This API is created using OpenCV and Django. It uses The Circle Hough Transform technique to find the circles and marks the center and radius of each circle.
The API returns response in JSON format i.e.
Test Image: {'success': True, 'Number of Coins': 17, 'coins(x,y,r)': [[847, 2357, 185], [1367, 2361, 190], [2699, 1153, 226], [2001, 667, 220], [2229, 1235, 189], [1771, 1467, 181], [1409, 1931, 190], [2777, 2263, 202], [2459, 1819, 186], [1125, 793, 221], [1585, 1077, 188], [3037, 1657, 222], [1925, 1931, 188], [795, 1709, 182], [585, 1183, 183], [1275, 1459, 191], [1839, 2519, 185]]}
Libraries and frameowrks required for the API to run are:
- Django
pip install django - OpenCV
pip install OpenCV
To run the api on your local server, you need to install the requirements and run the following command by opening command line from the project directory.
python manage.py runserver
The API can be tested using curl or the included api_test.py file.
Using api_test.py also displays the result image.
Using curl we get the response in JSON.
-
To test the API using included test_api python file,
- run the
api_test.pywith path to the image as second argument .i.e
python api_test.py [path_to_image]
- run the
-
To test the file using curl, use the following command
-
curl -X POST -F image=@[path_to_image] "[api_endpoint**]"** For local server use "http://127.0.0.1:8000/coin_detect/"```
-
Need to add some more funtionality. Will add soon