Django Document Recognition App

Overview

This AI-powered solution is designed to streamline the recognition and processing of documents. It supports various image formats, including PNG and JPEG, and offers robust features such as customizable field management and the ability to export and import field configurations. The application processes documents and outputs the extracted data in a structured JSON format, ensuring seamless workflows and easy integration with other systems, even when reloading or transferring data between sessions.

Main Tech Stacks

• Python;
• OpenAI Python library;
• Django;
• Tailwind CSS;
• Vanilla Javascript

Features

• Document Recognition: Automatically extracts and processes data from supported document types.
• Field Management: Customize and manage fields for document data extraction.

Usage

Getting Started

1. Configure fields as needed;
2. Select the file you want to process (ensure it meets the supported file type and size requirements);
3. Click on "Submit";
4. After a few seconds, the JSON response will appear on the bottom of the page.

Supported File Types

The application supports the following file types for recognition:

• Images (.jpg, .jpeg, .png);

File Size Limitations

• Maximum file size: 10 MB per document.

Fields

The application uses fields to extract and organize data from documents. Below are the key fields and their purposes:

• Name: The name/title of the field;
• Description: A brief explanation or context about the field;
• Type: The expected data type from the field. It supports some fields of OpenAI Structured Outputs
• Array Type: If the type is Array, you must specify the data type of the items;

Export/Import Feature

Export

• The export feature allows you to save processed document data into a JSON file.
• This is useful for backing up data or transferring it to another system.

Import

• The import feature enables you to load previously saved JSON files into the application.
• This allows you to restore data or continue working on previously processed documents.

Steps:

1. Export:
   • After processing a document, click the "Export" button.
   • Choose a location to save the JSON file.
2. Import:
   • Click the "Import" button.
   • Select a JSON file containing previously saved data.

Development

1. Create a .env file and save the OpenAI API key on it:

OPENAI_API_KEY=sk-proj-encrypted...

1. If you need to update Tailwind classes, run the watch mode:

python manage.py tailwind start

2. Run the server:

python manage.py runserver

3. http://localhost:8000

Contributing

Feel free to contribute to this project by submitting issues or pull requests.