OCR COMPTA (new Version) : Extraction of invoice data with IA & Generic Patterns

MBH OCR is an AI-powered intelligent document processing solution that combines OCR, Generative AI, and LLM technologies to automatically extract, understand, classify, and process accounting data from scanned documents and invoices. The platform evolved from a traditional rule-based OCR system into a smarter AI-driven solution, improving data extraction accuracy, reducing manual processing, and streamlining accounting workflows.

Présentation du projet IA

OCR COMPTA – This web application dedicated to accountants aims to save time in processing invoice binders. It permits :

Customers

This solution has been saled to several clients in Tunisia, Switzerland and France

Points on technologies​

For the new version of MBH OCR, I worked with a modern, AI-powered technology stack combining backend development, document processing, messaging, storage, databases, and DevOps tools.

  • Java 21

  • Spring Boot 3.5

  • Apache Kafka

  • Generative AI

  • LLM

  • PaddleOCR

  • Python

  • PostgreSQL

  • pgAdmin

  • MinIO

  • Docker

  • VPS

  • REST APIs

  • JetBrains IDEs with AI Assistant

  • Git

Video Presentation

Clone to the GIT

https://gitlab.com/java-jproject/gold-gps-backend.git

Get a quote on a project :

en_USEnglish