/Image-To-Text

Optical Character Recognition using Convolutional Recurrent Neural Network

Primary LanguagePythonMIT LicenseMIT

Image-To-Text

Datasets instructions

Go to the datasets folder

Link given in the download_and_extract file

Download the datasets and extract in the datasets folder only from the given link

Or you can use this commads in command line (your path must be in datasets folder only):

wget https://www.robots.ox.ac.uk/~vgg/data/text/mjsynth.tar.gz
tar -xvzf mjsynth.tar.gz

new-model.py usage

python new-model.py

To test a single image prediction

Change the filename in the source code and run

python new-model-test.py

Size of the image must be Height = 32 and Width = 128