A study on methods of evaluating urban street quality based on human perceptions : 인간인지를 중심으로 도시가로 질 평가방법에 관한 연구

국문 초록 (Abstract)

좋은 도시가로 공간의 질은 사회 발전을 촉진하는 데 핵심적인 역할을 한다. 중국의 도시화 과정에서 정부의 정책은 도시의 전반적인 질과 주민들의 생활 수준을 향상시키는 것을 목표로 가로 공간의 미세화를 강조했다. 이러한 변화는 지난 수십 년 동안 급속한 도시화 과정에서 발생한 도시 문제에 대한 직접적인 대응이다. 정책의 핵심은 보다 지속 가능한 현대적인 도시 환경을 조성하는 것이다. 그 중 가로 공간은 도시 기반 구조의 중요한 부분이며, 합리적인 설계는 도시의 전반적인 질을 향상시키는 열쇠로 간주된다. 코로나 이후 시대에 사람들은 도시가로에 대한 수요가 크게 변화했으며 '인간 중심' 디자인 개념이 더욱 강조되었다. 코로나 전염병은 공중 보건과 정신 건강에 지대한 영향을 미치며 안전, 아름다움, 풍요와 같은 가로환경에 대한 주민들의 요구를 증가시킨다. 이러한 환경은 기본 기능을 충족해야 할 뿐만 아니라 코로나 전염병으로 인한 외로움과 심리적 스트레스를 완화하기 위해 정서적 가치를 제공해야 한다.
한편, 빅데이터와 신기술의 적용은 도시가로 인지 연구에 새로운 가능성을 제시한다. 스트리트 뷰 이미지 등 빅데이터를 활용하고 딥러닝 기술을 접목해 도시가로 공간의 변화를 보다 정확하게 이해하고 분석할 수 있다. 이러한 기술은 데이터 수집 및 분석의 효율성을 향상시킬 뿐만 아니라 도시 계획 및 설계를 보다 과학적이고 정확하게 만든다. 이를 통해 도시 계획자는 주민들의 요구를 더 잘 이해하고 현대 생활에 더 적합한 가로 공간을 만들 수 있다.선전시 난산구(深圳市 南山區)는 중국에서 빠르게 발전하는 도시이기 때문에 기후가 가로 시각요소에 계절적 차이를 일으키지 않고 바이두 지도의 자원이 풍부하다. 이에 본 연구에서는 난산구를 대상으로 실증 연구를 수행하였다.
이에 따른 결과를 요약하면 다음과 같다.
첫째, cityscape 데이터 세트를 사용하여 deeplab v3+ 모델을 훈련하면 최종 정확도가 90%를 초과했다. 이러한 정확도는 스트리트 뷰의 효과적인 분할을 보장하기에 충분하며 분할된 시각요소는 도시가로 인지와 시각요소의 관계를 연구하는 특징 데이터로 사용할 수 있다.
둘째, 지원자들의 인지 데이터를 수집하여 인지 데이터 세트를 만든다. 다양한 랜덤 트리를 순회하는 방식으로 랜덤 포레스트 모델을 훈련시킨 결과, 여섯 가지 인지의 예측 정확도는 모두 90%를 넘었다. 이러한 높은 정확도는 도시가로 인지 예측의 유효성을 보장하며, 예측 결과는 도시가로 인지와 시각요소의 연관성을 연구하는 데 사용될 레이블 데이터로 활용될 수 있을 것으로 판단된다. 훈련된 랜덤 포레스트 모델을 사용하여 도시가로의 인지를 예측하면, 난산구의 부유함과 안전감이 일반적으로 높은 것을 발견할 수 있다. 이러한 인지는 주로 중부와 동부의 상업 지역, 그리고 남부의 주거 지역과 과학기술 혁신 지역에 집중되어 있다. 활발함은 중부의 종합 서비스 지역과 남부의 주거 지역에서 더욱 두드러지며, 서부의 상업 및 서비스 지역에서는 상대적으로 낮다. 아름다움 인지는 보통 수준이며, 지루함은 전반적으로 낮지만 서부의 상업 및 서비스 지역에서 더욱 두드러진다. 우울함은 중부 주거 지역에서 높으며 다른 지역에서는 낮다. 또한, 여섯 가지 인지 유형은 공간 분포에서 뚜렷한 자기상관성을 보여준다. 이러한 분석은 효과적인 거시적 개선 전략 수립에 견고한 데이터 기반을 제공한다.
셋째, 도시에서 선택도가 높은 지역은 일반적으로 도시의 핵심 상업 및 주거 지역이다. 이러한 지역은 인구가 밀집되어 있고 교통 네트워크가 잘 발달되어 있으며, 대부분 이차 도로와 가로로 구성된다. 선택도가 낮은 가로는 보통 도시 외곽이나 사람들의 발길이 드문 곳에 위치하며, 일상적인 이동의 우선 선택지가 되기 어렵다. 이러한 높은 선택도 지역에서는 유지 및 최적화 작업이 기존 상업 활동을 강화하고 주거 환경의 질을 향상시키는 데 집중되어야 한다. 낮은 선택도 지역에서는 교통 연결성을 향상시키고, 공공 시설을 추가하며, 지역의 매력을 높여 도시 내에서의 위치를 개선할 필요가 있다. 높은 선택도 도로의 배경 하에, 높은 아름다움 인지의 분포는 상대적으로 분산되어 있으며, 북부와 중부에 밀집되어 있고, 남부와 서부는 밀집되어 있지 않다. 지루함과 우울함 인지는 적게 나타나며, 난산구 전체적으로 긍정적인 인지 상태가 좋다는 것을 보여준다. 높은 선택도 도로의 배경 하에, 부유함, 안전함, 활기찬 인지의 분포는 상대적으로 넓다. 이러한 분석 결과는 계획자들에게 가치 있는 데이터를 제공하며, 도시 계획자들에게 귀중한 관점을 제공하여, 제한된 예산 하에서 더 현명한 결정을 내릴 수 있게 한다. 특히, 자원이 제한되고 경제적 압박이 큰 도시의 경우, 이러한 방법은 도시가로의 최적화를 실현하는 데 도움을 줄 수 있다.
넷째, 난산구의 개선이 필요해야 할 특정 지역에 대해서는 시각요소의 공간 이질성을 고려하여 개선 전략과 제안을 맞춤화할 수 있다. 긍정적인 인지를 향상시키기 위해, 난산구는 도시의 미학적 디자인, 생태 지속 가능성 및 공공 안전을 종합적으로 고려해야 한다. 이에는 친환경 건축 재료 사용, 도시 녹화 강화, 지역 예술 및 문화 요소와의 융합, 시각적 및 인지적 균형에 대한 중요성, 생태 지속 가능성 원칙의 적용 및 공공 안전 조치 강화가 포함된다. 이러한 조치들은 안전하고 아름답고 활기찬 부유한 도시 공간을 조성하고, 난산구의 도시 특성을 보여준다. 지루함과 우울함 문제를 해결하기 위해 난산구는 이러한 부정적인 인지를 줄이기 위한 일련의 종합적 조치를 시행해야 한다. 이러한 조치에는 보행로에 예술 설치물 및 그림 추가, 보행로의 기능성 및 편안함 향상, 환경 질을 개선하기 위한 녹지대 추가, 고층 건물 밀도를 줄이기 위한 건축 배치 재계획 및 건축 디자인에 자연 요소의 통합이 포함되어야 한다.
본 연구는 도시가로 인지 평가 분야에서 혁신적인 다학제적 접근 방식을 제시하며, 이론적 혁신, 방법론적 개선 및 실용적 가치를 보여주고자 하였다. 이론적으로는 컴퓨터 과학, 지리학, 수학 등 분야의 이론과 기술을 통합하여 도시가로 인지 평가에서의 학제간 통합의 중요성을 보여주었다. 방법론적으로는 일련의 기술 적용을 통해 전통적 연구 방법의 효율성과 비용 문제를 극복하고 데이터 처리 및 분석의 효율성과 정확성을 향상시킬 수 있다. 실천적으로는 도시 계획에 대한 정확하고 목적성 있는 제안 및 분석을 제공하는 포괄적인 실용적 분석 도구를 제공했다. 이러한 종합적 분석 방법은 도시 계획 및 디자인 실무에 중요한 의미를 가지며, 도시 환경의 최적화 및 발전에 대한 과학적 지침과 효과적인 지원을 제공한다.

목차 (Table of Contents)

• I. Introduction 1
• 1. Background and purpose of the study. 1
• 1.1. Study background. 1
• 1.1.1. Refined requirements for urban construction. 1
• 1.1.2. People-centric requirements for city streets 2
• 1.1.3. The emergence of new technologies and methods 3
• 1.2. Study purpose. 5
• 2. Methodology and framework of the study 7
• 2.1. Research methodology 7
• 2.2. Research framework. 11
• Ⅱ. Theory and Exploration. 14
• 1. Theories related to the spatial quality of urban streets. 14
• 1.1. Definition of urban street space 14
• 1.2. Characteristics of Urban Street Space. 16
• 1.3. Definition of spatial quality of urban streets 18
• 1.4. The connotation of spatial quality in urban streets 20
• 2. Image semantic segmentation techniques 22
• 2.1. Concept of image semantic segmentation technique 22
• 2.2. Application scenarios of image semantic segmentation technology 23
• 3. Machine learning techniques. 26
• 3.1. Machine learning technology concepts 26
• 3.2. Types of machine learning techniques. 26
• 3.3. State of the art in machine learning technology applications 29
• 4. Space syntax theory 31
• 4.1. Overview of space syntax concepts 31
• 4.2. Overview of space syntax applicability 32
• 4.3. Methods of space syntax analysis 33
• 4.4. Indicators related to space syntax 39
• 5. Literature review of relevant international research 42
• 5.1. Literature review on spatial quality of urban streets 42
• 5.1.1. Shifting spatial quality claims on urban streets 42
• 5.1.2. Multi-dimensional considerations on the quality of urban streets. 46
• 5.1.3. Quantifying the spatial quality of people-centric urban streets 49
• 5.1.4. Research trends in spatial quality of urban streets 53
• 5.2. Literature review on image semantic segmentation techniques 55
• 5.3. Literature review of machine learning techniques 58
• 5.4. Literature review of space syntax 63
• 5.5. Enlightenment 64
• Ⅲ. Data processing and street view semantic segmentation model training. 66
• 1. Focus of this chapter 66
• 2. Overview of the study area. 68
• 3. Data sources and pre-processing. 73
• 3.1. Urban street network data collection 73
• 3.2. Street view image data collection and processing. 74
• 3.2.1. Advantages of street view images. 74
• 3.2.2. Collection of street view images 77
• 3.2.3. Screening of Street view images 82
• 4. Semantic segmentation of street view images 84
• 4.1. Evaluation of commonly used semantic segmentation models 84
• 4.2. Training strategies for DeepLab v3+ models 90
• 4.2.1. Selection of training dataset. 90
• 4.2.2. Description of training parameters 94
• 4.2.3. Setting of training parameters 101
• 4.3. DeepLab v3+ model image segmentation results checking 103
• 4.3.1. Interpretation of performance assessment indicators 103
• 4.3.2. Evaluation of model performance 104
• 5. Summary of the chapter. 107
• Ⅳ. Machine learning-based predictions for urban street perception 108
• 1. Focus of this chapter 108
• 2. Selection of machine learning models 110
• 2.1. Advantages and limitations of different models 110
• 2.1.1. Linear Regression 110
• 2.1.2. Logistic Regression 110
• 2.1.3. Decision Trees 111
• 2.1.4. Random Forests 112
• 2.1.5. Support Vector Machines 113
• 2.1.6. K-Nearest Neighbors. 114
• 2.1.7. K-Means Clustering 115
• 2.2. Summary 116
• 3. Construction of random forest models 117
• 3.1. Random Forest model runtime environment building 117
• 3.2. Production of the random forest model dataset 119
• 3.2.1. Production of feature data 119
• 3.2.2. Production of label data 121
• 3.3. Training strategies for random forest models 123
• 4. Predictive analysis of perception based on random forests. 129
• 5. Analysis of urban street perception at different road classes. 141
• 6. Analysis of perception spatial characteristics of urban streets 146
• 6.1. Global spatial autocorrelation analysis of urban street perception. 146
• 6.2. Local spatial autocorrelation analysis of urban street perception 151
• 7. Spearman correlation analysis of urban street perception 156
• 8. Macro suggestion for urban street development in Nanshan District 160
• 9. Summary of the chapter. 163
• Ⅴ. Priority area identification for urban street perception and space syntax integration 165
• 1. Focus of this chapter 165
• 2. Space syntax model selection. 167
• 3. Choice analysis 169
• 3.1. Urban Street network processing. 169
• 3.1.1. Urban Street network classification 169
• 3.1.2. Urban Street network consolidation 170
• 3.1.3. Urban Street network simplification 171
• 3.1.4. Urban Street network centreline extraction. 172
• 3.2. Depthmap X choice export. 173
• 3.3. Urban Street choice analysis 174
• 4. Priority area recognition 177
• 4.1. Urban street extraction with high perception. 177
• 4.2. Urban Street extraction with high choice 178
• 4.3. Priority area recognition with high choice and perception. 179
• 5. Summary of the chapter. 182
• Ⅵ. Exploring the direction of improvement in priority areas based on heterogeneity of perceptual and visual elements 184
• 1. Focus of this chapter 184
• 2. Screening of factors influencing urban street perception. 186
• 3. Comparison of models for interpreting urban street perception 188
• 3.1. Introduction to different types of explanatory models 188
• 3.2. Description of model evaluation parameters 192
• 3.3. Test of spatial heterogeneity of visual elements for perceptions 194
• 3.4. Comparative analysis of three explanatory models. 200
• 3.4.1. Comparative analysis of the bandwidth of three explanatory models 200
• 3.4.2. Comparative analysis of the performance of the three explanatory models 208
• 3.4.3. Summary of three explanation models 209
• 4. Exploration of the improvement direction of priority areas based on the spatial heterogeneity of visual elements 211
• 4.1. Exploration of the improvement direction of priority areas with positive perception. 212
• 4.1.1. Exploration of the direction of improvement in the priority areas of the perception of beautiful 212
• 4.1.2. Exploration of the direction of improvement in the priority areas of the perception of wealthy 223
• 4.1.3. Exploration of the direction of improvement in the priority areas of the perception of lively. 230
• 4.1.4. Exploration of the direction of improvement in the priority area of the perception of safety 238
• 4.2. Exploration of the improvement direction of priority areas with negative perception. 244
• 4.2.1. Exploration of the direction of improvement in the priority area of the perception of boring 244
• 4.2.2. Exploration of the direction of improvement in the priority areas of the perception of depressing 249
• 5. Recommendations for the development of urban street rehabilitation 255
• 6. Summary of the chapter. 257
• Ⅶ. Conclusion 258
• 1. Conclusions. 258
• 1.1. The value of evaluation methods for urban street perception. 258
• 1.2. Summary of the evaluation of the perception quality of urban streets in Nanshan District 262
• 2. Research limitations and future prospects 265
• REFERENCES 268
• Abstract (In Korean) 288

참고문헌 (Reference)

1. Public Space, Carr S., Cambridge University Press, , 1992

2. Learning machines, Nilsson NJ, McGrawHill: New York, , 1965

3. The Ordinary City, Amin A, Graham S., Transactions of the Institute of British Geographers22: 411–429. doi:10.1111/j.0020-2754.1997.00411. x, , 1997

4. AI Powered Society, Bera RK, Rochester, NY doi:10.2139/ssrn.3256873, , 2018

5. The image of the city, Lynch K., 33. print Cambridge, Mass.: M. I. T. Press, , 2008

6. Evaluating Public Space, Mehta V., 19: 53–88. doi:10.1080/13574809.2013.854698, , 2014

7. Semi-supervised learning, Schölkopf B, Chapelle O, Zien A., Cambridge, Mass.: MIT Press, , 2006

8. The new science of cities, Batty, M., MIT Press, , 2013

9. A survey on image segmentation, Fu KS, Mui JK, 13: 3–16. doi:10.1016/0031-3203(81)90028-5, , 1981

10. Applications of Decision Trees, P., Vijayaraghavan A, Thomas T, Emmanuel S editors, Thomas T, Emmanuel S. In, P. Vijayaraghavan A, Machine Learning Approaches in Cyber Security Analytics Singapore: Springer pp. 157–184. doi:10.1007/978-981- 15-1706-8_9, , 2020

11. Geographic Information Analysis, Unwin D., O’Sullivan D, John Wiley & Sons, , 2003

12. Toward an Urban Design Manifesto., Appleyard AJ Donald, 6th ed The City Reader. 6th ed Routledge, , 2015

13. The role of big data in smart city, Hashem IAT, Adewole K, Anuar NB, Gani A et al, Chang V, Yaqoob I, 36: 748–758. doi:10.1016/j. ijinfomgt.2016.05.002, , 2016

14. Smart urban governance in smart city, Nhu DT, Toan NQ, IOP Conf Ser: Mater Sci Eng869: 022021. doi:10.1088/1757-899X/869/2/022021, , 2020

15. Edge Computing: Vision and Challenges, Zhang Q, Xu L., Li Y, Cao J, Shi W, 3: 637–646. doi:10.1109/JIOT.2016.2579198, , 2016

16. Quality of Urban Spaces and Wellbeing, Adams M., Wellbeing. John Wiley & Sons, Ltd pp. 1–21. doi:10.1002/9781118539415. wbwell064, , 2013

17. Machine Learning and Visual Perception, Zhang B, Walter de Gruyter GmbH & Co KG, , 2020

18. Managing Sustainable Development 2nd ed, Christie MC Ian, London Routledge doi:10.4324/9781315091525, , 2017

19. New Progress in Artificial Intelligence, Winston PH, cited Available https://dspace. mit. edu/handle/1721.1/6895, , 1974

20. Creativity as social and spatial process, P. Haynes B, Sailer K., editor Facilities29: 6–18. doi:10.1108/02632771111101296, , 2011

21. Semi-supervised learning: a brief review, Reddy Y, Pulabaigari V, B E., 7: 81. doi:10.14419/ijet. v7i1.8.9977, , 2018

22. A survey on evolutionary machine learning, Mei Y, Lensen A, Chen Q, Bi Y, Al-Sahaf H, Sun Y et al, 49: 205–228. doi:10.1080/03036758.2019.1609052, , 2019

23. Multicollinearity and Regression Analysis, Daoud JI, 949: 012009. doi:10.1088/1742-6596/949/1/012009, , 2017

24. Correlation Analysis in Biological Studies, Yadav S., 4: 116. doi:10.4103/jpcs. jpcs_31_18, , 2018

25. Deep Reinforcement Learning: A Brief Survey, Brundage M, Bharath AA, Arulkumaran K, Deisenroth MP, IEEE Signal Processing Magazine34: 26–38. doi:10.1109/MSP.2017.2743240, , 2017

26. Effects of COVID-19 on business and research, Donthu, N., Gustafsson, A., 117: 284–289. doi:10.1016/j. jbusres.2020.06.008, , 2020

27. 63 Research and Application of Urban Function, Li S., Aesthetic Expression and Human Behavior on Commercial Street in Chinese Old Downtown Area -- Renovation of Nanjing Zhujiang Road Available https//urn. kb. se/resolve?urn=urn:nbn:se:bth-4471, , 2014

28. Introduction to Space Syntax in Urban Studies, Yamu C., van Nes A, doi:10.1007/978-3-030-59140-3, , 2021

29. A Fast Learning Algorithm for Deep Belief Nets, Osindero S, Teh Y-W., Hinton GE, 18: 1527–1554. doi:10.1162/neco.2006.18.7.1527, , 2006

30. Local indicators of spatial association—LISA, Anselin, L., 27: 93–115. doi:10.1111/j.1538-4632.1995. tb00338. x, , 1995

31. The research of the fast SVM classifier method, Li J, Yang Y, Yang Y, 2015 12th International Computer Conference on Wavelet Active Media Technology and Information Processing (ICCWAMTIP) pp. 121–124. doi:10.1109/ICCWAMTIP.2015.7493959, , 2015

32. Transit Oriented Development: Making it Happen, Renne JL, Curtis C, Bertolini L., Ashgate Publishing, Ltd., , 2009

33. 10 Urban form resilience: A meso-scale analysis, Sharifi A, Cities93: 238–252. doi:10.1016/j. cities.2019.05.010, , 2019

34. COVID-19 Pandemic: Applying a Multisystemic Lens, Rolland JS, Family Process59: 922–936. doi:10.1111/famp.12584, , 2020

35. Complexity in Organizations: A Research Overview, Johannessen SO, Routledge, , 2022

36. Survey on semantic image segmentation techniques, Kapadia AD, Rahevar M., Shah A, Chavda JB, Sevak JS, 2017 International Conference on Intelligent Sustainable Systems (ICISS) pp. 306–313. doi:10.1109/ISS1.2017.8389420, , 2017

37. The Street: A Quintessential Social Public Space, Mehta V., Routledge, , 2013

38. Google Street View: navigating the operative image, Hoelzl I, Marie R., Visual Studies29: 261–271. doi:10.1080/1472586X.2014.941559, , 2014

39. Revitalization of Urban Public Spaces: An Overview, Yunus RM, Ramlee M, Omar D, Samadi Z., 201: 360–367. doi:10.1016/j. sbspro.2015.08.187, , 2015

40. Urban Forms: The Death and Life of the Urban Block, Depaule J-C, Samuels I, Castex J, Panerai P, Routledge, , 2004

41. Economic Restructuring and Suburbanization in China, Ma LJC, Zhou Y, 21: 205–236. doi:10.2747/0272-3638.21.3.205, , 2000

42. 71 THE KEY ISSUES IN TRANSPORT AND URBAN DEVELOPMENT, LICHFIELD DB NATHANIEL, Routledge, , 1995

43. Contour Detection and Hierarchical Image Segmentation, Fowlkes C, Arbeláez P, Malik J., Maire M, 33: 898–916. doi:10.1109/TPAMI.2010.161, , 2011

44. Machine Learning: An Artificial Intelligence Approach, Carbonell JG, Mitchell TM, Michalski RS, Springer Science & Business Media, , 2013

45. Machine learning: Trends, perspectives, and prospects, Jordan M, Mitchell T., 349: 255–260. doi:10.1126/science. aaa8415, , 2015

46. Pavement Crack Image Detection based on Deep Learning, Wang J, Wei R., Lyu P, Proceedings of the 2019 3rd International Conference on Deep Learning Technologies. New York, NY, USA: Association for Computing Machinery pp. 6–10. doi:10.1145/3342999.3343003, , 2019

47. Fully convolutional networks for semantic segmentation, Darrell, T., Shelhamer, E., Long, J., pp. 3431–3440 Available https://openaccess. thecvf. com/content_cvpr_2015/html/Long_Fully_Convolutional_Networ ks_2015_CVPR_paper. html, , 2015

48. Network and Psychological Effects in Urban Movement In, Iida S. AG Mark DM editors, Hillier B, Spatial Information Theory Berlin, Heidelberg: Springer pp. 475–490. doi:10.1007/11556114_30, , 2005

49. Economic growth and mental health in 21st century China, Wang Q, Tapia Granados JA, 220: 387–395. doi:10.1016/j. socscimed.2018.11.031, , 2019

50. Linking the quality of public spaces to quality of life, Beck H., 2: 240–248. doi:10.1108/17538330911013933, , 2009

51. Machine Learning from Theory to Algorithms: An Overview, Alzubi J, Nayyar A, Kumar A, 1142: 012012. doi:10.1088/1742-6596/1142/1/012012, , 2018

52. Life in the pandemic: Social isolation and mental health, Bhullar N, Usher K, Jackson D., 29: 2756–2757. doi:10.1111/jocn.15290, , 2020

53. Street view imagery in urban analytics and GIS: A review, Biljecki F, Ito K., 215: 104217. doi:10.1016/j. landurbplan.2021.104217, , 2021

54. 268 The elements of townscape and the art of urban design, Taylor N., 4: 195–209. doi:10.1080/13574809908724446, , 1999

55. 29 Towards a user-centred theory of the built environment, Vischer JC, Building Research & Information36: 231–240. doi:10.1080/09613210801936472, , 2008

56. 93 An overview of the supervised machine learning methods, Nasteski V., HORIZONSB4: 51–62. doi:10.20544/HORIZONS. B.04.1.17. P05, , 2017

57. COVID‑19 and its consequences on mental health (Review), Mueller C, Schizas D, Terniotis C et al, Ouranidis A, Tsiptsios D, Tsamakis K, 21: 1–1. doi:10.3892/etm.2021.9675, , 2021

58. Computer science as empirical inquiry: symbols and search, Newell A, Simon HA, ACM Turing Award Lectures. New York: Association of Computing Machinery p. 1975. doi:10.1145/1283920.1283930, , 2007

59. Literature review of Industry 4.0 and related technologies, Gursev S., Oztemel E, 31: 127–182. doi:10.1007/s10845-018-1433-8, , 2020

60. Public Places Urban Spaces: The Dimensions of Urban Design, Carmona M., 3rd ed New York: Routledge doi:10.4324/9781315158457, , 2021

61. A Deep Learning-Based Image Semantic Segmentation Algorithm, Sun CS and Z, 2023;19: 98–108. doi:10.3745/JIPS.02.0191, , 2023

62. Cities Full of Symbols: A Theory of Urban Space and Culture, Nas PJM, Leiden University Press doi:10.5117/9789087281250, , 2011

63. Multimodal urban mobility and multilayer transport networks, Battiston F., Alessandretti L, Natera Orozco LG, Szell M, Saberi M, 23998083221108190. doi:10.1177/23998083221108190, , 2022

64. Using Google Street View to Audit Neighborhood Environments, Neckerman KM, Rundle AG, Bader MDM, Teitler JO, Richards CA, 40: 94–100. doi:10.1016/j. amepre.2010.09.034, , 2011

65. 202 Space Syntax in Healthcare Facilities Research: A Review, Luo Y, Haq S, HERD5: 98–117. doi:10.1177/193758671200500409, , 2012

66. Embracing Change: Continual Learning in Deep Neural Networks, Hadsell R, Pascanu R., Rusu AA, Rao D, 24: 1028–1040. doi:10.1016/j. tics.2020.09.004, , 2020

67. From Abstract to Concrete: Subjective Reading of Urban Space, Kallus R., 6: 129–150. doi:10.1080/13574800120057818, , 2001

68. Conservation machine learning: a case study of random forests, Sipper M, Moore JH, 11: 3629. doi:10.1038/s41598-021-83247-4, , 2021

69. The Cityscapes Dataset for Semantic Urban Scene Understanding, Cordts M, Ramos S, Rehfeld T, Omran M, Benenson R et al, Enzweiler M, pp. 3213–3223 Available: https://openaccess. thecvf. com/content_cvpr_2016/html/Cordts_The_Cityscapes_Dataset_C VPR_2016_paper. html, , 2016

70. Space is the machine: a configurational theory of architecture, Hillier B, Space Syntax: London, UK London, UK: Space Syntax Available https//discovery. ucl. ac. uk/id/eprint/3881/, , 2007

71. 180 Some Studies in Machine Learning Using the Game of Checkers, Samuel AL, 3: 210–229. doi:10.1147/rd.33.0210, , 1959

72. 59 Streets and social life in cities: a taxonomy of sociability, Mehta V., Urban Des Int24: 16–37. doi:10.1057/s41289-018-0069-9, , 2019

73. Object-Contextual Representations for Semantic Segmentation J-M, Chen X, Brox T, Yuan Y, Wang J. A, Frahm editors, Bischof H, In Computer Vision – ECCV Cham: Springer International Publishing pp. 173–190. doi:10.1007/978-3-030- 58539-6_11, , 2020

74. Walkable streets: pedestrian behavior, perceptions and attitudes, Mehta V., Journal of Urbanism: International Research on Placemaking and Urban Sustainability1: 217– 245. doi:10.1080/17549170802529480, , 2008

75. Leisure public space and quality of life in the urban environment, Lloyd K, Auld C., Urban Policy and Research21: 339–356. doi:10.1080/0811114032000147395, , 2003

76. People and Environment: Behavioural Approaches in Human Geography, Lewis GJ, Walmsley DJ, Routledge, , 2014

77. A Review of Deep-Learning-Based Medical Image Segmentation Methods, Liu S, Song L, Zhang Y, Liu X, Sustainability13: 1224. doi:10.3390/su13031224, , 2021

78. Big Data Technology and Applications in Intelligent Transportation, Kim T-H, Arabnia HR, Zhang D, Qu X, Zhao J., Mohammed S, IEEE Access Special Section Editorial IEEE Access8: 201331–201344. doi:10.1109/ACCESS.2020.3035440, , 2020

79. Jane Jacobs Critique of Zoning: From Euclid to Portland and Beyond, Wickersham J., 28: 547, , 2000

80. Public Space, Public Life: an Interaction How To Study Public Life, Gehl J, Svarre B, Svarre B editors, Gehl J, In Washington, DC: Island Press/Center for Resource Economics pp. 1–8. doi:10.5822/978-1-61091-525-0_1, , 2013

81. Smart health: Big data enabled health paradigm within smart cities, Lau RYK, Demirkan H, Azad MdAK, Pramanik MI, Expert Systems with Applications87: 370–383. doi:10.1016/j. eswa.2017.06.027, , 2017

82. Urbanization and industrialization impact of CO2 emissions in China, Liu X, Bae J., 172: 178–186. doi:10.1016/j. jclepro.2017.10.156, , 2018

83. Sensing Cities: Regenerating Public Life in Barcelona and Manchester, Degen MM, Psychology Press, , 2008

84. Built Environments, Constructed Societies: Inverting Spatial Analysis, Vis BN, Sidestone Press, , 2009

85. Rated preference and complexity for natural and urban visual material, Wendt JS, Kaplan S, Kaplan R, 12: 354–356. doi:10.3758/BF03207221, , 1972

86. Event Camera Survey and Extension Application to Semantic Segmentation, Jia S., Proceedings of the 4th International Conference on Image Processing and Machine Vision New York, NY, USA: Association for Computing Machinery pp. 115–121. doi:10.1145/3529446.3529465, , 2022

87. Limits of space syntax for urban design: Axiality, scale and sinuosity, Aschwanden GD, Pafka E, Dovey K, Environment and Planning B: Urban Analytics and City Science47: 508–522. doi:10.1177/2399808318786512, , 2020

88. Support Vector Machines In Data Mining and Knowledge Discovery Handbook, Maimon O Rokach L editors, Shmilovici A, Boston, MA: Springer US pp. 231–247. doi:10.1007/978-0-387-09823-4_12, , 2010

89. 260 The Fitting of Straight Lines if Both Variables are Subject to Error, Wald A, 11: 284–300. doi:10.1214/aoms/1177731868, , 1940

90. Landslide susceptibility prediction based on image semantic segmentation, Hu X, Zhao Z, Du B, Sun L et al, Wu G, Han L, 155: 104860. doi:10.1016/j. cageo.2021.104860, , 2021

91. The Mapillary Vistas Dataset for Semantic Understanding of Street Scenes, Neuhold G, Kontschieder P., Ollmann T, Rota Bulo S, pp. 4990–4999 Available https://openaccess. thecvf. com/content_iccv_2017/html/Neuhold_The_Mapillary_Vistas_IC CV_2017_paper. html, , 2017

92. Applications of machine vision in agricultural robot navigation: A review, Chen B, Zhang M., Li H, Zhang Z, Wang T, Computers and Electronics in Agriculture198: 107085. doi:10.1016/j. compag.2022.107085, , 2022

93. Deep Learning the City: Quantifying Urban Perception at a Global Scale In, Parikh D, Hidalgo CA B, Naik N, Dubey A, Raskar R, Leibe Matas J Sebe N Welling M editors Computer Vision – ECCV Cham: Springer International Publishing; 2016. pp. 196–212. doi:10.1007/978-3-319-46448-0_12, , 2016

94. Effects of Smartphone Usage on Walking Speed using Machine Learning Method, Do MS, Jin HR, Hanbat National University, Center of Infrastructure Asset Management, 18: 93–103. doi:10.12815/kits.2019.18.2.93, , 2019

95. Observed and Potential Impacts of the COVID-19 Pandemic on the Environment, Georgiadis T, Legates DR, Piticar A, Mihai Adamescu C, Cheval S, Herrnegger M, 17: 4140. doi:10.3390/ijerph17114140, , 2020

96. Understanding Smart City Practice in Urban China: A Governance Perspective, Ma E, Du S, Han Y, Lin J, Cai J, Sustainability. 2023;15: 7034. doi:10.3390/su15097034, , 2023

97. A comprehensive framework for evaluating the quality of street view imagery, Biljecki F., Hou Y, 115: 103094. doi:10.1016/j. jag.2022.103094, , 2022

98. Strategies for image segmentation combining region and boundary information, Freixenet J, Martı́ J, Muñoz X, Cufı́ X, 24: 375–392. doi:10.1016/S0167-8655(02)00262-3, , 2003

99. 236 Application of Neural and Regression Models in Sports Results Prediction, Roczniok R, Stanula A, Pietraszewski P, Maszczyk A, Zając A, Gołaś A, Procedia - Social and Behavioral Sciences117: 482–487. doi:10.1016/j. sbspro.2014.02.249, , 2014

100. Creating Shared Public Space in the Contested City: The Role of Urban Design, Sterrett K., Gaffikin F, Mceldowney M, 15: 493–513. doi:10.1080/13574809.2010.502338, , 2010

101. Deep Neural Network Self-training Based on Unsupervised Learning and Dropout, Kim N, Lee J-H, Lee H-W, 17: 1–9. doi:10.5391/IJFIS.2017.17.1.1, , 2017

102. Machine Learning Aspects and its Applications Towards Different Research Areas, Kumar Y, Kaur K, Singh G., 2020 International Conference on Computation, Automation and Knowledge Management (ICCAKM) pp. 150–156. doi:10.1109/ICCAKM46823.2020.9051502, , 2020

103. The Role of Context for Object Detection and Semantic Segmentation in the Wild, Mottaghi R, Liu X, Fidler S et al, Chen X, Cho N-G, Lee S-W, pp. 891–898. Available: https://www. cvfoundation. org/openaccess/content_cvpr_2014/html/Mottaghi_The_Role_of_2014_CVPR_ paper. html, , 2014

104. 30 years of adaptive neural networks: perceptron, Madaline, and backpropagation, Widrow B, Lehr MA, 78: 1415–1442. doi:10.1109/5.58323, , 1990

105. Intelligent Semantic Segmentation for Self-Driving Vehicles Using Deep Learning, Kautish S, Vidyarthi A, Bisoy S, Sellat Q, Priyadarshini R, Barik RK, 2022: e6390260. doi:10.1155/2022/6390260, , 2022

106. Research on Financial Technology Innovation and Application Based on 5G Network, Tian M-W, Tian X-X, Wang L, Rodrigues JJPC, Yan S-R, Liu Z-Q, IEEE Access7: 138614–138623. doi:10.1109/ACCESS.2019.2936860, , 2019

107. Chinese Customers Evaluation of Travel Website Quality: A Decision-Tree Analysis, Harrill R., Cárdenas DA, Sun P, 25: 476–497. doi:10.1080/19368623.2015.1037977, , 2016

108. Measuring human perceptions of a large-scale urban region using machine learning, Liu L, Lin H et al, Fung HH, Zhou B, Zhang F, Liu Y, 180: 148–160. doi:10.1016/j. landurbplan.2018.08.020, , 2018

109. Using Space Syntax to Model Pedestrian Movement in Urban Transportation Planning, Rofè Y, Omer I, Lerman Y, Geographical Analysis46: 392–410. doi:10.1111/gean.12063, , 2014

110. A Review of the Evolution of Shared (Street) Space Concepts in Urban Environments, Karndacharuk A, Dunn R., Wilson DJ, 34: 190–220. doi:10.1080/01441647.2014.893038, , 2014

111. Semantic Image Segmentation with Deep Convolutional Nets and Fully Connected CRFs, Kokkinos I, Papandreou G, Chen L-C, Murphy K, Yuille AL, arXiv doi:10.48550/arXiv.1412.7062, , 2016

112. Sites of Vision: The Discursive Construction of Sight in the History of Philosophy, Kleinberg-Levin DM, MIT Press, , 1999

113. Space Syntax and buried cities: The case of the Roman town of Falerii Novi (Italy), Battistin F., 35: 102712. doi:10.1016/j. jasrep.2020.102712, , 2021

114. Urbanisation and its impact on building energy consumption and efficiency in China, Li B, Yao R., 34: 1994–1998. doi:10.1016/j. renene.2009.02.015, , 2009

115. Predicting House Price in India Using Linear Regression Machine Learning Algorithms, Dongariya N, Sethi N., Verma A, Singhi N, Nagar C, 2022 3rd International Conference on Intelligent Engineering and Management (ICIEM) pp. 917–924. doi:10.1109/ICIEM54221.2022.9853185, , 2022

116. Rethinking Urban Parks: Public Space and Cultural Diversity. Rethinking Urban Parks, Low SM, Taplin D, Scheld S., University of Texas Press doi:10.7560/706859, , 2009

117. 104 Space Syntax—A Synopsis of Basic Concepts, Measures, and Empirical Application, Garau C., Yamu C, Bill Hilliers Legacy, van Nes A, Sustainability13: 3394. doi:10.3390/su13063394, , 2021

118. Cyclable Cities: Building Feasible Scenario through Urban Space Morphology Assessment, Fortunato G., Scorza F, 147: 05021039. doi:10.1061/(ASCE)UP.1943-5444.0000713, , 2021

119. Urban Planning in China: Continuity and Change: What the future holds may surprise you, Abramson DB, 72: 197–215. doi:10.1080/01944360608976739, , 2006

120. 68 Remaking the Image: The Changing Legibility and Visibility of Derbys City Center, UK, Cheshmehzangi A., 24: 910–923. doi:10.1080/10911359.2014.914994, , 2014

121. Beyond two dimensions: architecture through three dimensional visibility graph analysis, Psarra S., Varoudis T, The Journal of Space Syntax5: 91–108, , 2014

122. Mathematical Approach to Representation of Locations Using K-Means Clustering Algorithm, Yogeesh N., International Journal of Mathematics And its Applications9: 127–136, , 2021

123. Learning with Kernels: Support Vector Machines, Regularization, Optimization, and Beyond, Scholkopf B, Smola AJ, MIT Press, , 2018

124. Meta-analysis of the relationships between space syntax measures and pedestrian movement, Kamruzzaman Md, Sharmin S, 38: 524–550. doi:10.1080/01441647.2017.1365101, , 2018

125. Review the state-of-the-art technologies of semantic segmentation based on deep learning, Mo Y, Liu F, Liao Y, Yang X, Wu Y, 493: 626–646. doi:10.1016/j. neucom.2022.01.005, , 2022

126. 205 Understanding place holistically: Cities, synergistic relationality, and space syntax, Seamon D., 6 Available https//krex. kstate. edu/handle/2097/35311, , 2015

127. 262 Multiscale Geographically Weighted Regression Computation, Inference, and Application, Li Z., Ph. D. Arizona State University Available https://www. proquest. com/docview/2444854594/abstract/AF498C250F3F4019PQ/1, , 2020

128. Geographical information system parallelization for spatial big data processing: a review, He J., Chen L, Ranjan R, Choo K-KR, Zhao L, 19: 139–152. doi:10.1007/s10586-015-0512-2, , 2016

129. Land fragmentation under rapid urbanization: A cross-site analysis of Southwestern cities, Zhang S, Shrestha M, York AM, Boone CG, Prebyl TJ, et al, Harrington JA, Urban Ecosyst14: 429–455. doi:10.1007/s11252-011-0157-8, , 2011

130. A comparative study of recursive partitioning algorithm and analog concept learning system, Bin Lee S, Hyun Oh S., Expert Systems with Applications1: 403–416. doi:10.1016/0957-4174(90)90049-Z, , 1990

131. Cities as emergent models: the morphological logic of Manhattan and Barcelona In In andeds, Al Sayed K, Steen, J., Marcus L. and, Turner A, Hanna S. Koch D., Proceedings of the 7th International Space Syntax Symposiumpp. p. 1 Royal Institute of Technology (KTH): Stockholm, SwedenInternet Stockholm Sweden Royal Institute of Technology (KTH); 2009 cited 8 Oct 2023 p. 1. Available: http://www. sss7. org/Proceedings_list. html, , 2009

132. Discovering the homogeneous geographic domain of human perceptions from street view images, Hong Y, Wang J, Qian C, Liang X, et al, Guan Q, Yao Y, 212: 104125. doi:10.1016/j. landurbplan.2021.104125, , 2021

133. The impact of the COVID‐19 pandemic on child and adolescent development around the world, Fisher PA, Rao N, Child Dev92: e738–e748. doi:10.1111/cdev.13653, , 2021