This is a simple version of OCR (Optical Character Recognition) by using Tensorflow and OpenCV within Python3.
Firstly OCR (Optical Character Recognition) is usually made by combining several image processing techniques such as:
De-Skewing,
Cleaning,
Normalization,
Character isolation,
etc.
Then it will goes into the pattern match or pattern recognition process to know the pattern of that character.
Thus, this notebook will implement Tensorflow 2.0 as CNN within OpenCV2 to make a Simple OCR; whereas:
1. Tensorflow for the core (pattern recognition) which trained using the dataset from Sueiras[1]
2. The model from tensorflow will be used to identify each character detected in the result of OpenCV process
Contact me at LinkedIn
[1] J. Sueiras, et al.: "Using Synthetic Character Database for Training Deep Learning Models Applied to Offline Handwritten Character Recognition", Proc. Intl. Conf.Intelligent Systems Design and Applications (ISDA), Springer, 2016.