DocTR

docTR 1.1.0 is an open-source deep learning library for optical character recognition (OCR), designed to detect and recognize text in images and scanned documents. Built with Python and PyTorch, docTR provides pretrained models and configurable OCR pipelines for automated document text extraction.

docTR can be used by developers, data-processing teams, document-management applications, intelligent document-processing workflows, and AI solutions that need to extract text from invoices, forms, reports, scanned pages, photographs, and other image-based documents.

Features:

• Text detection and recognition capabilities
• Pretrained deep learning OCR models
• Python-based API for application integration
• Support for images and document-processing workflows
• PyTorch-based deep learning backend
• Configurable OCR model architectures
• Suitable for automated document and data extraction

Usage Instructions:

#Activate virtual environment: 
$ sudo su
$ cd /opt/doctr
$ source doctr-venv/bin/activate

#Verify the version of DocTR: $ pip show python-doctr

Disclaimer: This marketplace image provides docTR and its required open-source dependencies for deployment on Ubuntu. Model accuracy and processing performance depend on the input documents, selected models, available CPU/GPU resources, and application configuration. Users are responsible for validating extracted information before using it in production or business-critical workflows.