Automatically extract invoice data from scanned PDF or image files and generate Sales or Purchase Invoices in ERPNext using Tesseract OCR.

🚀 Features

• 📄 Parses invoice number, date, line items, and total
• 🧾 Creates:
• ✅ Sales Invoice (for Customer)
• ✅ Purchase Invoice (for Supplier)
• 📦 Automatically creates Items if not found
• 🔄 Dynamic Party Link (Customer or Supplier)
• 📂 JSON preview of extracted data for audit

📁 Doctypes

Invoice Upload

⚙️ Full Installation Guide

✅ 1. Prerequisites

Install required system packages:

sudo apt update

sudo apt install -y poppler-utils tesseract-ocr

cd ~/frappe-bench/apps

git clone https://github.com/Mohtashim-1/Invoice-OCR.git

Activate your Frappe virtual environment

source ~/frappe-bench/env/bin/activate

Install required Python libraries

pip install -r invoice_ocr/requirements.txt

Or manually:

pip install pytesseract pdf2image Pillow PyPDF2

4. Install the app on your site

cd ~/frappe-bench

bench new-site invoice_ocr

bench --site invoice_ocr install-app invoice_ocr

bench migrate

bench restart