milwright/historical-ocr

Historical OCR

An advanced OCR application for historical document analysis using Mistral AI.

Note: This tool is designed to assist scholars in historical research by extracting text from challenging documents. While it may not achieve 100% accuracy for all materials, it serves as a valuable research aid for navigating historical documents, particularly historical newspapers, handwritten documents, and photos of archival materials.

Features

• —OCR with Context: AI-enhanced OCR optimized for historical documents
• —Document Type Detection: Automatically identifies handwritten letters, recipes, scientific texts, and more
• —Advanced Image Preprocessing:
• —Automatic deskewing to correct document orientation
• —Smart thresholding with Otsu and adaptive methods
• —Morphological operations to clean up text
• —Document-type specific optimization
• —Custom Prompting: Tailor the AI analysis with document-specific instructions
• —Structured Output: Returns organized, structured information based on document type

Using This App

1. 1.Upload a historical document (image or PDF)
2. 2.Add optional context or special instructions
3. 3.Get detailed, structured OCR results with historical context

Supported Document Types

• —Handwritten letters and correspondence
• —Historical recipes and cookbooks
• —Travel accounts and exploration logs
• —Scientific papers and experiments
• —Legal documents and certificates
• —Historical newspaper articles
• —General historical texts

Technical Details

Built with Streamlit and Mistral AI's OCR and large language model capabilities.

Created by Zach Muhlbauer, CUNY Graduate Center