Skip to main content
SpringerLink
Log in

Measuring the Built Environment with Google Street View and Machine Learning: Consequences for Crime on Street Segments

  • Original Paper
  • Published:
Journal of Quantitative Criminology Aims and scope Submit manuscript

Abstract

Objectives

Despite theoretical interest in how dimensions of the built environment can help explain the location of crime in micro−geographic units, measuring this is difficult.

Methods

This study adopts a strategy that first scrapes images from Google Street View every 20 meters in every street segment in the city of Santa Ana, CA, and then uses machine learning to detect features of the environment. We capture eleven different features across four main dimensions, and demonstrate that their relative presence across street segments considerably increases the explanatory power of models of five different Part 1 crimes.

Results

The presence of more persons in the environment is associated with higher levels of crime. The auto−oriented measures—vehicles and pavement—were positively associated with crime rates. For the defensible space measures, the presence of walls has a slowing negative relationship with most crime types, whereas fences did not. And for our two greenspace measures, although terrain was positively associated with crime rates, vegetation exhibited an inverted−U relationship with two crime types.

Conclusions

The results demonstrate the efficacy of this approach for measuring the built environment.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7

Similar content being viewed by others

Notes

  1. https://paperswithcode.com/sota/semantic-segmentation-on-cityscapes

  2. The python code for implementing this algorithm can be obtained here: https://github.com/lexfridman/mit-deep-learning.

  3. Source: Cityscapes website, https://www.cityscapes-dataset.com/dataset-overview/#labeling-policy (accessed on Jan. 28, 2020)

  4. Deeplab3+ extracts 19 elements, although there were no trains in our images and therefore we had 18 elements. Further, “person” and “rider” are collapsed into the category of humans. And “car”, “truck”, “bus”, “motorcycle”, and “bicycle” are collapsed into vehicles. And “poles”, “traffic signs”, and “traffic lights” are collapsed into objects.

  5. GSV images acquired from wide roads can be distorted when viewing the other side of the street, as the google vehicle only travels in one direction. However, our medium-sized city has major streets that only rarely have three lanes in each direction, and thus this is a less salient problem in our case study area. Furthermore, this distortion problem is mitigated to some extent by our strategy of computing the average of images taken in four different directions when measuring the features.

  6. These measures were constructed based on blocks and the exponential decay of blocks around a focal block. We then average the values for blocks adjacent to a particular segment. Of course, the buffer values in these adjacent blocks are extremely highly correlated, so this strategy does not introduce problems.

  7. The logged alpha term capturing overdispersion was highly significant in all models, indicating the need to use a negative binomial regression rather than a Poisson model.

  8. For example, in our study site, about 60% of robberies occur after dark, and about 50% of aggravated assaults, 45% of burglaries and motor vehicle thefts, and 40% of larcenies occur after dark.

References

  • Badrinarayanan Vijay, Kendall Alex, Cipolla Roberto (2017) Segnet: a deep convolutional encoder − decoder architecture for image segmentation. IEEE Trans Pattern Anal Mach Intell 39(12):2481–2495

    Article  Google Scholar 

  • Beavon, Daniel J.K., P.L. Brantingham, and P.J. Brantingham. (1994) The Influence of Street Networks on the Patterning of Property Offenses. In Crime Prevention Studies, edited by R. V. Clarke

  • Bernasco Wim, Block Richard L (2011) Robberies in Chicago: a Block − level analysis of the influence of crime generators, crime attractors, and offender anchor points. J Res Crime Delinq 48(1):33–57

    Article  Google Scholar 

  • Boessen Adam, Hipp John R (2015) Close − ups and the scale of ecology: land uses and the geography of social context and crime. Criminology 53(3):399–426

    Article  Google Scholar 

  • Boessen Adam, Hipp John R (2018) Parks as crime inhibitors or generators: examining Parks and the Role of their Nearby Context. Soc Sci. Res. 76(1):186–201

    Article  Google Scholar 

  • Bowers Kate (2014) Risky facilities: crime radiators or crime absorbers? a comparison of internal and external levels of theft. J Quan Criminol 30(1):389–414

    Article  Google Scholar 

  • Brantingham Paul J, Brantingham Patricia L (1984) Patterns in crime. MacMillan, New York

    Google Scholar 

  • Brantingham, Patricia L., and Paul J. Brantingham. (2008) Crime Pattern Theory. In Environmental Criminology and Crime Analysis, edited by R. Wortley and L. Mazerolle. Portland, OR: Willan Publishing

  • Browning Christopher R, Byron Reginald A, Calder Catherine A, Krivo Lauren J, Kwan Mei-Po, Lee Jae Yong, Peterson Ruth D (2010) Commercial density, residential concentration, and crime: land use patterns and violence in neighborhood context. J. Res. Crime Delinq 47(3):329–357

    Article  Google Scholar 

  • Chen, Liang − Chieh, Yukun Zhu, George Papandreou, Florian Schroff, and Hartwig Adam. (2018) Encoder − decoder with atrous separable convolution for semantic image segmentation. Paper read at Proceedings of the European conference on computer vision (ECCV)

  • Clarke Philippa, Ailshire Jennifer A, Melendez Robert, Bader Michael, Morenoff Jeffrey (2010) Using google earth to conduct a neighborhood audit: reliability of a virtual audit instrument. Health Place 16(6):1224–1229

    Article  Google Scholar 

  • Cohen Deborah A, Inagami Sanae, Finch Brian (2008) The built environment and collective efficacy. Health Place 14(2):198–208

    Article  Google Scholar 

  • Cordts, Marius, Mohamed Omran, Sebastian Ramos, Timo Rehfeld, Markus Enzweiler, Rodrigo Benenson, Uwe Franke, Stefan Roth, and Bernt Schiele. (2016) The cityscapes dataset for semantic urban scene understanding. Paper read at proceedings of the IEEE conference on computer vision and pattern recognition

  • Curtis Andrew, Mills Jacqueline W (2011) Crime in urban post − disaster environments: a methodological framework from New Orleans. Urban Geogr 32(4):488–510

    Article  Google Scholar 

  • Davies Toby, Johnson Shane D (2014) Examining the relationship between road structure and burglary risk via quantitative network analysis. J Quan Criminol 31(3):481–507

    Article  Google Scholar 

  • Deryol Rustu, Wilcox Pamela, Logan Matthew, Wooldredge John (2016) Crime places in context: an illustration of the multilevel nature of hot spot development. J Quan Criminol 32(2):305–325

    Article  Google Scholar 

  • Du, Kun, Jinsheng Ning, and Li Yan. (2020). How long is the sun duration in a street canyon? — Analysis of the view factors of street canyons. Building and Environment 172

  • Edwards Nicole, Hooper Paula, Trapp Gina S, Bull Fiona, Boruff Bryan, Giles-Corti Billie (2013) Development of a public open space desktop auditing tool (POSDAT): a remote sensing approach. Appl Geogr 38:22–30

    Article  Google Scholar 

  • Felson Marcus (2002) Crime and everyday life. Third, Edition. Sage, Thousand Oaks, CA

    Google Scholar 

  • Feng Shun Q, Piza Eric L, Kennedy Leslie W, Caplan Joel M (2018) Aggravating effects of alcohol outlet types on street robbery and aggravated assault in New York City. J Crime Justice 42(3):257–273

    Article  Google Scholar 

  • Gong Fang-Ying, Zeng Zhao-Cheng, Zhang Fan, Li Xiaojiang, Ng Edward, Norford Leslie K (2018) Mapping sky, tree, and building view factors of street canyons in a high − density urban environment. Build Environ 134:155–167

    Article  Google Scholar 

  • Gracia Enrique, Herrero Juan (2006) Perceived neighborhood social disorder and residents’ attitudes toward reporting child physical abuse. Child Abuse Neglect 30(4):357–365

    Article  Google Scholar 

  • Groff Elizabeth (2011) Exploring ‘near’: characterizing the spatial extent of drinking place influence on crime. Aust New Zealand J Criminol 44(2):156–179

    Article  Google Scholar 

  • Groff, Elizabeth, and Eric S. McCord. (2011) The Role of Neighborhood Parks as Crime Generators. Security Journal:1 − 24

  • He Li, Páez Antonio, Liu Desheng (2017) Built environment and violent crime: an environmental audit approach using google street view. Comput, Environ Urban Syst 66:83–95

    Article  Google Scholar 

  • Hipp John R (2010) A dynamic view of neighborhoods: the reciprocal relationship between crime and neighborhood structural characteristics. Soc Problem 57(2):205–230

    Article  Google Scholar 

  • Hipp John R, Bates Christopher J, Lichman Moshe, Smyth Padhraic (2019) Using social media to measure temporal ambient population: does it help explain local crime rates? Justice Quar 36(4):718–748

    Article  Google Scholar 

  • Hipp John R, Kim Young-an (2019) Explaining the temporal and spatial dimensions of robbery: differences across measures of the physical and social environment. J Crim Justice 60(1):1–12

    Article  Google Scholar 

  • Hipp John R, Wo James, Kim Young-an (2017) Studying neighborhood crime across different macro spatial scales: the case of robbery in four cities. Soc Sci Res 68(1):15–29

    Article  Google Scholar 

  • Hipp, John R., Jonathan Corcoran, Rebecca Wickes, and Tiebei Li. (2014) Examining the social porosity of environmental features on neighborhood sociability and attachment. PLOS: One 9 (1):1 − 13

  • Hipp, John R., Young − an Kim, and James Wo. (2020) Micro − scale, meso − scale, macro−scale, and temporal scale: Comparing the relative importance for robbery risk in New York City. Justice Quarterly Forthcoming

  • Infogroup. (2015) Reference USA historical business data, edited by Infogroup. Reference USA, Papillion, NE

    Google Scholar 

  • Jacobs Jane (1961) The death and life of great american cities. Random House, New York

    Google Scholar 

  • Kang Hyeon-Woo, Kang Hang-Bong (2017) Prediction of crime occurrence from multi−modal data using deep learning. PLoS One 12(4):e0176244

    Article  Google Scholar 

  • Kestens Yan, Thériault Marius, Rosiers François Des (2004) The impact of surrounding land use and vegetation on single−family house prices. Environ Plann B: Plann Design 31(4):539–567

    Article  Google Scholar 

  • Kim Young-an (2018) Examining the relationship between the structural characteristics of place and crime by imputing census block data in street segments: is the pain worth the gain? J Quant Criminol 34(1):67–110

    Article  Google Scholar 

  • Kim Young-an, Hipp John R (2017) Physical boundaries and City boundaries: consequences for crime patterns on street segments? Crime Delinq 64(2):227–254

    Article  Google Scholar 

  • Kimpton Anthony, Corcoran Jonathan, Wickes Rebecca (2017) Greenspace and crime: an analysis of greenspace types, neighboring composition, and the temporal dimensions of crime. J Res Crime Delinq 54(3):303–337

    Article  Google Scholar 

  • Krylov Vladimir A, Kenny Eamonn, Dahyot Rozenn (2018) Automatic discovery and geotagging of objects from street view imagery. Remote Sens 10(5):661

    Article  Google Scholar 

  • Lee Sang-Woo, Ellis Christopher D, Kweon Byoung-Suk, Hong Sung-Kwon (2008) Relationship between landscape structure and neighborhood satisfaction in urbanized areas. Landsc Urban Plann 85(1):60–70

    Article  Google Scholar 

  • Li Xiaojiang, Zhang Chuanrong, Li Weidong, Ricard Robert, Meng Qingyan, Zhang Weixing (2015) Assessing street−level urban greenery using google street view and a modified green view index. Urban Urban Green14(3):675–685

    Article  Google Scholar 

  • Liu, Penghua, X. Liu, M. Liu, Q. Shi, Jinxing Yang, Xiaocong Xu, and Y. Zhang. (2019) Building footprint extraction from High−resolution images via spatial residual inception convolutional neural network. Remote Sens 11 (7):830

  • Long, Jonathan, Evan Shelhamer, Trevor Darrell. (2015) Fully convolutional networks for semantic segmentation. Paper read at proceedings of the IEEE conference on computer vision and pattern recognition

  • Lu Yi (2018) Using google street view to investigate the association between street greenery and physical activity. Landscape Urban Plann 191:1–9

    Google Scholar 

  • Malleson, Nick, and Martin A. Andresen. (2015) Spatio−temporal crime hotspots and the ambient population. Crime Sci 4 (1)

  • Marco Miriam, Gracia Enrique Gómez, Martín-Fernández Manuel, López-Quílez Antonio (2017) Validation of a google street view−based neighborhood disorder observational scale. J Urban Health 94(2):190–198

    Article  Google Scholar 

  • Mooney SJ, DiMaggio CJ, Lovasi GS, Neckerman KM, Bader MD, Teitler JO, Rundle AG, Mooney Stephen J, DiMaggio Charles J, Lovasi Gina S, Neckerman Kathryn M, Bader Michael D. M, Teitler Julien O, Sheehan Daniel M, Jack Darby W, Rundle Andrew G (2016) Use of google street view to assess environmental contributions to pedestrian injury. Am J Public Health 106(3):462–469

    Article  Google Scholar 

  • Nagata, Shohei, Tomoki Nakaya, Tomoya Hanibuchi, Shiho Amagasa, Hiroyuki Kikuchi, and Shigeru Inoue. (2020) Objective scoring of streetscape walkability related to leisure walking: statistical modeling approach with semantic segmentation of Google Street View images. Health & Place 66 (102428)

  • Nesoff Elizabeth D, Milam Adam J, Pollack Keshia M, Curriero Frank C, Bowie Janice V, Gielen Andrea C, Furr-Holden Debra M (2018) Novel methods for environmental assessment of pedestrian injury: creation and validation of the inventory for pedestrian safety infrastructure. J Urban Health 95(2):208–221

    Article  Google Scholar 

  • Newman Oscar (1972) Defensible space: people and design in the violent city. Architectural Press, London

    Google Scholar 

  • Odgers Candice L, Caspi Avshalom, Bates Christopher J, Sampson Robert J, Moffitt Terrie E (2012) Systematic social observation of children’s neighborhoods using google street view: a reliable and cost−effective method. J Child Psychol Psychiatr 53(10):1009–1017

    Article  Google Scholar 

  • Patino Jorge E, Duque Juan C, Pardo-Pascual Josep E, Ruiz Luis A (2014) Using remote sensing to assess the relationship between crime and the urban layout. Appl Geograp 55(1):48–60

    Article  Google Scholar 

  • Ratcliffe Jerry H (2012) The spatial extent of criminogenic places: a changepoint regression of violence around bars. Geograp Anal 44:302–320

    Article  Google Scholar 

  • Rogers Shannon H, Halstead John M, Gardner Kevin H, Carlson Cynthia H (2010) Examining walkability and social capital as indicators of quality of life at the municipal and neighborhood scales. Appl Res Qual Life 6(2):201–213

    Article  Google Scholar 

  • Rundle Andrew G, Bader Michael D.M., Richards Catherine A, Neckerman Kathryn M, Teitler Julien O (2011) Using google street view to audit neighborhood environments. Am J Prev Med 40(1):94–100

    Article  Google Scholar 

  • Stucky Thomas D, Ottensmann John R (2009) Land use and violent crime. Criminology 47(4):1223–1264

    Article  Google Scholar 

  • Sung Hyungun, Lee Sugie (2015) Residential built environment and walking activity: empirical evidence of Jane Jacobs’ urban vitality. Trans Res Part D: Trans Environ 41:318–329

    Article  Google Scholar 

  • Wang Ruoyu, Yi Lu, Zhang Jinbao, Liu Penghua, Yao Yao, Liu Ye (2019) The relationship between visual enclosure for neighbourhood street walkability and elders’ mental health in China: using street view images. J Trans Health 13:90–102

    Article  Google Scholar 

  • Wang Haoying, Yunfeng Hu, Tang Li, Zhuo Qi (2020) Distribution of urban blue and green space in beijing and its influence factors. Sustainability 12(6):2252

    Article  Google Scholar 

  • Wang, Mingshu, and Floris Vermeulen. (2020) Life between buildings from a street view image: What do big data analytics reveal about neighbourhood organisational vitality? Urban Studies Online first:004209802095719

  • Weisburd David, Bernasco Wim, Bruinsma Gerben (eds) (2009) Putting crime in its place: units of analysis in spatial crime research. Springer Verlaag, New York

    Google Scholar 

  • Yang, Yiyang, Dongsheng He, Zhonghua Gou, Ruoyu Wang, Ye Liu, and Yi Lu. (2019) Association between street greenery and walking behavior in older adults in Hong Kong. Sustainable Cities and Society 51: (101747)

  • Ye, Yu, Daniel Richards, Yi Lu, Xiaoping, Zhuang Song, Yu, Wei Zeng, and Teng Zhong. (2018) Measuring daily accessed street greenery: A human-scale approach for informing better urban planning practices. Landscape and Urban Planning 191:1-13.  

  • Zhang, Yihong, Panote Siriaraya, Yukiko Kawai, and Adam Jatowt. (2019) Analysis of street crime predictors in web open data. J Intell Inform Syst:1−25

  • Zhao, Hengshuang, Jianping Shi, Xiaojuan Qi, Xiaogang Wang, and Jiaya Jia. (2017) Pyramid scene parsing network. Paper read at proceedings of the IEEE conference on computer vision and pattern recognition

  • Zhuang Song Yu, Zeng Wei, Zhong Teng (2018) Measuring daily accessed street greenery: a human−scale approach for informing better urban planning practices. Landsc Urban Plann 191:1–13

    Google Scholar 

  • Zhuang Song Yu, Zeng Wei, Zhong Teng. (2019) Pathways: Examining street network configurations, structural characteristics and spatial crime patterns in street segments. J Quan Criminol Forthcom

Download references

Funding

This research is supported in part by the Metropolitan Futures Initiative (MFI) at the University of California, Irvine and the Korea Research Foundation (NRF − 2018R1A2A2A05023583).

Author information

Authors and Affiliations

  1. Department of Criminology, Law and Society and Department of Sociology, University of California, Irvine, 3311 Social Ecology II, Irvine, CA, 92697, USA

    John R. Hipp

  2. Department of Urban Planning and Engineering, Hanyang University, 222 Wangsimni-ro, Seongdong-gu, Seoul, 04763, Korea

    Sugie Lee & Donghwan Ki

  3. Department of Urban Planning and Public Policy, University of California, Irvine, 206E Social Ecology I, Irvine, CA, 92697, USA

    Jae Hong Kim

Authors
  1. John R. Hipp
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Sugie Lee
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Donghwan Ki
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Jae Hong Kim
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to John R. Hipp.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Hipp, J.R., Lee, S., Ki, D. et al. Measuring the Built Environment with Google Street View and Machine Learning: Consequences for Crime on Street Segments. J Quant Criminol 38, 537–565 (2022). https://doi.org/10.1007/s10940-021-09506-9

Download citation

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s10940-021-09506-9

Keywords

Search

Navigation