Open-sourced tool that uses opencv and tensorflow to 'learn' and identify stuff from images

• OpenCV is used to manipulate the properties of the image
• Tensorflow is used for the learning of 'elements' within the image

• v1.0

Able to extract digits out of an image

1. Fix contour regions (don't scan for the contour within the zero)
2. Increase accuracy for digit recognition

For ubuntu: Refer to http://www.pyimagesearch.com/2015/07/20/install-opencv-3-0-and-python-3-4-on-ubuntu/ to install OpenCV3, and numpy on your system

Refer to https://www.tensorflow.org/versions/master/get_started/os_setup.html#download-and-setup to setup tensorflow

To run an example:

cd ~
git clone https://github.com/kendricktan/openimif.git
cd openimif
python setup.py image.png