Azure Document Intelligence Sample

This project demonstrates the usage of Azure AI Document Intelligence (formerly Form Recognizer) to analyze and extract information from documents using Python.

Features

• Document layout analysis
• Handwritten text detection
• Text content extraction
• Table analysis
• Selection marks (checkboxes, radio buttons) detection

Prerequisites

• Python 3.8 or later
• Azure Document Intelligence resource
• Required packages:
  • azure-ai-documentintelligence
  • azure-core

Installation

1. Clone this repository:

git clone https://github.com/lphan3/Doc.int.AI.git
cd Doc.int.AI

2. Create and activate a virtual environment (recommended):

# Windows
python -m venv venv
.\venv\Scripts\activate

3. Install the required packages:
```bash
# Install all required packages
pip install azure-ai-documentintelligence azure-core

# Verify installation
pip list | findstr azure

4. Configure your Azure credentials:

   • Go to Azure Portal
   • Create or select your Document Intelligence resource
   • Copy the endpoint URL and API key
   • Update Starting.py with your credentials:

     endpoint = "YOUR_ENDPOINT_HERE"  # e.g., "https://your-resource.cognitiveservices.azure.com/"
     key = "YOUR_KEY_HERE"  # Your Azure Document Intelligence API key

5. Test the installation:

python Starting.py

Configuration

Before running the code, you need to configure your Azure Document Intelligence credentials:

1. Get your endpoint URL and API key from the Azure Portal
2. Update the following variables in Starting.py:
   • endpoint: Your Azure Document Intelligence endpoint URL
   • key: Your Azure Document Intelligence API key

Usage

The sample code demonstrates several features:

1. Document Analysis

   • Uses the "prebuilt-layout" model for general document analysis
   • Supports URL-based document analysis
   • Sample document URL provided for testing

2. Style Analysis

   • Detects handwritten content in documents
   • Processes document styles

3. Page Content Analysis

   • Extracts text lines from each page
   • Processes selection marks (checkboxes, radio buttons)
   • Handles UTF-8 encoded content

4. Table Analysis

   • Extracts table structure
   • Processes table cells and their content
   • Provides row and column information

Sample Output

The code will output:

• Document style information (handwritten content detection)
• Text content from each line
• Selection mark states and confidence levels
• Table dimensions and cell contents

Code Structure

The code is organized into clear sections:

1. Imports and Dependencies
2. Azure Configuration
3. Sample Document Setup
4. Client Initialization
5. Document Analysis
6. Style Analysis
7. Page Content Analysis
8. Table Analysis

Resources

• Azure Document Intelligence Documentation
• Azure Portal

Note

This is a sample implementation for presentation purposes. For production use, ensure proper error handling and security measures are in place.

Excel Integration Options

For Excel interaction, you can use the following packages:

This is base on my research, there're more out there

1. pandas

• Most popular and powerful data manipulation library
• Excellent for Excel file handling and lookup operations
• Installation: pip install pandas openpyxl
• Features:
  • Read/write Excel files
  • VLOOKUP-like operations using merge()
  • Data filtering and transformation
  • Large dataset handling

2. openpyxl

• Direct Excel file manipulation
• Installation: pip install openpyxl
• Features:
  • Read/write Excel files
  • Cell formatting
  • Formula support
  • Charts and images

3. xlwings

• Excel automation with Python
• Installation: pip install xlwings
• Features:
  • Interact with Excel application
  • Run Excel macros
  • Real-time Excel updates
  • Excel formula execution

4. xlrd and xlwt

• Lightweight Excel file handling
• Installation: pip install xlrd xlwt
• Features:
  • Read Excel files (xlrd)
  • Write Excel files (xlwt)
  • Basic Excel operations

Recommended Setup

For most use cases, we recommend using pandas with openpyxl:

pip install pandas openpyxl

Example usage with pandas for lookup operations:

import pandas as pd

# Read Excel files
df1 = pd.read_excel('source.xlsx')
df2 = pd.read_excel('lookup.xlsx')

# Perform VLOOKUP-like operation
result = pd.merge(df1, df2, on='key_column', how='left')