- Dataset is not provided for this task. So there must be some reason for that. Reasons can be :
- We have to find dataset for this task from somewhere.
- We have to make our own dataset.
- We have to approach to find the solution for this task from some other method that can be possible to be done in the given time-frame.
I have borrowed the images of college IdCard from my friend as my phone camera is not working properly right now.
-
ExplainOneImage.ipynb -> This file contains all the detailed information how i have done the task with detailed description.
-
BulkImages.py -> This python file contains the code for performing the task on all images inside the folder.
-
images -> Put all your images inside this folder.
-
images-result -> Resultant images will be stored inside this folder.
- numpy
- PIL
- OpenCV Version: 3.4.3
- Simply to reduce the noise
- it removes noise while keeping edges relatively sharp.
- It performs good in all the conditions where image has different lighting conditions in different areas.
For more details on this read on docs :- https://docs.opencv.org/2.4/doc/tutorials/imgproc/erosion_dilatation/erosion_dilatation.html
- cv2.RETR_EXTERNAL => If we use this flag, it returns only extreme outer flags. All child contours are left behind.
- cv2.CHAIN_APPROX_SIMPLE => It will only give us the endPoints.It will compresses horizontal, vertical, and diagonal segments and leaves only their end points.