A Comprehensive Study on Text Detection and Extraction from Images and PDF Documents

Authors: Mayank Deshmukh, Saloni Rabde, Priyanka Makode, Sourabh Jasuja, Bhavesh Khasdev

Country: India

Full-text Research PDF File:   View   |   Download

Abstract: The growing need for digitization and intelligent document processing has led to significant advancements in text detection and extraction technologies. This paper reviews methodologies and tools employed for extracting textual information from images and Portable Document Format (PDF) files. Both traditional Optical Character Recognition (OCR) techniques and modern deep learning-based approaches are discussed. Five major research contributions in this area are analyzed in detail. The paper further explores challenges in handling complex document layouts, multilingual text, and low-quality images, and highlights research gaps and future directions that emphasize the potential of artificial intelligence and multimodal learning to enhance text extraction accuracy and efficiency.

Keywords: OCR, Text Extraction, Deep Learning, Layout LM, Scene Text Detection, Document Analysis.

Paper Id: 232796

Published On: 2025-11-07

Published In: Volume 13, Issue 6, November-December 2025

Share this