1
$\begingroup$

I'm new to ML world and been reading about ML and TensorFlow. My goal is to read the following example in real time with Android phone:

enter image description here

So I tried firebase ML OCR and it works really good, it reads the complete value but it does not read the decimal point and also reads a lot of the surrounding text. So my idea is that I should first detect black and red bounding boxes and then detect individual numbers inside

  1. is this the right way to go? How would I accomplish this?
  2. Also how do you use two kinds of a model, one to extract a part of the image (black and red bounding areas) and then pass them to OCR model?
  3. What about last digit which can always be in between two numbers (example: 1 and 2)?
$\endgroup$
0
$\begingroup$

Two options :

  1. Use pre-built libraries for OCR + Bounding Box detection (E.g.: https://www.pyimagesearch.com/2018/08/20/opencv-text-detection-east-text-detector/ for Bouding Box detection and then OpenCV / Tessract for OCR)

enter image description here

  1. Train a Deep learning model for Text Detection in scene . Examples : https://github.com/qjadud1994/CRNN-Keras and https://github.com/mvoelk/ssd_detectors
| improve this answer | |
$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.