You can use tessaract-python to convert image into text and then use pandas to create the excel & fill the cells, as per your requirement.
It is useful (but not required) to create a new .XML file for each project using Word 2007/Word 2010. If the .PDF was saved from a .doc file that you can access then copy the table from the Word file into an .XML spreadsheet. Sometimes, a .pdf is saved from a .doc and the text in the .pdf is selectable. Then you don't need the .doc because you can select text from the .pdf file. However, some .pdf files are just scanned bitmap (.bmp) files that are saved as a .pdf file. In that case, use text recognition software with optical character recognition (OCR) to pull text from the .pdf file. Then paste it into a .doc and format it into a table and then copy/paste into the .XML file. It is useful (but not required) to create a new .XML file for each project using Word.