An Efficient Deep Learning-Based Approach to Automating Invoice Document Validation

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Amari, Aziz, Makni, Mariem, Fnaich, Wissal, Lahmar, Akram, Koubaa, Fedi, Charrad, Oumayma, Zormati, Mohamed Ali, Douss, Rabaa Youssef
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866913738349608960
author Amari, Aziz
Makni, Mariem
Fnaich, Wissal
Lahmar, Akram
Koubaa, Fedi
Charrad, Oumayma
Zormati, Mohamed Ali
Douss, Rabaa Youssef
author_facet Amari, Aziz
Makni, Mariem
Fnaich, Wissal
Lahmar, Akram
Koubaa, Fedi
Charrad, Oumayma
Zormati, Mohamed Ali
Douss, Rabaa Youssef
contents In large organizations, the number of financial transactions can grow rapidly, driving the need for fast and accurate multi-criteria invoice validation. Manual processing remains error-prone and time-consuming, while current automated solutions are limited by their inability to support a variety of constraints, such as documents that are partially handwritten or photographed with a mobile phone. In this paper, we propose to automate the validation of machine written invoices using document layout analysis and object detection techniques based on recent deep learning (DL) models. We introduce a novel dataset consisting of manually annotated real-world invoices and a multi-criteria validation process. We fine-tune and benchmark the most relevant DL models on our dataset. Experimental results show the effectiveness of the proposed pipeline and selected DL models in terms of achieving fast and accurate validation of invoices.
format Preprint
id arxiv_https___arxiv_org_abs_2503_12267
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle An Efficient Deep Learning-Based Approach to Automating Invoice Document Validation
Amari, Aziz
Makni, Mariem
Fnaich, Wissal
Lahmar, Akram
Koubaa, Fedi
Charrad, Oumayma
Zormati, Mohamed Ali
Douss, Rabaa Youssef
Computer Vision and Pattern Recognition
In large organizations, the number of financial transactions can grow rapidly, driving the need for fast and accurate multi-criteria invoice validation. Manual processing remains error-prone and time-consuming, while current automated solutions are limited by their inability to support a variety of constraints, such as documents that are partially handwritten or photographed with a mobile phone. In this paper, we propose to automate the validation of machine written invoices using document layout analysis and object detection techniques based on recent deep learning (DL) models. We introduce a novel dataset consisting of manually annotated real-world invoices and a multi-criteria validation process. We fine-tune and benchmark the most relevant DL models on our dataset. Experimental results show the effectiveness of the proposed pipeline and selected DL models in terms of achieving fast and accurate validation of invoices.
title An Efficient Deep Learning-Based Approach to Automating Invoice Document Validation
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2503.12267