🔍 OCR Module Guide

This page is currently only available in Chinese. Please switch to 简体中文 for the full content.

🔍 OCR Module Guide

The OCR module calls the Baidu AI Cloud API and provides features like VAT invoice recognition → Excel.

Core Scenarios

Recognize a Single Invoice

import office
office.ocr.VatInvoiceOCR2Excel(
    input_path='./invoice_001.jpg',
    output_path='./output/'
)

Batch-Recognize a Folder

office.ocr.VatInvoiceOCR2Excel(
    input_path='./all_invoices/',
    output_path='./output/',
    output_excel='monthly_invoice_summary.xlsx',
    file_name=True
)

Recognize Image from URL

office.ocr.VatInvoiceOCR2Excel(
    img_url='https://example.com/invoice.jpg',
    output_path='./output/'
)

Recognition Result

The Excel automatically includes the following fields:

• Invoice code, invoice number, issue date
• Seller / buyer info (name, tax ID)
• Amount, tax amount, tax-exclusive amount, tax rate

Configuring Baidu OCR API

1. Visit Baidu AI Cloud and register an account
2. Create a "Text Recognition OCR" application
3. Get the API Key and Secret Key
4. Configure:

office.ocr.VatInvoiceOCR2Excel(
    input_path='./invoices/',
    id='your_api_id',
    key='your_api_secret'
)

Full API see OCR API Reference