Bubble Answers Selection Using Document Understanding

How to read the OMR bubble answers using Document Understanding?

Issue Description: There is a need to understand the bubbling response on each question.


Use ML model(Document understanding Model) by labelling the entire row: "5 A C D" in which the Robot would parse by determining this is question 5, letter B is missing, therefore the choice is B. Refer the Machine Learning Packages .

Note:: The model first needs to be trained with the required dataset. Refer KB Article - "How to Train Document Understanding Model in AI-Fabric" to know more about the procedure.

However, this will only work if these options are only laid out vertically on the page, so that there are no multiple questions on the same row of text.

** If there are multiple of these sets of options on the same row, the above suggested approach will not work.
Also, options which are crossed out, and others selected instead, that will need to be addressed by a human using Validation Station.