AI based signage classification for linguistic landscape studies

2025-10-26

AI Summary

Traditional language landscape research relies on manual data collection and annotation, resulting in low efficiency and poor scalability. This study proposes a human–AI collaborative hybrid analytical framework, implemented in Honolulu’s Chinatown, integrating optical character recognition (OCR), multilingual language classification models, and georeferenced image data—augmented by systematic human verification. Results show an overall text recognition accuracy of 79% and identify five recurrent OCR error patterns—including background text interference and failures induced by font variability or suboptimal lighting—thereby elucidating AI behavior and limitations in complex real-world environments. Crucially, this framework pioneers the integration of interpretable, human-led validation directly into AI-driven language landscape analysis. It not only demonstrates the feasibility of large-scale automated analysis but also advances the field toward intelligent, reproducible, and methodologically transparent scholarship.

Technology Category

• Natural Language Processing: Language Grounding & Multi-modal NLP
• Computer Vision: Language and Vision
• Humans and AI: Crowd Sourcing and Human Computation

Application Category

• Search and Retrieval-Augmented AI: Multilingual and cross-lingual Web search
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• Web Mining and Content Analysis: Mining multimedia, multimodal, multilingual, cross-lingual Web data

Abstract

Linguistic Landscape (LL) research traditionally relies on manual photography and annotation of public signages to examine distribution of languages in urban space. While such methods yield valuable findings, the process is time-consuming and difficult for large study areas. This study explores the use of AI powered language detection method to automate LL analysis. Using Honolulu Chinatown as a case study, we constructed a georeferenced photo dataset of 1,449 images collected by researchers and applied AI for optical character recognition (OCR) and language classification. We also conducted manual validations for accuracy checking. This model achieved an overall accuracy of 79%. Five recurring types of mislabeling were identified, including distortion, reflection, degraded surface, graffiti, and hallucination. The analysis also reveals that the AI model treats all regions of an image equally, detecting peripheral or background texts that human interpreters typically ignore. Despite these limitations, the results demonstrate the potential of integrating AI-assisted workflows into LL research to reduce such time-consuming processes. However, due to all the limitations and mis-labels, we recognize that AI cannot be fully trusted during this process. This paper encourages a hybrid approach combining AI automation with human validation for a more reliable and efficient workflow.

Problem

• Automating linguistic landscape analysis using AI language detection methods
• Addressing time-consuming manual signage classification in urban studies
• Identifying AI limitations in text recognition for reliable landscape research

Innovation

• AI automates linguistic landscape analysis with OCR
• Hybrid approach combines AI automation with human validation
• Georeferenced photo dataset enables language classification