 229ms
  creator CPython3Posix(dest=/home/ra/src/omr/venv, clear=False, no_vcs_ignore=False, global=False)
  seeder FromAppData(download=False, pip=bundle, setuptools=bundle, wheel=bundle, via=copy, app_data_dir=/home/ra/.local/share/virtualenv)
    added seed packages: pip==21.3.1, setuptools==60.1.0, wheel==0.37.1
  activators BashActivator,CShellActivator,FishActivator,NushellActivator,PowerShellActivator,PythonActivator

[ra@x omr]$ source venv/bin/activate

(venv) [ra@x omr]$ pip install -r requirements.txt                  

Collecting atomicwrites==1.3.0
  Using cached atomicwrites-1.3.0-py2.py3-none-any.whl (5.9 kB)
Collecting attrs==19.3.0
  Using cached attrs-19.3.0-py2.py3-none-any.whl (39 kB)
Collecting importlib-metadata==0.23
  Using cached importlib_metadata-0.23-py2.py3-none-any.whl (28 kB)
Collecting more-itertools==7.2.0
  Using cached more_itertools-7.2.0-py3-none-any.whl (57 kB)
Collecting numpy==1.17.4
  Using cached numpy-1.17.4.zip (6.4 MB)
  Preparing metadata (setup.py) ... done
ERROR: Could not find a version that satisfies the requirement opencv-python==4.2.0.32 (from versions: 3.4.0.14, 3.4.10.37, 3.4.11.39, 3.4.11.41, 3.4.11.43, 3.4.11.45, 3.4.13.47, 3.4.15.55, 3.4.16.57, 3.4.16.59, 3.4.17.61, 4.3.0.38, 4.4.0.40, 4.4.0.42, 4.4.0.44, 4.4.0.46, 4.5.1.48, 4.5.3.56, 4.5.4.58, 4.5.4.60, 4.5.5.62)
ERROR: No matching distribution found for opencv-python==4.2.0.32

(venv) [ra@x omr]$ python -V
Python 3.10.1

answer sheet coordinate settings

Hi, can you explain in detail all the settings below to create a new answer sheet? For example, why is the answer sheet 740x1049 according to which unit?
or why ANSWER_PATCH_HEIGHT_WITH_MARGIN what does 80 represent and in what unit?
Thanks..
ANSWER_SHEET_WIDTH = 740
ANSWER_SHEET_HEIGHT = 1049

ANSWER_PATCH_HEIGHT = 50
ANSWER_PATCH_HEIGHT_WITH_MARGIN = 80
ANSWER_PATCH_LEFT_MARGIN = 200
ANSWER_PATCH_RIGHT_MARGIN = 90
FIRST_ANSWER_PATCH_TOP_Y = 200

ALTERNATIVE_HEIGHT = 50
ALTERNATIVE_WIDTH = 50
ALTERNATIVE_WIDTH_WITH_MARGIN = 100

rbaron / omr Goto Github PK

omr's Introduction

omr

Usage

Example

Installation

Using pipenv with pyenv

Using virtualenv

Anwer Sheet

Development

Run unit tests

omr's People

Contributors

Stargazers

Watchers

Forkers

omr's Issues

Recommend Projects

Recommend Topics

Recommend Org