Keywords
built environment
deep learning
image classification
google street view
How to Cite
Abstract
The environment’s composition can have an impact on human behavior, however, this relationship remains uncertain until the cities' qualities and landscape can be analyzed empirically. Images obtained through Google Street View (GSV) enable a large volume of data for automated assessment of environmental characteristics. Deep learning techniques have advanced in the identification of compositional elements of the built environment. In this sense, this study seeks to investigate and test a procedure for identifying the configuration and composition of the urban landscape, classifying images obtained from GSV through a deep learning approach. From an image dataset of three different neighborhoods in Londrina-PR, a deep learning model for image classification was proposed. The model had a good performance, correctly attributing 87.6% of the samples to the corresponding neighborhoods in the case study. Compositional characteristics were empirically identified, considering the distribution of the samples in the obtained search space. The proposed model contributes to the definition of spatial units as well as in the measurement of environmental qualities, optimizing data collection, expanding sample sizes, and providing objectivity to results. This approach contributes to the expansion of city's analytical scales, identifying compositional and relational patterns in the understanding of elements influent in human behavior.
References
Alexander, C. (1979). The Timeless Way of Building, 1o ed. Oxford University Press, New York.
Amorim, R. R., Oliveira, R. C. de. (2008). As unidades de paisagem como uma categoria de análise geográfica: o exemplo do município de São Vicente-SP. Soc. Nat. 20, 177–198.
Anderson, P., He, X., Buehler, C., Teney, D., Johnson, M., Gould, S. (2018). Bottom-Up and Top-Down Attention for Image Captioning and Visual Question Answering, in: IEEE Conference on Computer Vision and Pattern Recognition (CVPR). pp. 6077–6086.
Badland, H. M., Opit, S., Witten, K., Kearns, R. A., Mavoa, S. (2010). Can Virtual Streetscape Audits Reliably Replace Physical Streetscape Audits? J. Urban Heal. 87, 1007–1016. Disponível em: https://doi.org/10.1007/s11524-010-9505-x
Beidack, A. R. dos S., Fresca, T. M. (2011). Urban Restructuring and new centralities: a study about the north zone of Londrina – PR. Bol. Geográfico 29, 147–163. Disponível em: https://doi.org/10.4025/bolgeogr.v29i2.9898
Ben-joseph, E., Lee, J. S., Seoul, M., Cromley, E. K., Laden, F., Troped, P. J. (2015). Virtual and Actual: Relative Accuracy of On-Site and Web-based Instruments in Auditing the Environment for Physical Activity. Heal. Place. 19, 138–150. Disponível em: https://doi.org/10.1016/j.healthplace.2012.11.001.
Bishop, C. M. (2006). Pattern Recognition and Machine Learning, 1o ed. Springer Science + Business Media, New York, NY.
Burges, C. J. C. (2010). Geometric Methods for Feature Extraction and Dimensional Reduction - A Guided Tour, in: Maimon, Oded, Rokach, L. (Eds.), Data mining and knowledge discovery handbook: a complete guide for practitioners and researchers. Springer Science + Business Media, pp. 53–82. Disponível em: https://doi.org/10.1007/978-0-387-09823-4
Chan, D. M., Rao, R., Huang, F., Canny, J. F. (2018). t-SNE-CUDA: GPU-Accelerated t-SNE and its Applications to Modern Data. 2018 30th Int. Symp. Comput. Archit. High Perform. Comput. 330–338. Disponível em: https://doi.org/10.1109/SBAC-PAD.2018.00060
Deng, J., Dong, W., Socher, R., Li, L., Li, K., Fei-fei, L. (2009). ImageNet: A large-scale hierarchical image database, in: IEEE Conference on Computer Vision and Pattern Recognition. Miami, FL, pp. 248–255. Disponível em: https://doi.org/10.1109/CVPR.2009.5206848
Doersch, C., Singh, S., Gupta, A., Sivic, J., Efros, A., Doersch, C., Singh, S., Gupta, A., Sivic, J., Efros, A., Makes, W., Look, P. (2012). What Makes Paris Look like Paris? ACM Trans. Graph. (SIGGRAPH 2012) 31. Disponível em: https://doi.org/10.1145/2185520.2185597
Ewing, R., Handy, S. (2009). Measuring the Unmeasurable: Urban Design Qualities Related to Walkability. J. Urban Des. 14, 65–84. Disponível em: https://doi.org/10.1080/13574800802451155
Fan, H., Zheng, L., Yan, C., Yang, Y. (2018). Unsupervised Person Re-identification: Clustering and Fine-tuning. arXiv Preprint. Disponível em: https://arxiv.org/abs/1705.10444
Gehrke, S. R., Wang, L. (2020). Operationalizing the neighborhood effects of the built environment on travel behavior. J. Transp. Geogr. 82, 12. Disponível em: https://doi.org/10.1016/j.jtrangeo.2019.102561
Glorot, X., Bengio, Y. (2010). Understanding the difficulty of training deep feedforward neural networks, in: Proceedings of the 13th International Conference on Artificial Intelligence and Statistics (AISTATS) 2010, Chia La- guna Resort, Sardinia, Italy., pp.249–256.
Gowda, S. N., Yuan, C. (2019). ColorNet: Investigating the Importance of Color Spaces for Image Classification, in: Jawahar C., Li H., Mori G., S.K. (Ed.), Computer Vision – ACCV 2018. Springer International Publishing, pp.581–596. Disponível em: https://doi.org/10.1007/978-3-030-20870-7
He, K., Sun, J. (2016). Deep Residual Learning for Image Recognition, in: EEE Conference on Computer Vision and Pattern Recognition (CVPR). pp.1–9.
IBGE (2018). Panorama municipal: Londrina-Paraná [WWW Document]. URL https://cidades.ibge.gov.br/brasil/pr/londrina/panorama (accessed 3.26.18).
Kamalipour, H., Faizi, M., Memarian, G. (2014). Safe place by design: Urban crime in relation to spatiality and sociality. Curr. Urban Stud. 2, pp.152–162. Disponível em: https://doi.org/10.4236/cus.2014.22015
Kamilaris, A., Prenafeta-Boldú, F. X. (2018). Deep Learning in Agriculture: A Survey. Comput. Electron. Agric. 147, pp.70–90. Disponível em: https://doi.org/10.1016/j.compag.2018.02.016
Kent, J. L., Ma, L., Mulley, C. (2017). The objective and perceived built environment: What matters for happiness? Cities Heal. 8834, pp.1–13. Disponível em: https://doi.org/10.1080/23748834.2017.1371456
Kingma, D. P., Ba, J. L. (2014). Adam: A method for stochastic optimization. arXiv Preprint. pp.1–15. Disponível em: https://arxiv.org/abs/1412.6980
Lecun, Y., Bengio, Y., Hinton, G. (2015). Deep learning. Nature 521, pp.436–444. Disponível em: https://doi.org/10.1038/nature14539
Lee, S. M., Conway, T. L., Frank, L. D., Saelens, B. E., Cain, K. L., Sallis, J. F. (2017). The Relation of Perceived and Objective Environment Attributes to Neighborhood Satisfaction. Environ. Behav. 49, pp.136–160. Disponível em: https://doi.org/10.1177/0013916515623823
Litjens, G., Kooi, T., Bejnordi, B. E., Arindra, A., Setio, A., Ciompi, F., Ghafoorian, M., Laak, J. A. W. M., Van Der, Ginneken, B. Van, Sánchez, C. I. (2017). A survey on deep learning. Medical image analysis 42, pp.60–88. Disponível em: https://doi.org/10.1016/j.media.2017.07.005
Liu, L., Silva, E. A., Wu, C., Wang, H. (2017). A machine learning-based method for the large-scale evaluation of the qualities of the urban environment. Comput. Environ. Urban Syst. 65, pp.113–125. Disponível em: https://doi.org/10.1016/j.compenvurbsys.2017.06.003
Maaten, L. V. D., Hinton, G. (2008). Visualizing data using t-SNE. Journal of machine learning research, 9, pp.2579-2605.
Medeiros, F. F., Grigio, A. M. (2019). Identificação das Unidades Homogêneas e Padrão da Ocupação Urbana como subsídio ao ordenamento territorial em Mossoró, RN – Brasil. EURE (Santiago), 45, pp.245–270.
Middel, A., Lukasczyk, J., Zakrzewski, S., Arnold, M., Maciejewski, R. (2019). Urban form and composition of street canyons: A human-centric big data and deep learning approach. Landsc. Urban Plan. 183, pp.122–132. Disponível em:https://doi.org/10.1016/j.landurbplan.2018.12.001
Moosavi, V. (2017). Urban morphology meets deep learning: Exploring urban forms in one million cities, town and villages across the planet. arXiv Preprint. arXiv:1709, 1–10. Disponível em: https://arxiv.org/abs/1709.02939.
Oura, K. (2006). Verticalização em Londrina - Paraná (1950-2005): A produção do espaço urbano e seu desenvolvimento pelos edifícios verticais. Dissertação de Mestrado - Universidade Presbiteriana São Paulo.
Poggio, T., Mhaskar, H., Rosasco, L., Miranda, B., Liao, Q. (2017). Why and When Can Deep – but Not Shallow – Networks Avoid the Curse of Dimensionality: a Review. Int. J. Autom. Comput. 14, pp.503–519. Disponível em: https://doi.org/10.1007/s11633-017-1054-2
Powers, D. M. W. (2011). Evaluation: from Precision, Recall and F-measure to ROC, Informedness, Markedness and Correlation. J. Mach. Learn. Technol. 2, pp.37–63.
Prefeitura do Município de Londrina (2020). Sistema de Informação Geográfica de Londrina – SIGLON [WWW Document]. Disponível em: http://www1.londrina.pr.gov.br/index.php?option=com_content&view=article&id=20114&Itemid=1988 (Acessado em 15 de janeiro de 2020).
Sallis, J. F., Cain, K. L., Conway, T. L., Gavand, K. A., Millstein, R. A., Geremia, C. M., Frank, L. D., Saelens, B. E., Glanz, K., King, A. C. (2015). Is Your Neighborhood Designed to Support Physical Activity? A Brief Streetscape Audit Tool. Prev. Chronic Dis. Disponível em: https://doi.org/10.5888/pcd12.150098
Seresinhe, C. I., Preis, T., Mackerron, G., Moat, H. S. (2019). Happiness is Greater in More Scenic Locations. Sci. Rep. pp.1–11. Disponível em: https://doi.org/10.1038/s41598-019-40854-6
Shen, Q., Member, S., Zeng, W., Ye, Y., Stefan, M., Schubiger, S., Burkhard, R., Qu, H. (2018). StreetVizor?: Visual Exploration of Human-Scale Urban Forms Based on Street Views. IEEE Trans. Vis. Comput. Graph. 24, pp.1004–1013. Disponível em: https://doi.org/10.1109/TVCG.2017.2744159
Shorten, C., Khoshgoftaar, T. M. (2019). A survey on Image Data Augmentation for Deep Learning. J. Big Data. Disponível em: https://doi.org/10.1186/s40537-019-0197-0
Takahashi, R., Matsubara, T., Uehara, K. (2019). Data Augmentation using Random Image Cropping and Patching for Deep CNNs. ArXiv Preprint. abs/1811.0, pp.1–16. Disponível em: https://arxiv.org/abs/1811.09030.
Tan, Y., Tang, P., Zhou, Y., Luo, W., Kang, Y., Li, G. (2017). Neurocomputing Photograph aesthetical evaluation and classi fication with deep convolutional neural networks. Neurocomputing 228, pp.165–175. Disponível em: https://doi.org/10.1016/j.neucom.2016.08.098
Töws, R. L., Mendes, C. M., Vercezi, J. T. (2010). The city as a business: the case from Londrina-PR and from Maringá-PR. Bol. Geográfico 28, pp.91–103.
Yamaki, H. T. (2017). Terras do Norte: paisagem e morfologia, 1 ed. Ed. H. Yamaki e UEL, Londrina.
Yin, L., Cheng, Q., Wang, Z., Shao, Z. (2015). ‘Big data’ for pedestrian volume: Exploring the use of Google Street View images for pedestrian counts. Appl. Geogr. 63, pp.337–345. Disponível em: https://doi.org/10.1016/j.apgeog.2015.07.010
Yin, L., Wang, Z. (2016). Measuring visual enclosure for street walkability: Using machine learning algorithms and Google Street View imagery. Appl. Geogr. 76, pp.147–153. Disponível em: https://doi.org/10.1016/j.apgeog.2016.09.024
Yin, R. K. (2001). Estudo de caso: Planejamento e Métodos, 2o. ed. Bookman Companhia Editora, São Paulo.
Zhang, F., Zhou, B., Liu, L., Liu, Y., Fung, H. H., Lin, H., Ratti, C. (2018). Landscape and Urban Planning Measuring human perceptions of a large-scale urban region using machine learning. Landsc. Urban Plan. 180, pp.148–160. Disponível em: https://doi.org/10.1016/j.landurbplan.2018.08.020
Zhang, W., Li, W., Zhang, C., Hanink, D. M., Li, X., Wang, W. (2017). Parcel feature data derived from Google Street View images for urban land use classification in Brooklyn, New York City. Data in Brief. 12, pp.175–179. Disponível em:https://doi.org/10.1016/j.dib.2017.04.002
Zhou, B., Khosla, A., Lapedriza, A., Torralba, A., Oliva, A. (2016). Places: An Image Database for Deep Scene Understanding. J. Vis. 17, pp.1–12.
This work is licensed under a Creative Commons Attribution 4.0 International License.
Copyright (c) 2020 Ana Luiza Favarão Leão, Hugo Queiroz Abonizio, Prof. Dr. Sylvio Barbon Júnior, Profa. Dra. Milena Kanashiro