Data Labelling for Comparing Pictures

Description

We have 100 PDF documents from which we have extract two things
1. An image of some tabular data
2. An excel spreadsheet, based on that tabular data.
The labelling task will be to compare these images against excel spreadsheets.
We want to know
1. Was the image converted to a non-empty table
2. Were the rows detected ? How many are missing?
3. Were the columns detected ? How many are missing?
4. Was the numerical information extracted ?