Text region detection in historical astronomical diagrams

2026-06-14

AI Summary

This study addresses the significant challenges in detecting text regions within historical astronomical charts, a task hindered by the absence of dedicated datasets and benchmark methods, particularly in multilingual and cross-century visual contexts. To bridge this gap, the authors introduce a novel dataset comprising 948 charts spanning the 8th to 18th centuries across seven linguistic traditions, annotated with 10,940 precise, reading-order-aware polygonal labels and enriched with 20 fine-grained typographic categories for Latin-script regions. They propose Poly-DETR, an extension of DINO-DETR capable of predicting ordered vertices, and evaluate it alongside TEXTSTR and DeepSolo++. Poly-DETR achieves state-of-the-art performance on both MTHv2 and cBAD2019 benchmarks and demonstrates superior text detection accuracy on the newly introduced dataset.

Abstract

Text detection is a crucial task in the analysis of historical documents. While datasets and benchmarks exist for text detection in manuscripts and maps, the study of text in mathematical diagrams has received little attention. To address this, we introduce a large-scale, diverse, open-access dataset of 948 historical astronomical diagrams containing 10,940 oriented polygonal text regions. Our dataset spans ten centuries (8th to 18th) and seven main linguistic traditions: Arabic and Persian (115), Chinese (332), Byzantine (233), Latin (185), Hebrew (48), and Sanskrit (35). It captures a wide range of diagram styles and textual content, from symbols to multi-line paragraphs. Each text instance is annotated with ordered polygons that precisely delineate text regions and encode the reading direction. In addition, we annotated the 2,293 regions in Latin diagrams with 20 class labels. We evaluated several strong baselines on our dataset, including TESTR, DeepSolo++, and Poly-DETR, a simple extension of DINO-DETR that we design to predict ordered polygon vertices. Poly-DETR achieves state-of-the-art performance on the MTHv2 and cBAD2019 benchmarks and provides a solid, simple baseline on our dataset. Code and dataset available online.

Problem

• text region detection
• historical astronomical diagrams
• mathematical diagrams
• document analysis
• text annotation

Innovation

• historical astronomical diagrams
• oriented polygonal text detection
• Poly-DETR
• multilingual document analysis
• reading direction annotation