🤖 AI Summary
This study addresses authorship identification in historical Arabic manuscripts by proposing a convolutional neural network integrated with an attention mechanism. The model performs closed-set author classification under two evaluation protocols: line-level and page-independent settings. Notably, it innovatively treats rare manuscript lines containing contributions from two authors as composite categories. Leveraging a substantially expanded and manually verified version of the Muharaf dataset, this work establishes the first baseline results for both protocols. Extensive experiments employing backbone architectures such as DenseNet201 and diverse training strategies—validated through fourteen ablation configurations—achieve a Top-1 accuracy of 99.05% (F1: 97.44%) under the line-level protocol and 78.61% (F1: 66.55%) under the page-independent protocol, demonstrating significant performance gains in manuscript authorship identification.
📝 Abstract
Handwritten Arabic manuscripts preserve the Arab world's intellectual and cultural heritage, and writer identification supports provenance, authenticity verification, and historical analysis. Using the Muharaf dataset of historical Arabic manuscripts, we evaluate writer identification from individual line images and, to the best of our knowledge, provide the first baselines reported under both line-level and page-disjoint evaluation protocols. Since the dataset is only partially labeled for writer identification, we manually verified and expanded writer labels in the public portion from 6,858 (28.00%) to 21,249 lines (86.75%) out of 24,495 line images, correcting inconsistencies and removing non-handwritten text. After further filtering, we retained 18,987 lines (77.51%). We propose a Convolutional Neural Network (CNN)-based model with attention mechanisms for closed-set writer identification, including rare two-writer lines modeled as composite writer-pair classes. We benchmark fourteen configurations and conduct ablations across different feature extractors and training regimes. To assess generalization to unseen pages, the page-disjoint protocol assigns all lines from each page to a single split. Under the line-level protocol, a fine-tuned DenseNet201 with attention achieves 99.05% Top-1 accuracy, 99.73% Top-5 accuracy, and 97.44% F1-score. Under the more challenging page-disjoint protocol, the best observed results are 78.61% Top-1 accuracy, 87.79% Top-5 accuracy, and 66.55% F1-score, thus quantifying the impact of page-level cues. By expanding the Muharaf dataset's labeled subset and reporting both protocols, we provide a clearer benchmark and a practical resource for historians and linguists engaged with culturally and historically significant documents. The code and implementation details are available on GitHub.