OCR - Digit Slant Detection and Slant Correction

Hi,

I am working on Autoamted Handwritten Digit Recognition with deep learning algorithm. Since the handwritten digits have a lot of variation, in terms of the slanted angles, so I would like to look for the potential solutions for digit slant detectiona and slant correction.

I have tried Hough Transform, however, it requires edges as reference and digit doesn't have a proper edge.

The example of slanted digit is shown in below:

I would like to correct its orientation so that my NN can recognize it accurately.

Thank You :D

Answer 1

[Shashank Gupta]

Hi,

There are generally 2 ways to deal with such problems. Either you rectify the orientation of image and train the model for straight up digits or you can add images with different orientation in your training data and let your deep model handle such complexity. I would prefer second approach because correcting the orientation is difficult task and cannot be generalised effectively which will result in error and in turn affect the detection also. So I would suggest you to augment your data to take care of such random rotation. it will make your model more robust. The easy way to deal with this is to use imageDataAugmentor function during training. It will help you create digits image with random rotation and then you can train your model with such variety, in the end you will obtain a robust model which can take of rotation by itself.

I hope this helps.

Cheers.

[Tan Wai Chun]

Thanks Shashank Gupta. I able to make it with Radon Transform algorithm. Btw, thanks for your valuable info ! :D