PDF Table Extractor is an innovative Python project designed to tackle the challenge of extracting tables from scanned PDF documents. Leveraging advanced optical character recognition (OCR) and image processing techniques.

Extract tables precisely from PDFs and convert them to clean HTML for RAG pipelines, running fast on CPU without external dependencies.

PDF table extraction for RAG — convert to clean HTML. Fast, local, no GPU.

Trained a custom PaddleOCR model for handwritten text detection using a handwritten image dataset. The project focuses on improving accuracy across diverse handwriting styles through fine-tuning, preprocessing, and evaluation. Built with Python, PaddleOCR, OpenCV, and NumPy for efficient text recognition.