Arabic Handwritten Document OCR Solution with Binarization and Adaptive Scale Fusion Detection

TL;DR

This paper presents a comprehensive OCR system for Arabic handwritten text, combining advanced line segmentation and recognition techniques, achieving high accuracy and setting new benchmarks in the field.

Contribution

The authors introduce a novel OCR pipeline with improved line segmentation and recognition methods tailored for Arabic handwriting, trained on a new dataset.

Findings

CRR of 99.20% on character recognition

WRR of 93.75% on single words

CRR of 83.76% on sentences

Abstract

The problem of converting images of text into plain text is a widely researched topic in both academia and industry. Arabic handwritten Text Recognation (AHTR) poses additional challenges due to diverse handwriting styles and limited labeled data. In this paper we present a complete OCR pipeline that starts with line segmentation using Differentiable Binarization and Adaptive Scale Fusion techniques to ensure accurate detection of text lines. Following segmentation, a CNN-BiLSTM-CTC architecture is applied to recognize characters. Our system, trained on the Arabic Multi-Fonts Dataset (AMFDS), achieves a Character Recognition Rate (CRR) of 99.20% and a Word Recognition Rate (WRR) of 93.75% on single-word samples containing 7 to 10 characters, along with a CRR of 83.76% for sentences. These results demonstrate the system's strong performance in handling Arabic scripts, establishing a new benchmark for AHTR systems.