Leveraging Transfer Learning and Mobile-enabled Convolutional Neural Networks for Improved Arabic Handwritten Character Recognition

2025-09-05

AI Summary

Addressing the dual challenges of high computational cost and scarce labeled data in Arabic handwritten character recognition, this paper proposes an efficient recognition framework integrating transfer learning with mobile-optimized lightweight CNNs. We systematically evaluate MobileNet, SqueezeNet, MnasNet, and ShuffleNet under three training paradigms—full fine-tuning, partial fine-tuning, and from-scratch training—across three benchmark datasets: AHCD, HIJJA, and IFHCDB. Results demonstrate that full fine-tuning achieves the best trade-off between accuracy and convergence speed. MnasNet attains 99% accuracy on IFHCDB (97% on AHCD; 92% on HIJJA), while MobileNet exhibits superior generalization across datasets. To the best of our knowledge, this is the first systematic study validating the synergistic efficacy of lightweight CNNs and transfer learning for low-resource Arabic handwriting recognition. The work establishes a reproducible, edge-deployable methodology, advancing practical deployment of Arabic OCR on resource-constrained devices.

Technology Category

• Machine Learning: Transfer, Domain Adaptation, Multi-Task Learning
• Natural Language Processing: Learning & Optimization for NLP
• Computer Vision: Learning & Optimization for CV

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

The study explores the integration of transfer learning (TL) with mobile-enabled convolutional neural networks (MbNets) to enhance Arabic Handwritten Character Recognition (AHCR). Addressing challenges like extensive computational requirements and dataset scarcity, this research evaluates three TL strategies--full fine-tuning, partial fine-tuning, and training from scratch--using four lightweight MbNets: MobileNet, SqueezeNet, MnasNet, and ShuffleNet. Experiments were conducted on three benchmark datasets: AHCD, HIJJA, and IFHCDB. MobileNet emerged as the top-performing model, consistently achieving superior accuracy, robustness, and efficiency, with ShuffleNet excelling in generalization, particularly under full fine-tuning. The IFHCDB dataset yielded the highest results, with 99% accuracy using MnasNet under full fine-tuning, highlighting its suitability for robust character recognition. The AHCD dataset achieved competitive accuracy (97%) with ShuffleNet, while HIJJA posed significant challenges due to its variability, achieving a peak accuracy of 92% with ShuffleNet. Notably, full fine-tuning demonstrated the best overall performance, balancing accuracy and convergence speed, while partial fine-tuning underperformed across metrics. These findings underscore the potential of combining TL and MbNets for resource-efficient AHCR, paving the way for further optimizations and broader applications. Future work will explore architectural modifications, in-depth dataset feature analysis, data augmentation, and advanced sensitivity analysis to enhance model robustness and generalizability.

Problem

• Enhancing Arabic handwritten character recognition accuracy
• Addressing computational demands and dataset scarcity issues
• Evaluating transfer learning with mobile-optimized neural networks

Innovation

• Transfer learning with mobile CNNs for Arabic recognition
• Evaluated full and partial fine-tuning strategies
• MobileNet achieved top accuracy and efficiency