Journal of the Architectural Institute of Korea (대한건축학회논문집)

Architectural Institute of Korea (대한건축학회)
권호 표지
DOI QR코드

DOI QR Code

Developing Geo-coded Street-level Pedestrian Volume Data Using Google Street View Data and Artificial Intelligence Models

Google Street View 데이터와 인공지능을 활용한 위치기반 가로공간 보행자 데이터 수집 방법 개발

  • Kim, Youngwoo (Institute of Engineering Research, Seoul National University) ;
  • Hwang, Yongha (Space Information and Planning, University of Michigan) ;
  • Jeong, Eunseok (Spatial Design, College of Design, Sangmyung University) ;
  • Kang, Bumjoon (Department of Architecture and Architectural Engineering, Seoul National University)
  • 김영우 (서울대 공학연구원) ;
  • 황용하 (미시간대 공간정보계획부) ;
  • 정은석 (상명대 디자인대학 스페이스디자인전공) ;
  • 강범준 (서울대 건축학과)
  • Received : 2023.06.15
  • Accepted : 2023.08.15
  • Published : 2023.09.30
⟨ Previous Next ⟩

Abstract

Pedestrian count data serves various purposes within architectural, urban planning, and related fields. Typically, this data is collected by government agencies and commercial survey companies. However, conventional methods of recording pedestrian data demand significant time and effort. Consequently, data availability is restricted to specific timeframes and limited locations. In response to this, we conducted feasibility tests for an object-based pedestrian detection procedure. Google Street View data was used to capture geocoded pedestrian counts at street levels in New York City, the U.S. A validation study was performed against historical pedestrian count data recorded officially in the city at 114 different locations. The results indicated a high agreement rate of over 0.8, suggesting that street-level image data could effectively and economically replace conventional pedestrian counting methods.

Keywords

Acknowledgement

이 연구는 2020년도 한국연구재단 연구비 지원에 의한 결과의 일부임 (과제번호: 2020R1C1C1013021)

References

  1. Aikoh, T., Homma, R., & Abe, Y. (2023). Comparing conventional manual measurement of the green view index with modern automatic methods using google street view and semantic segmentation. Urban Forestry & Urban Greening, 80, 127845. 
  2. Burr, A., Schaeg, N., & Hall, D. M. (2018). Assessing residential front yards using Google Street View and geospatial video: A virtual survey approach for urban pollinator conservation. Applied Geography, 92, 12-20.  https://doi.org/10.1016/j.apgeog.2018.01.010
  3. Byun, M.-R., & Seo, U. (2011). How to measure daytime population in urban streets?: Case of Seoul pedestrian flow survey. Survey Research, 12(2), 27-50. 
  4. Chen, F.-C., Subedi, A., Jahanshahi, M. R., Johnson, D. R., & Delp, E. J. (2022a). Deep learning-based building attribute estimation from Google Street View images for flood risk assessment using feature fusion and task relation encoding. Journal of Computing in Civil Engineering, 36(6), 04022031. 
  5. Chen, L., Chen, C., Ewing, R., McKnight, C. E., Srinivasan, R., & Roe, M. (2013). Safety countermeasures and crash reduction in New York City-Experience and lessons learned. Accident Analysis & Prevention, 50, 312-322.  https://doi.org/10.1016/j.aap.2012.05.009
  6. Chen, L., Lu, Y., Ye, Y., Xiao, Y., & Yang, L. (2022b). Examining the association between the built environment and pedestrian volume using street view images. Cities, 127, 103734. 
  7. City of Seoul. (n.d.). CCTV floating population data collection in Bukchon, Seoul, Seoul open data plaza Homepage. Retrieved August 21, 2023 from http://data.seoul.go.kr/dataList/OA-12838/A/1/datasetView.do 
  8. Deng, M., Yang, W., Chen, C., & Liu, C. (2022). Exploring associations between streetscape factors and crime behaviors using Google Street View images. Frontiers of Computer Science, 16(4), 164316. 
  9. Ewing, R., & Cervero, R. (2010). Travel and the built environment: a meta-analysis. Journal of the American Planning Association, 76(3), 265-294.  https://doi.org/10.1080/01944361003766766
  10. Ewing, R., Hajrasouliha, A., Neckerman, K. M., Purciel-Hill, M., & Greene, W. (2016). Streetscape features related to pedestrian activity. Journal of Planning Education and Research, 36(1), 5-15.  https://doi.org/10.1177/0739456X15591585
  11. Gao, G., Ye, X., Li, S., Huang, X., Ning, H., Retchless, D., & Li, Z. (2023). Exploring flood mitigation governance by estimating first-floor elevation via deep learning and google street view in coastal Texas. Environment and Planning B: Urban Analytics and City Science, 23998083231175681. 
  12. He, K., Zhang, X., Ren, S., & Sun, J. (2016). Deep residual learning for image recognition. Proceedings of the IEEE conference on computer vision and pattern recognition, 770-778. 
  13. Helbich, M., Yao, Y., Liu, Y., Zhang, J., Liu, P., & Wang, R. (2019). Using deep learning to examine street view green and blue spaces and their associations with geriatric depression in Beijing, China. Environment International, 126, 107-117. 
  14. Jiao, L., Zhang, F., Liu, F., Yang, S., Li, L., Feng, Z., & Qu, R. (2019). A survey of deep learning-based object detection. IEEE Access, 7, 128837-128868.  https://doi.org/10.1109/ACCESS.2019.2939201
  15. Kang, B. (2019). Identifying street design elements effective in reducing vehicle-to-pedestrian collisions at intersections in New York City. Accident Analysis and Prevention, 122: 308-317.  https://doi.org/10.1016/j.aap.2018.10.019
  16. Kang, J., Korner, M., Wang, Y., Taubenbock, H., & Zhu, X. X. (2018). Building instance classification using street view images. ISPRS Journal of Photogrammetry and Remote Sensing, 145, 44-59.  https://doi.org/10.1016/j.isprsjprs.2018.02.006
  17. Ki, D., & Lee, S. (2021). Analyzing the effects of Green View Index of neighborhood streets on walking time using Google Street View and deep learning. Landscape and Urban Planning, 205, 103920. 
  18. Lin, T. Y., Maire, M., Belongie, S., Hays, J., Perona, P., Ramanan, D., ... & Zitnick, C. L. (2014). Microsoft coco: Common objects in context. In Computer Vision-ECCV 2014: 13th European Conference, Zurich, Switzerland, September 6-12, 2014, Proceedings, Part V 13 (pp. 740-755). Springer International Publishing. 
  19. New York City. (n.d.). Bi-annual pedestrian counts, NYC OpenData. Retrieved March 25, 2020 from https://data.cityofnewyork.us/Transportation/Bi-Annual-Pedestrian-Counts/2de2-6x2h 
  20. Ozbay, K., Bartin, B., Yang, H., Walla, R., & Williams, R. (2010). Automated Pedestrian Counter: Final Report, February 2010. 
  21. Park, K., & Lee, S. (2018). Application and Validation of a Deep Learning Model to Predict the Walking Satisfaction on Street Level. Journal of the Urban Design Institute of Korea, 19(6), 19-34. 
  22. Ren, S., He, K., Girshick, R., & Sun, J. (2015). Faster R-CNN: Towards real-time object detection with region proposal networks. Advances in Neural Information Processing Systems. 
  23. Retting, R. A., Ferguson, S. A., & McCartt, A. T. (2003). A review of evidence-based traffic engineering measures designed to reduce pedestrian-motor vehicle crashes. American Journal of Public Health, 93(9), 1456-1463.  https://doi.org/10.2105/AJPH.93.9.1456
  24. Ryus, P., Ferguson, E., Laustsen, K. M., Schneider, R. J., Proulx, F. R., Hull, T., & Miranda-Moreno, L. (2014). Guidebook on Pedestrian and Bicycle Volume Data Collection.
  25. Wang, R., Liu, Y., Lu, Y., Zhang, J., Liu, P., Yao, Y., & Grekousis, G. (2019). Perceptions of built environment and health outcomes for older Chinese in Beijing: A big data approach with street view images and deep learning technique. Computers, Environment and Urban Systems, 78, 101386. 
  26. Wang, X., Liono, J., McIntosh, W., & Salim, F. D. (2017). Predicting the city foot traffic with pedestrian sensor data. Proceedings of the 14th EAI International Conference on Mobile and Ubiquitous Systems: Computing, Networking and Services. 
  27. Yue, H., Xie, H., Liu, L., & Chen, J. (2022). Detecting people on the street and the streetscape physical environment from Baidu street view images and their effects on community-level street crime in a Chinese city. ISPRS International Journal of Geo-Information, 11(3), 151. 
  28. 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. Applied Geography, 63, 337-345. 
  29. Yin, L., & Wang, Z. (2016). Measuring visual enclosure for street walkability: Using machine learning algorithms and Google Street View imagery. Applied Geography, 76, 147-153.  https://doi.org/10.1016/j.apgeog.2016.09.024
  30. Zhang, L., Lin, L., Liang, X., & He, K. (2016). Is faster R-CNN doing well for pedestrian detection? European Conference on Computer Vision. 
  31. Zhou, H., He, S., Cai, Y., Wang, M., & Su, S. (2019). Social inequalities in neighborhood visual walkability: Using Street View imagery and deep learning technologies to facilitate healthy city planning. Sustainable Cities and Society, 50, 101605.