Share
ReportHello everyone!
I have the following case.
I have a variety of unstructured documents in the form of pdf, containing scanned images or photos of tables of data. The tables contain similar data. For example each table will contain specific data corresponding to diesel engine (Pressures, temperatures etc), but not in the same position or similar length or table size and with minor title variations. So far I have already identify over 100 different types of documents, in my database.
Here are some table examples.
As you can see the tables are
- Not the same size, position and elements can be scattered through out the page.
- Each page is further divided into sub tables.
The scope is to use a trained model in a power automate flow, capable of identifying the data from such type of documents.
Thank you in advance for your time and effort.
Categories:
I have the same question (0)
All responses (
Answers (
997
