Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

From Press to Pixels: Evolving Urdu Text Recognition

Samee Arif, Sualeha Farid

Abstract

This paper presents a comparative analysis of Large Language Models (LLMs) and traditional Optical Character Recognition (OCR) systems on Urdu newspapers, addressing challenges posed by complex multi-column layouts, low-resolution scans, and the stylistic variability of the Nastaliq script. To handle these challenges, we fine-tune YOLOv11x models for article- and column-level text block extraction and train a SwinIR-based super-resolution module that enhances image quality for downstream text recognition, improving accuracy by an average of 50%. We further introduce the Urdu Newspaper Benchmark (UNB), a manually annotated dataset for Urdu OCR comprising 829 paragraph images with a total of 9,982 sentences. Using UNB and the OpenITI corpus, we conduct a systematic comparison between traditional CNN+RNN-based OCR systems and modern LLMs, presenting detailed insertion, deletion, and substitution error analyses alongside character-level confusion patterns. We find that Gemini-2.5-Pro achieves the best performance on UNB (WER 0.133), while fine-tuning GPT-4o on just 500 in-domain samples yields a 6.13% absolute WER improvement, demonstrating the adaptability of LLMs to low-resource, morphologically complex scripts like Urdu. The UNB dataset and fine-tuned models are publicly available at https://github.com/sameearif/urdu-newspaper-benchmark.

From Press to Pixels: Evolving Urdu Text Recognition

Abstract

This paper presents a comparative analysis of Large Language Models (LLMs) and traditional Optical Character Recognition (OCR) systems on Urdu newspapers, addressing challenges posed by complex multi-column layouts, low-resolution scans, and the stylistic variability of the Nastaliq script. To handle these challenges, we fine-tune YOLOv11x models for article- and column-level text block extraction and train a SwinIR-based super-resolution module that enhances image quality for downstream text recognition, improving accuracy by an average of 50%. We further introduce the Urdu Newspaper Benchmark (UNB), a manually annotated dataset for Urdu OCR comprising 829 paragraph images with a total of 9,982 sentences. Using UNB and the OpenITI corpus, we conduct a systematic comparison between traditional CNN+RNN-based OCR systems and modern LLMs, presenting detailed insertion, deletion, and substitution error analyses alongside character-level confusion patterns. We find that Gemini-2.5-Pro achieves the best performance on UNB (WER 0.133), while fine-tuning GPT-4o on just 500 in-domain samples yields a 6.13% absolute WER improvement, demonstrating the adaptability of LLMs to low-resource, morphologically complex scripts like Urdu. The UNB dataset and fine-tuned models are publicly available at https://github.com/sameearif/urdu-newspaper-benchmark.

Paper Structure

This paper contains 32 sections, 8 figures, 5 tables.

Table of Contents

  1. Introduction
  2. Related Work
  3. Open-source OCR Engines
  4. Prior Work in Urdu OCR
  5. Recent Advances in Multilingual OCR
  6. Urdu Text Recognition Datasets
  7. Methodology
  8. Data Collection and Preparation
  9. Text Recognition Datasets.
  10. Segmentation Datasets.
  11. OCR Architecture
  12. Experimental Setup
  13. Image Segmentation.
  14. Image Super-Resolution.
  15. Text Recognition.
  16. ...and 17 more sections

Figures (8)

  • Figure 1: Example of Naskh (top) and Nastaliq (bottom).
  • Figure 2: OCR pipeline with article and column segmentation, super-resolution, and LLM-based text recognition.
  • Figure 3: The sequence of text extracted by the LLMs from multi-column image.
  • Figure : (a) Input
  • Figure : (a) Input
  • ...and 3 more figures