This study is specifically designed to model diacritical variations and long‐range spatial dependencies within isolated Urdu characters through a Feature Shift Attention (FSA) architecture that integrates a three‐block convolutional frontend for fine‐grained local stroke and diacritical dot extraction with a four‐stage FSA backend employing shifted window multi‐head self‐attention across the full character image.
Abstract
Handwritten Urdu script recognition remains one of the most demanding challenges in Optical Character Recognition (OCR) due to the fully cursive nature of the script, context‐sensitive allographic character forms, overlapping ligature structures, and semantically discriminative diacritical marks. Existing OCR systems, predominantly designed for Latin‐based non‐cursive scripts, fail to generalize effectively to Urdu handwriting under diverse writers, document conditions, and image quality variations. The proposed study is specifically designed to model diacritical variations and long‐range spatial dependencies within isolated Urdu characters through a Feature Shift Attention (FSA) architecture that integrates a three‐block convolutional frontend for fine‐grained local stroke and diacritical dot extraction with a four‐stage FSA backend employing shifted window multi‐head self‐attention across the full character image. A validated and class‐balanced training corpus of approximately 113,800 samples is aggregated from the UCOM, CENPARMI, and UNHD datasets and expanded to 400,000 samples through class‐conditional geometric and photometric augmentation. A two‐phase staged fine‐tuning protocol using the AdamW optimizer with cosine‐annealed scheduling has been employed to adapt pre‐trained FSA representations to the Urdu script domain without catastrophic forgetting. When evaluated against CNN‐only, SVM, and KNN baselines on a stratified blind held‐out test set spanning all 39 Urdu base character classes, the proposed system achieved a macro‐averaged accuracy of 96.3%, precision of 95.8%, recall of 96.1%, and F1 score of 95.9%, outperforming all competing systems and enabling robust discrimination between visually similar characters differentiated by dot patterns.
The automatic assessment of handwritten answers in low‐resource Indic scripts is a difficult problem because of the complexity of the orthographic ligatures, the variability of the writers, limited availability of annotated resources, and the error propagation from handwritten text recognition (HTR) to the grading mo...
Padmavathi Pragada, D. Chamala· International Conference on...· 0 citations
Offline handwritten text recognition (HTR) remains one of the most challenging problems in document image analysis because of the enormous variability of individual handwriting styles, inconsistent stroke widths, irregular slant and the scarcity of large annotated corpora. Data augmentation has become the de-facto stra...
Lalita Kumari· International journal on eme...· 0 citations
Arabic handwriting recognition (AHR) remains challenging due to the context-dependent, cursive script that’s affected by diacritic marks, overlapping strokes, and high writer variability. This review presents an overview of deep learning approaches for AHR in both online and offline settings. The literature search was...
Enas Munther Altai, Ali Mohammed Sahan· Iraqi Journal for Computers...· 0 citations
The results point towards the potential feasibility of writer adaptation for the HMER task through a writer-adaptive fine-tuning pipeline to MFH-CoMER, with the model struggling for handwriting articles that the base model can already evaluate well.
P. Silva, Lorenz Bernard Marqueses, J. Ilao· arXiv.org· 0 citations
Handwritten mathematical expression recognition (HMER) focuses on automatically generating LaTeX sequences from provided images. Currently, attention-based encoder-decoder models are extensively adopted for this task. However, due to the ever-changing nature of handwriting styles, existing decoding methods, which merel...
Jie-Rui Huang, Cheng-Xu Liu, Yan-Ji Song et al.· Proceedings of the 34th ACM...· 0 citations
Recognizing handwritten medical prescriptions is a challenging problem due to the diverse styles of handwriting, similar character shapes, image noise and the absence of annotated data. To overcome these problems, two major changes are suggested: in the paper, a transfer learning-based CNN-BiLSTM network is introduc...
N. A, A. S· Computer and Decision Making...· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.