I have 3 tables (image pasted) all 3 table(have same columns) look same and i want data of address column (yellow colour) of 3 tables stored inside a variable.

There are different ways to handle extraction of tables from pdf. The final solution will depend primarily on individual pdf that you need to read. Some variables to think about when choosing the solution are:

• is the pdf just an image saved as pdf (rastered image of a scanned document)?
• what is the quality of pdf?
• is there any noise in the pdf files (e.g. spots caused by a printer) you need to get rid of?
• is the table in pdf skewed?
• how many pages has a pdf?
• how many pages a table spans across?
• how many documents do you need to scan?

There are many solutions to extract tables from pdf ranging from table-specialized OCR services to python utility libraries to help you build your own extraction program.

An example of a powerful tool to convert data from tables from pdf to excel is Camelot, which you have included in your question's tags. It abstracts a lot of complexity involved in the task at hand. You just install it and access it for example like that:

import camelot
file = 'https://www.w3.org/WAI/WCAG21/working-examples/pdf-table/table.pdf'
tables = camelot.read_pdf(file)
tables[0].to_excel('table.xlsx')

As I mentioned, the devil lies in the individual characteristics of a table and a pdf file.