smart-ocr

Extract text and spatial data from images and scanned documents using PaddleOCR.

3|Updated Apr 2, 2026

Search Tags:#ocr #extraction #image processing #multilingual #paddleocr #digitization

Ready to run

npx skills add https://github.com/legout/pi-config --skill smart-ocr-legout

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: smart-ocr
Source: https://github.com/legout/pi-config/tree/main/installed-skills/smart-ocr
Command: npx skills add https://github.com/legout/pi-config --skill smart-ocr-legout

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires paddlepaddle, paddleocr, pdf2image, Pillow, requests, tqdm.

What problem does it solve?

This skill solves the challenge of digitizing information trapped in non-text formats like photos, scanned PDFs, and handwritten notes by converting them into structured, machine-readable text.

Core Features & Use Cases

Multilingual OCR: Supports over 100 languages, including English, Chinese, Japanese, and more, using the powerful PaddleOCR engine.
Layout Reconstruction: Intelligently groups extracted text into logical lines and paragraphs to maintain document structure.
Use Case: Quickly process a stack of business cards or receipts by extracting contact details or line items directly into a structured JSON format for your database.

Quick Start

Use the smart-ocr skill to extract all text from the provided image file and return it as a structured list.

Frequently Asked Questions about smart-ocr

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I extract text from a scanned PDF or image?▼

To extract text from a scanned PDF or image, the smart-ocr Skill uses PaddleOCR to perform high-accuracy optical character recognition, converting non-text formats like photos and receipts into structured, machine-readable text.

Does PaddleOCR support multilingual text extraction for documents?▼

PaddleOCR supports multilingual text extraction, processing over 100 languages including English, Chinese, and Japanese to digitize diverse document types into structured spatial data.

What's the best way to digitize business cards into structured JSON?▼

To digitize business cards into structured JSON, use the smart-ocr Skill to extract contact details and intelligently group text into logical lines and paragraphs, maintaining the original document layout.

Do I need to install PaddlePaddle and pdf2image to process scanned documents?▼

Yes, you need to install PaddlePaddle, PaddleOCR, pdf2image, and Pillow, as these dependencies are required to execute text detection and complex layout reconstruction tasks.

How does layout reconstruction work when extracting text from images?▼

Layout reconstruction works by intelligently grouping extracted text into logical lines and paragraphs, preserving the spatial data and document structure of complex formats like receipts and PDFs.