receipt-parser-ocr 1.1.1

A lightweight Python library to parse and categorize receipt data from OCR-extracted text

Receipt Parser OCR

A lightweight and extensible Python library designed to parse and categorize receipt data from OCR-extracted text.

This library is built to transform unstructured text (generated from OCR engines such as Tesseract) into structured financial data, making it ideal for applications involving expense tracking, financial analysis, and document processing.

Features

• OCR Text Parsing
  Extract structured information from raw OCR output (images or PDFs)

• Key Data Extraction
  Automatically identify:

  • Store name
  • Total amount
  • Payment mode (e.g., VISA, CASH)
  • Purchase date

• Automatic Categorization
  Classifies receipts into categories such as:

  • Groceries
  • Food
  • Travel
  • Shopping
  • Other

• Robust Text Cleaning
  Handles common OCR issues such as:

  • Encoding errors (€, £, ₹ symbols)
  • Noise and irregular formatting

• Extensible Design
  Easily extendable for:

  • New store-category mappings
  • Custom parsing logic
  • Integration with other systems

• Framework Independent
  Designed with zero dependency on Django or any web framework, allowing reuse across different Python applications.

Installation

pip install receipt-parser-ocr