Scanning of structured documents

Challenge

Allow access to structured data (test results presented in tabular form) contained in images in a wide range of formats.

Resolution

An algorithmic solution for image pre-processing, Character Recognition (OCR) and post processing of text data was created in order to compile results in data frames. This solution will allow the Client to significantly lower the time and cost dedicated to extracting and classifying the data.

Techno

Python/Pandas/Tesseract

Facing the technical challenges of this project was a real kick!

Scanning of structured documents

Challenge

Resolution

Techno

Data Scientist

Louvain-la-Neuve or Brussels

Project Manager

Louvain-la-Neuve or Brussels