{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,1,2]],"date-time":"2026-01-02T07:43:26Z","timestamp":1767339806812,"version":"build-2065373602"},"reference-count":49,"publisher":"MDPI AG","issue":"9","license":[{"start":{"date-parts":[[2022,8,29]],"date-time":"2022-08-29T00:00:00Z","timestamp":1661731200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"Sichuan University","award":["2021SCU12125"],"award-info":[{"award-number":["2021SCU12125"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IJGI"],"abstract":"<jats:p>Urban-safety perception is crucial for urban planning and pedestrian street preference studies. With the development of deep learning and the availability of high-resolution street images, the use of artificial intelligence methods to deal with urban-safety perception has been considered adequate by many researchers. However, most current methods are based on the feature-extraction capability of convolutional neural networks (CNNs) with large-scale annotated data for training, mainly aimed at providing a regression or classification model. There remains a lack of interpretable and complete evaluation systems for urban-safety perception. To improve the interpretability of evaluation models and achieve human-like safety perception, we proposed a complete decision-making framework based on reinforcement learning (RL). We developed a novel feature-extraction module, a scalable visual computational model based on visual semantic and functional features that could fully exploit the knowledge of domain experts. Furthermore, we designed the RL module\u2014comprising a combination of a Markov decision process (MDP)-based street-view observation environment and an intelligent agent trained using a deep reinforcement-learning (DRL) algorithm\u2014to achieve human-level perception abilities. Experimental results using our crowdsourced dataset showed that the framework achieved satisfactory prediction performance and excellent visual interpretability.<\/jats:p>","DOI":"10.3390\/ijgi11090465","type":"journal-article","created":{"date-parts":[[2022,8,29]],"date-time":"2022-08-29T21:01:31Z","timestamp":1661806891000},"page":"465","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":6,"title":["A Complete Reinforcement-Learning-Based Framework for Urban-Safety Perception"],"prefix":"10.3390","volume":"11","author":[{"given":"Yaxuan","family":"Wang","sequence":"first","affiliation":[{"name":"School of Computer Science, Sichuan University, Chengdu 610207, China"}]},{"given":"Zhixin","family":"Zeng","sequence":"additional","affiliation":[{"name":"School of Cyber Science and Engineering, Sichuan University, Chengdu 610207, China"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8659-0464","authenticated-orcid":false,"given":"Qiushan","family":"Li","sequence":"additional","affiliation":[{"name":"Institute for Disaster Management and Reconstruction, Sichuan University, Chengdu 610207, China"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-2208-0833","authenticated-orcid":false,"given":"Yingrui","family":"Deng","sequence":"additional","affiliation":[{"name":"School of Computer Science, Sichuan University, Chengdu 610207, China"}]}],"member":"1968","published-online":{"date-parts":[[2022,8,29]]},"reference":[{"key":"ref_1","first-page":"343","article-title":"Environment and Crime in the Inner City: Does Vegetation Reduce Crime?","volume":"33","author":"Kuo","year":"2001","journal-title":"Environ. Behav."},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"262","DOI":"10.1016\/j.landurbplan.2012.03.010","article-title":"The relationship between tree canopy and crime rates across an urban\u2013rural gradient in the greater Baltimore region","volume":"106","author":"Troy","year":"2012","journal-title":"Landsc. Urban Plan."},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"2624","DOI":"10.1109\/TVCG.2014.2346446","article-title":"City Forensics: Using Visual Elements to Predict Non-Visual City Attributes","volume":"20","author":"Arietta","year":"2014","journal-title":"IEEE Trans. Vis. Comput. Graph."},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"101631","DOI":"10.1016\/j.compenvurbsys.2021.101631","article-title":"Using Google Street View imagery to capture micro built environment characteristics in drug places, compared with street robbery","volume":"88","author":"Zhou","year":"2021","journal-title":"Comput. Environ. Urban Syst."},{"key":"ref_5","unstructured":"Kelling, G.L., and Coles, C.M. (1997). Fixing Broken Windows: Restoring Order and Reducing Crime in Our Communities, Simon & Schuster. A Touchstone Book."},{"key":"ref_6","unstructured":"Quercia, D., O\u2019Hare, N.K., and Cramer, H. (February, January 25). Aesthetic capital: What makes london look beautiful, quiet, and happy?. Proceedings of the 17th ACM Conference on Computer Supported Cooperative Work & Social Computing, Portland, OR, USA."},{"key":"ref_7","doi-asserted-by":"crossref","unstructured":"Naik, N., Philipoom, J., Raskar, R., and Hidalgo, C. (2014, January 23\u201328). Streetscore\u2014Predicting the Perceived Safety of One Million Streetscapes. Proceedings of the 2014 IEEE Conference on Computer Vision and Pattern Recognition Workshops, Columbus, OH, USA.","DOI":"10.1109\/CVPRW.2014.121"},{"key":"ref_8","doi-asserted-by":"crossref","unstructured":"Cheng, L., Chu, S., Zong, W., Li, S., Wu, J., and Li, M. (2017). Use of Tencent Street View Imagery for Visual Perception of Streets. ISPRS Int. J. Geo-Inf., 6.","DOI":"10.3390\/ijgi6090265"},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"153","DOI":"10.1016\/j.compenvurbsys.2018.05.005","article-title":"Representing place locales using scene elements","volume":"71","author":"Zhang","year":"2018","journal-title":"Comput. Environ. Urban Syst."},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"26","DOI":"10.1016\/j.jenvp.2012.09.003","article-title":"Architectural variation, building height, and the restorative quality of urban residential streetscapes","volume":"33","author":"Lindal","year":"2013","journal-title":"J. Environ. Psychol."},{"key":"ref_11","doi-asserted-by":"crossref","unstructured":"Dubey, A., Naik, N., Parikh, D., Raskar, R., and Hidalgo, C.A. (2016). Deep Learning the City: Quantifying Urban Perception At A Global Scale. arXiv.","DOI":"10.1007\/978-3-319-46448-0_12"},{"key":"ref_12","doi-asserted-by":"crossref","unstructured":"Liu, X., Chen, Q., Zhu, L., Xu, Y., and Lin, L. (2017, January 23\u201327). Place-centric Visual Urban Perception with Deep Multi-instance Regression. Proceedings of the 25th ACM International Conference on Multimedia, Mountain View, CA, USA.","DOI":"10.1145\/3123266.3123271"},{"key":"ref_13","doi-asserted-by":"crossref","unstructured":"Sch\u00f6lkopf, B., Platt, J., and Hofmann, T. (2007). TrueSkill: A Bayesian Skill Rating System. Advances in Neural Information Processing Systems 19, The MIT Press.","DOI":"10.7551\/mitpress\/7503.001.0001"},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"529","DOI":"10.1038\/nature14236","article-title":"Human-level control through deep reinforcement learning","volume":"518","author":"Mnih","year":"2015","journal-title":"Nature"},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1016\/j.robot.2019.01.003","article-title":"Advanced planning for autonomous vehicles using reinforcement learning and deep inverse reinforcement learning","volume":"114","author":"You","year":"2019","journal-title":"Robot. Auton. Syst."},{"key":"ref_16","doi-asserted-by":"crossref","first-page":"484","DOI":"10.1038\/nature16961","article-title":"Mastering the game of Go with deep neural networks and tree search","volume":"529","author":"Silver","year":"2016","journal-title":"Nature"},{"key":"ref_17","doi-asserted-by":"crossref","first-page":"679","DOI":"10.1512\/iumj.1957.6.56038","article-title":"A Markovian Decision Process","volume":"6","author":"Bellman","year":"1957","journal-title":"Indiana Univ. Math. J."},{"key":"ref_18","unstructured":"Lynch, K. (2008). The Image of the City, MIT Press. [33rd ed.]. Publication of the Joint Center for Urban Studies."},{"key":"ref_19","unstructured":"Jacobs, J. (1992). The Death and Life of Great American Cities, Vintage Books."},{"key":"ref_20","doi-asserted-by":"crossref","first-page":"127","DOI":"10.1016\/j.ufug.2013.01.005","article-title":"Perceived personal safety in relation to urban woodland vegetation\u2014A review","volume":"12","author":"Jansson","year":"2013","journal-title":"Urban For. Urban Green."},{"key":"ref_21","doi-asserted-by":"crossref","first-page":"558","DOI":"10.1136\/jech.2004.028399","article-title":"Multilevel modelling of built environment characteristics related to neighbourhood walking activity in older adults","volume":"59","author":"Li","year":"2005","journal-title":"J. Epidemiol. Community Health"},{"key":"ref_22","doi-asserted-by":"crossref","first-page":"2076","DOI":"10.2105\/AJPH.2006.097154","article-title":"Association Between Fear of Crime and Mental Health and Physical Functioning","volume":"97","author":"Stafford","year":"2007","journal-title":"Am. J. Public Health"},{"key":"ref_23","doi-asserted-by":"crossref","first-page":"832","DOI":"10.1093\/bjc\/azp033","article-title":"Public Health and Fear of Crime: A Prospective Cohort Study","volume":"49","author":"Jackson","year":"2009","journal-title":"Br. J. Criminol."},{"key":"ref_24","doi-asserted-by":"crossref","first-page":"113","DOI":"10.1016\/j.compenvurbsys.2017.06.003","article-title":"A machine learning-based method for the large-scale evaluation of the qualities of the urban environment","volume":"65","author":"Liu","year":"2017","journal-title":"Comput. Environ. Urban Syst."},{"key":"ref_25","doi-asserted-by":"crossref","first-page":"83","DOI":"10.1016\/j.compenvurbsys.2017.08.001","article-title":"Built environment and violent crime: An environmental audit approach using Google Street View","volume":"66","author":"He","year":"2017","journal-title":"Comput. Environ. Urban Syst."},{"key":"ref_26","doi-asserted-by":"crossref","first-page":"46","DOI":"10.1016\/j.irle.2007.12.002","article-title":"Who wants safer cities? Perceptions of public safety and attitudes to migrants among China\u2019s urban population","volume":"28","author":"Nielsen","year":"2008","journal-title":"Int. Rev. Law Econ."},{"key":"ref_27","doi-asserted-by":"crossref","first-page":"S18","DOI":"10.1016\/j.ypmed.2009.08.015","article-title":"Do you see what I see?\u2014Correlates of multidimensional measures of neighborhood types and perceived physical activity\u2013related neighborhood barriers and facilitators for urban youth","volume":"50","author":"Yan","year":"2010","journal-title":"Prev. Med."},{"key":"ref_28","unstructured":"Porzi, L., Rota Bul\u00f2, S., Lepri, B., and Ricci, E. (2022, January 10\u201314). Predicting and Understanding Urban Perception with Convolutional Neural Networks. Proceedings of the 23rd ACM International Conference on Multimedia, Lisboa, Portugal."},{"key":"ref_29","doi-asserted-by":"crossref","first-page":"494","DOI":"10.1007\/978-3-319-10599-4_32","article-title":"Learning High-Level Judgments of Urban Perception","volume":"Volume 8694","author":"Fleet","year":"2014","journal-title":"Computer Vision\u2014ECCV 2014"},{"key":"ref_30","doi-asserted-by":"crossref","unstructured":"Acosta, S., and Camargo, J.E. (2019). Predicting city safety perception based on visual image content. arXiv.","DOI":"10.1007\/978-3-030-13469-3_21"},{"key":"ref_31","doi-asserted-by":"crossref","unstructured":"Salesses, P., Schechtner, K., and Hidalgo, C.A. (2013). The Collaborative Image of The City: Mapping the Inequality of Urban Perception. PLoS ONE, 8.","DOI":"10.1371\/journal.pone.0068400"},{"key":"ref_32","doi-asserted-by":"crossref","unstructured":"Kaur, T., and Gandhi, T.K. (2019, January 19\u201321). Automated Brain Image Classification Based on VGG-16 and Transfer Learning. Proceedings of the 2019 International Conference on Information Technology (ICIT), Bhubaneswar, India.","DOI":"10.1109\/ICIT48102.2019.00023"},{"key":"ref_33","doi-asserted-by":"crossref","unstructured":"Sutton, R.S. (1992). Simple Statistical Gradient-Following Algorithms for Connectionist Reinforcement Learning. Reinforcement Learning, Springer.","DOI":"10.1007\/978-1-4615-3618-5"},{"key":"ref_34","unstructured":"Solla, S., Leen, T., and M\u00fcl-ler, K. (1999). Actor-Critic Algorithms. Advances in Neural Information Processing Systems, MIT Press."},{"key":"ref_35","unstructured":"Sutton, R.S., McAllester, D.A., Singh, S.P., and Mansour, Y. (1999). Policy Gradient Methods for Reinforcement Learning with Function Approximation 7, AT&T Labs."},{"key":"ref_36","unstructured":"Solla, S., Leen, T., and M\u00fcller, K. (1999). Policy Gradient Methods for Reinforcement Learning with Function Approximation. Advances in Neural Information Processing Systems, MIT Press."},{"key":"ref_37","doi-asserted-by":"crossref","first-page":"204166952091305","DOI":"10.1177\/2041669520913052","article-title":"Seven Myths on Crowding and Peripheral Vision","volume":"11","author":"Strasburger","year":"2020","journal-title":"i-Perception"},{"key":"ref_38","doi-asserted-by":"crossref","unstructured":"Sch\u00f6lkopf, B., Platt, J., and Hofmann, T. (2007). Learning to Rank with Nonsmooth Cost Functions. Advances in Neural Information Processing Systems 19, The MIT Press.","DOI":"10.7551\/mitpress\/7503.001.0001"},{"key":"ref_39","doi-asserted-by":"crossref","first-page":"589","DOI":"10.1177\/0013916583155003","article-title":"Adult Viewers\u2019 Preferences in Residential Scenes: A Study of the Relationship of Environmental Attributes to Preference","volume":"15","author":"Nasar","year":"1983","journal-title":"Environ. Behav."},{"key":"ref_40","doi-asserted-by":"crossref","unstructured":"Altman, I., and Wohlwill, J.F. (1976). Environmental Aesthetics: The Environment as a Source of Affect. Human Behavior and Environment, Springer US.","DOI":"10.1007\/978-1-4684-2550-5"},{"key":"ref_41","doi-asserted-by":"crossref","first-page":"65","DOI":"10.1080\/13574800802451155","article-title":"Measuring the Unmeasurable: Urban Design Qualities Related to Walkability","volume":"14","author":"Ewing","year":"2009","journal-title":"J. Urban Des."},{"key":"ref_42","first-page":"408","article-title":"Understanding Place Characteristics in Geographic Contexts through Graph Convolutional Neural Networks","volume":"110","author":"Zhu","year":"2020","journal-title":"Ann. Am. Assoc. Geogr."},{"key":"ref_43","unstructured":"Xie, E., Wang, W., Yu, Z., Anandkumar, A., Alvarez, J.M., and Luo, P. (2021). SegFormer: Simple and Efficient Design for Semantic Segmentation with Transformers. arXiv."},{"key":"ref_44","doi-asserted-by":"crossref","first-page":"173","DOI":"10.1007\/978-3-030-58539-6_11","article-title":"Object-Contextual Representations for Semantic Segmentation","volume":"Volume 12351","author":"Vedaldi","year":"2020","journal-title":"Computer Vision\u2014ECCV 2020"},{"key":"ref_45","doi-asserted-by":"crossref","first-page":"833","DOI":"10.1007\/978-3-030-01234-2_49","article-title":"Encoder-Decoder with Atrous Separable Convolution for Semantic Image Segmentation","volume":"Volume 11211","author":"Ferrari","year":"2018","journal-title":"Computer Vision\u2014ECCV 2018"},{"key":"ref_46","doi-asserted-by":"crossref","unstructured":"Cordts, M., Omran, M., Ramos, S., Rehfeld, T., Enzweiler, M., Benenson, R., Franke, U., Roth, S., and Schiele, B. (2016, January 27\u201330). The Cityscapes Dataset for Semantic Urban Scene Understanding. Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Las Vegas, NV, USA.","DOI":"10.1109\/CVPR.2016.350"},{"key":"ref_47","doi-asserted-by":"crossref","unstructured":"Ren, Z., Wang, X., Zhang, N., Lv, X., and Li, L.-J. (2017, January 21\u201326). Deep Reinforcement Learning-Based Image Captioning with Embedding Reward. Proceedings of the 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Honolulu, HI, USA.","DOI":"10.1109\/CVPR.2017.128"},{"key":"ref_48","doi-asserted-by":"crossref","unstructured":"Van Hasselt, H., Guez, A., and Silver, D. (2016, January 12\u201317). Deep Reinforcement Learning with Double Q-Learning. Proceedings of the Thirtieth AAAI Conference on Artificial Intelligence, Phoenix, AR, USA.","DOI":"10.1609\/aaai.v30i1.10295"},{"key":"ref_49","unstructured":"Wang, Z., Schaul, T., Hessel, M., van Hasselt, H., Lanctot, M., and de Freitas, N. (2016, January 19\u201324). Dueling Network Architectures for Deep Reinforcement Learning. Proceedings of the 33rd International Conference on Machine Learning, New York, NY, USA."}],"container-title":["ISPRS International Journal of Geo-Information"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/2220-9964\/11\/9\/465\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,11]],"date-time":"2025-10-11T00:19:38Z","timestamp":1760141978000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/2220-9964\/11\/9\/465"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2022,8,29]]},"references-count":49,"journal-issue":{"issue":"9","published-online":{"date-parts":[[2022,9]]}},"alternative-id":["ijgi11090465"],"URL":"https:\/\/doi.org\/10.3390\/ijgi11090465","relation":{},"ISSN":["2220-9964"],"issn-type":[{"type":"electronic","value":"2220-9964"}],"subject":[],"published":{"date-parts":[[2022,8,29]]}}}