Research Article
Automated Document Processing: Combining OCR and Generative AI for Efficient Text Extraction and Summarization
@INPROCEEDINGS{10.4108/eai.28-4-2025.2357770, author={C Shyamala Kumari and V. Yeswanth Gupta and G. Manikanta and B. V. Abhishikth}, title={Automated Document Processing: Combining OCR and Generative AI for Efficient Text Extraction and Summarization}, proceedings={Proceedings of the 4th International Conference on Information Technology, Civil Innovation, Science, and Management, ICITSM 2025, 28-29 April 2025, Tiruchengode, Tamil Nadu, India, Part I}, publisher={EAI}, proceedings_a={ICITSM PART I}, year={2025}, month={10}, keywords={optical character recognition (ocr) generative artificial intelligence document processing text summarization machine learning ai summarization document workflow automation}, doi={10.4108/eai.28-4-2025.2357770} }- C Shyamala Kumari
V. Yeswanth Gupta
G. Manikanta
B. V. Abhishikth
Year: 2025
Automated Document Processing: Combining OCR and Generative AI for Efficient Text Extraction and Summarization
ICITSM PART I
EAI
DOI: 10.4108/eai.28-4-2025.2357770
Abstract
The implementation of new digital document creation media formats throughout technological development requires improved processing techniques and methods to be developed. Traditional OCR tools remain helpful but they cannot successfully process complicated documents or deteriorated scan copies. The present manual summarization technique proves both slow and error-prone when companies work to compile textual content from various sources. An integrated solution for text extraction and summarization will be developed by applying enhanced OCR alongside Google Gemini’s generative AI technology. Advanced OCR software enables the system to better detect text patterns in incomplete or difficult-to-read document images. The implementation of generative AI models enables clinicians to produce brief yet appropriate summaries and enhances their ability to locate and handle documents. This method has proven its effectiveness by conducting tests on different documents which showed improved accuracy along with higher utility compared to conventional approaches.
