Streetscape Evaluation Using Deep Learning in Light of Local Biases
DOI:
https://doi.org/10.31185/wjes.Vol14.Iss2.971Keywords:
Streetscape Quality, Urban Visual Perception, Deep Learning, ImageabilityAbstract
The evaluation of Streetscapes has seen increasing development in the use of artificial intelligence and deep learning techniques to analyze Streetscape in a more objective and continuous manner. Many studies still rely on human assessments as a primary reference to verify the accuracy of intelligent models, with limited research addressing the impact of residents’ spatial background and local expertise on urban evaluation outcomes. This study aims to compare the assessments of local experts and external experts in evaluating urban landscape quality based on imageability and enclosure indicators, and to analyze the extent to which each group aligns with the results of a deep learning model. The research relied on a set of urban images evaluated by urban design experts, which were then compared with the outputs of a model based on convolutional neural networks. The results showed a clear difference between the two groups, with external experts achieving higher agreement rates with the model, reaching 82.26% for the Imageability indicator and 88.71% for the containment index, compared to local experts, whose agreement rates were 69.11% and 62.60%, respectively. The results indicate that spatial familiarity and direct experience with a place may influence urban perception and produce a certain type of perceptual bias, while external evaluations tend to show a greater degree of neutrality and consistency with AI-based analysis.
References
[1] Lynch, K., The Image of the City. MIT Press, 1960.
[2] Carmona, M., Public Places Urban Spaces: The Dimensions of Urban Design. Routledge, 2010.
[3] Dubey, A., Naik, N., Parikh, D., Raskar, R., and Hidalgo, C. A., “Deep learning the city: Quantifying urban perception at a global scale,” in European Conference on Computer Vision (ECCV), pp. 196–212, Springer, 2016.
[4] Li, Y., Zhang, Y., and Li, Z., “The visual quality of streets: A human-centred continuous measurement based on machine learning algorithms and street view images,” Cities, vol. 115, p. 103238, 2021.
[5] Nasar, J. L., The Evaluative Image of the City. Sage Publications, 1998.
[6] Rapoport, A., The Meaning of the Built Environment: A Nonverbal Communication Approach. University of Arizona Press, 1990.
[7] Ewing, R., and Clemente, O., Measuring Urban Design: Metrics for Livable Places. Island Press, 2013.
[8] Salesses, P., Schechtner, K., and Hidalgo, C. A., “The collaborative image of the city: Mapping the inequality of urban perception,” PLOS ONE, vol. 8, no. 7, p. e68400, 2013.
[9] Mehta, V., The Street: A Quintessential Social Public Space. Routledge, 2014.
[10] Gehl, J., Cities for People. Island Press, 2010.
[11] Appleyard, D., “Why buildings are known: A predictive tool for architects and planners,” Environment and Behavior, vol. 1, no. 2, pp. 131–156, 1969.
[12] Cullen, G., The Concise Townscape. Architectural Press, 1961.
[13] Ewing, R., and Handy, S., “Measuring the unmeasurable: Urban design qualities related to walkability,” Journal of Urban Design, vol. 14, no. 1, pp. 65–84, 2009.
[14] Wang, R., Lu, Y., Weng, Q., and Zhang, X., “A comparison of two deep-learning-based urban perception models: Which one is better?” Computational Urban Science, vol. 1, no. 1, pp. 1–14, 2021.
[15] Zhang, Y., Li, Z., and Li, Y., “Evaluation of spatial visual perception of streets based on deep learning and spatial syntax,” Sustainable Cities and Society, vol. 59, p. 102194, 2020.
[16] Relph, E., Place and Placelessness. Pion Limited, 1976.
[17] Tuan, Y. F., Space and Place: The Perspective of Experience. University of Minnesota Press, 1977.
[18] Lang, J., Creating Architectural Theory: The Role of the Behavioral Sciences in Environmental Design. Van Nostrand Reinhold, 1987.
[19] Herzog, T. R., and Leverich, O. L., “Searching for legibility,” Environment and Behavior, vol. 35, no. 4, pp. 459–477, 2003.
[20] Montgomery, J., “Making a city: Urbanity, vitality and urban design,” Journal of Urban Design, vol. 3, no. 1, pp. 93–116, 1998.
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Roqaia Salim, Ihsan Abbas, Manaf K. H. Altaleb
This work is licensed under a Creative Commons Attribution 4.0 International License.
