Python ML Image Recognition Expert Needed

I'm searching for a skilled Python developer with a solid background in machine learning and image recognition to work on a project to help me write a python code that can detect the questions on a scanned image and crop them into different images. You can use any library or package.


Key Tasks:
- Object Detection: I require the developer to implement object detection capabilities within my image recognition system.
- Text Blocks Detection: I need a method to accurately detect and interpret text blocks within images.

The ideal candidate should have:
- A strong background in Machine Learning, especially with Python.
- Extensive experience in object detection and image classification.
- Expertise in text blocks detection would be a great advantage.

The data available for this project includes some scanned exam files.

The output I need requires cropped images from the required coordinates so that each different question is saved into different folder, and the name of the files are given as the number of questions. For example: "Math_14", "Math_15". So the most important text that is required to be accurately read is the number/order of questions.

Very close example of what I've tried so far is the following. It did not work well with the examples I used but the methodology should be something similar:
https://stackoverflow.com/questions/60223779/find-question-text-block-in-image-with-python-opencv

I am also attaching some pdf files as I've mentioned.

I also have a draft code which does not work as accurate but I can share it with you if needed.
I believe that the code to be written should not take more than 5 hours of an experienced python developer but please let me know if you can do that within a longer period.

Thank you in advance!