Mapping language at scale: combining street-view imagery, geospatial analysis, and machine learning in the study of Indonesian urban landscapes

Aulia Cika Kartika

doi:10.67490/ijls.v1i2.895

Authors

Aulia Cika Kartika Universitas Negeri Yogyakarta Author

DOI:

https://doi.org/10.67490/ijls.v1i2.895

Keywords:

GeoAI, linguistic landscape, machine learning, spatial analysis, street-view imagery

Abstract

Background: Urban linguistic landscapes remain difficult to examine at metropolitan scale because conventional fieldwork offers interpretive depth but limited spatial coverage, while automated image analysis often overlooks linguistic and social context. Objective: This study aims to integrate street-view imagery, machine-assisted text recognition, geospatial analysis, and multimodal interpretation to map publicly visible languages across differentiated urban environments in Jakarta. Method: This study applies a spatially explicit multimethod design to ninety validated signs extracted from nine eligible public images, combining sign detection, OCR, language identification, human validation, corridor-based mapping, and social-semiotic coding. Results: Findings indicate that machine-readable visibility is uneven, with standardized regulatory and commercial signs more reliably detected than temporary, occluded, stylized, or mixed-script signs. Spatial patterns show Indonesian concentrated in regulatory corridors, English more visible in corporate and retail settings, and Chinese localized within Glodok’s commercial landscape. Implication: Institutional analysis further demonstrates that language hierarchy is produced through sign ownership, communicative function, placement, typography, colour, materiality, and logo positioning rather than frequency alone. Novelty: This study contributes a reproducible framework that connects computational accuracy, exposure-aware spatial comparison, and socially grounded interpretation while preventing linguistic visibility from being conflated with speaker population, language vitality, or causal urban influence within a transparent and human-validated urban GeoAI research architecture.

References

[1] F. Zhang et al., “Measuring human perceptions of a large-scale urban region using machine learning,” Landscape and Urban Planning, Dec. 2018, doi: 10.1016/j.landurbplan.2018.08.020.

[2] J. C. Rangel, E. Cruz, and M. Cazorla, “Automatic Understanding and Mapping of Regions in Cities Using Google Street View Images,” Applied Sciences, Mar. 2022, doi: 10.3390/app12062971.

[3] Y. Liu et al., “An interpretable machine learning framework for measuring urban perceptions from panoramic street view images,” iScience, vol. 26, Feb. 2023, doi: 10.1016/j.isci.2023.106132.

[4] Z. Wang, H. Li, and R. Rajagopal, “Urban2Vec: Incorporating Street View Imagery and POIs for Multi-Modal Urban Neighborhood Embedding,” presented at the ArXiv, Jan. 2020. doi: 10.1609/aaai.v34i01.5450.

[5] Y. Xu, S. Jin, Z. Chen, X. Xie, S. Hu, and Z. Xie, “Application of a graph convolutional network with visual and semantic features to classify urban scenes,” International Journal of Geographical Information Science, vol. 36, pp. 2009–2034, Mar. 2022, doi: 10.1080/13658816.2022.2048834.

[6] Z. Li, B. Chen, S. Wu, M. Su, J. Chen, and B. Xu, “Deep learning for urban land use category classification: A review and experimental assessment,” Remote Sensing of Environment, Sep. 2024, doi: 10.1016/j.rse.2024.114290.

[7] P. Liu and F. Biljecki, “A review of spatially-explicit GeoAI applications in Urban Geography,” Int. J. Appl. Earth Obs. Geoinformation, vol. 112, p. 102936, Aug. 2022, doi: 10.1016/j.jag.2022.102936.

[8] H. Aini, “Visualizing identity: linguistic landscapes and ethnolinguistic signage in indonesian multicultural neighborhoods,” ijls, vol. 1, no. 1, pp. 51–60, Mar. 2026.

[9] D. Sartono, “Visual discourse of urban space: a semiotic analysis of street art and linguistic graffiti in Jakarta,” ijls, vol. 1, no. 1, pp. 1–11, Mar. 2026.

[10] E. Papadakis, B. Adams, S. Gao, B. Martins, G. Baryannis, and A. Ristea, “Explainable artificial intelligence in the spatial domain (X‐GeoAI),” Transactions in GIS, vol. 26, Sep. 2022, doi: 10.1111/tgis.12996.

[11] R. Marasinghe, T. Yigitcanlar, S. Mayere, T. Washington, and M. Limb, “Towards Responsible Urban Geospatial AI: Insights From the White and Grey Literatures,” Journal of Geovisualization and Spatial Analysis, vol. 8, Jun. 2024, doi: 10.1007/s41651-024-00184-2.

[12] Fawaid and P. Handayani, “Visualizing inequality: uncovering gender bias in Islamic textbooks of Indonesia,” Cogent Education, vol. 12, no. 1, 2025, doi: 10.1080/2331186X.2025.2593342.

[13] P. A. Andini, “Multimodal representation of environmental messages in Indonesian eco-murals,” ijls, vol. 1, no. 1, pp. 21–30, Mar. 2026.

[14] P. Hidayat, “Spatial framing in tourism advertising: the linguistic construction of exoticism in Indonesian destination campaigns,” ijls, vol. 1, no. 1, pp. 31–40, Mar. 2026.

[15] R. Pahlawan, “Language of typography in political posters: visual rhetoric in Indonesian electoral design,” ijls, vol. 1, no. 1, pp. 41–50, Mar. 2026.

[16] Daroji, “Architectural narratives and spatial language: a discourse study of traditional house descriptions in indonesia,” ijls, vol. 1, no. 1, pp. 12–20, Mar. 2026.

[17] F. Fang et al., “Synthesizing location semantics from street view images to improve urban land-use classification,” International Journal of Geographical Information Science, vol. 35, pp. 1802–1825, Oct. 2020, doi: 10.1080/13658816.2020.1831515.

[18] T. Arulananth et al., “Semantic segmentation of urban environments: Leveraging U-Net deep learning model for cityscape image analysis,” PLOS ONE, vol. 19, Apr. 2024, doi: 10.1371/journal.pone.0300767.

[19] M. Fleischmann and D. Arribas-Bel, “Decoding (urban) form and function using spatially explicit deep learning,” Comput. Environ. Urban Syst., vol. 112, p. 102147, Sep. 2024, doi: 10.1016/j.compenvurbsys.2024.102147.

[20] Chen et al., “Mapping essential urban land use categories with open big data: Results for five metropolitan areas in the United States of America,” ISPRS Journal of Photogrammetry and Remote Sensing, Aug. 2021, doi: 10.1016/j.isprsjprs.2021.06.010.

[21] W. Huang, L.-Z. Cui, M. Chen, D. Zhang, and Y. Yao, “Estimating urban functional distributions with semantics preserved POI embedding,” International Journal of Geographical Information Science, vol. 36, pp. 1905–1930, Mar. 2022, doi: 10.1080/13658816.2022.2040510.

[22] Y. Guo, J. Tang, H. Liu, X. Yang, and M. Deng, “Identifying up-to-date urban land-use patterns with visual and semantic features based on multisource geospatial data,” Sustainable Cities and Society, Jan. 2024, doi: 10.1016/j.scs.2024.105184.

[23] S. Qu, H. Wang, Z. Hu, Z. Wang, and S. Hu, “Mixed-use urban land parcels identification integrating geospatial data and machine learning,” Geo-spatial Information Science, vol. 28, pp. 1743–1756, Sep. 2024, doi: 10.1080/10095020.2024.2374996.

[24] X. Zou et al., “Deep Learning for Cross-Domain Data Fusion in Urban Computing: Taxonomy, Advances, and Outlook,” Inf. Fusion, vol. 113, p. 102606, Feb. 2024, doi: 10.1016/j.inffus.2024.102606.

[25] Z. Li, J. Xu, S. Wang, Y. Wu, and H. Li, “StreetviewLLM: Extracting Geographic Information Using a Chain-of-Thought Multimodal Large Language Model,” ArXiv, vol. abs/2411.14476, Nov. 2024, doi: 10.48550/arxiv.2411.14476.