A Deep Learning Framework for Emotion Recognition and Semantic Interpretation of Social Media Images in Urban Parks: The ULEAF Approach
Source abstract
Abstract
This study proposes the Urban Landscape Emotion Analysis Framework (ULEAF) based on images of urban parks shared on social media. This framework integrates an emotion recognition module driven by a convolutional neural network (ConvNeXt Tiny) with a semantic extraction module supported by multimodal semantic matching models (CLIP and DeepSentiBank ANP lexicon). It constructs a systematic analysis pathway from semantic understanding to emotional perception, effectively overcoming the limitations of traditional research methods. Results indicate that positive emotion images predominantly correlate with nature, health, and openness, while negative emotion images are closely associated with the characteristics of decay, abandonment, and oppression, as well as loneliness and calmness, estrangement and disharmony, and gloom and bleakness. Findings reveal trends consistent with prior research, further validating the stable association between urban landscape visual features and emotional perception. The analytical framework developed in this study facilitates the systematic revelation of semantic characteristics and affective perception mechanisms in large-scale urban park imagery, providing scientific reference for optimizing urban park landscapes and implementing emotion-oriented design.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Aesthetic Perception and Analysis · Urban Green Space and Health · Color perception and design
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
Department of Landscape Architecture, Kyungpook National University, Daegu 41566, Republic of Korea
College of Ecology and Environment, Inner Mongolia University, Hohhot 010010, China
References (77)
Show source-provided references
1. Efficiency of parks in mitigating urban heat island effect: An example from Addis Ababa10.1016/j.landurbplan.2013.12.008 ↗
2. Designing urban parks that ameliorate the effects of climate change10.1016/j.landurbplan.2015.02.006 ↗
3. Central Park as a model for social control: Urban parks, social class and leisure behavior in nineteenth-century America10.1080/00222216.1999.11949875 ↗
4. The role of urban parks for the sustainable city10.1016/j.landurbplan.2003.08.003 ↗
5. Physiological and psychological responses of young males during spring-time walks in urban parks10.1186/1880-6805-33-8 ↗
6. Preferences and emotion perceptions of ornamental plant species for green space designing among urban park users in Iran10.1016/j.ufug.2018.12.007 ↗
7. Zhang, L., Liu, S., and Liu, S. (2021). Mechanisms underlying the effects of landscape features of urban community parks on health-related feelings of users. Int. J. Environ. Res. Public Health, 18.10.3390/ijerph18157888 ↗
8. How do different types and landscape attributes of urban parks affect visitors’ positive emotions?10.1016/j.landurbplan.2022.104482 ↗
9. Public emotions and visual perception of the East Coast Park in Singapore: A deep learning method using social media data10.1016/j.ufug.2024.128285 ↗
10. Modeling emotion influence in image social networks10.1109/TAFFC.2015.2400917 ↗
11. Nguyen-Dinh, N., and Zhang, H. (2025). How Landscape Preferences and Emotions Shape Environmental Awareness: Perspectives from University Experiences. Sustainability, 17.10.3390/su17073161 ↗
12. Wen, B., and Burley, J.B. (2020). Expert opinion dimensions of rural landscape quality in Xiangxi, Hunan, China: Principal component analysis and factor analysis. Sustainability, 12.10.3390/su12041316 ↗
13. Survey research: Process and limitations10.12968/ijtr.2009.16.1.37935 ↗
14. Wang, Z., Jin, Y., Liu, Y., Li, D., and Zhang, B. (2018). Comparing social media data and survey data in assessing the attractiveness of Beijing Olympic Forest Park. Sustainability, 10.10.3390/su10020382 ↗
15. Are social media data and survey data consistent in measuring park visitation, park satisfaction, and their influencing factors? A case study in Shanghai10.1016/j.ufug.2023.127869 ↗
16. How to quantify multidimensional perception of urban parks? Integrating deep learning-based social media data analysis with questionnaire survey methods10.1016/j.ufug.2025.128754 ↗
17. Assessing and interpreting perceived park accessibility, usability and attractiveness through texts and images from social media10.1016/j.scs.2024.105619 ↗
18. Huang, Y., and Zheng, B. (2024). Social media users’ visual and emotional preferences of internet-famous sites in urban riverfront public spaces: A case study in Changsha, China. Land, 13.10.20944/preprints202403.1855.v1 ↗
19. Jia, J., Wu, S., Wang, X., Hu, P., Cai, L., and Tang, J. (November, January 29). Can we understand van gogh’s mood? learning to infer affects from images in social networks. Proceedings of the 20th ACM International Conference on Multimedia, Nara, Japan.
20. Chen, M., Arribas-Bel, D., and Singleton, A. (2020). Quantifying the characteristics of the local urban environment through geotagged flickr photographs and image recognition. ISPRS Int. J. Geo-Inf., 9.10.3390/ijgi9040264 ↗
21. The image of the City on social media: A comparative study using “Big Data” and “Small Data” methods in the Tri-City Region in Poland10.1016/j.landurbplan.2020.103977 ↗
22. Zhang, X., Xu, D., and Zhang, N. (2022). Research on landscape perception and visual attributes based on social media data—A case study on Wuhan University. Appl. Sci., 12.10.3390/app12168346 ↗
23. Krizhevsky, A., Sutskever, I., and Hinton, G.E. (2012). Imagenet classification with deep convolutional neural networks. Adv. Neural Inf. Process. Syst., 25, Available online: https://proceedings.neurips.cc/paper_files/paper/2012/file/c399862d3b9d6b76c8436e924a68c45b-Paper.pdf.
24. Deep convolutional neural networks for image classification: A comprehensive review10.1162/neco_a_00990 ↗
25. Multi-level region-based convolutional neural network for image emotion classification10.1016/j.neucom.2018.12.053 ↗
26. Yang, J., She, D., and Sun, M. (2017, January 19–25). Joint Image Emotion Classification and Distribution Learning via Deep Convolutional Neural Network. Proceedings of the IJCAI, Melbourne, Australia.10.24963/ijcai.2017/456 ↗
27. Classification of urban morphology with deep learning: Application on urban vitality10.1016/j.compenvurbsys.2021.101706 ↗
28. Street-Frontage-Net: Urban image classification using deep convolutional neural networks10.1080/13658816.2018.1555832 ↗
29. Li, J., Li, D., Savarese, S., and Hoi, S. (2023, January 23–29). Blip-2: Bootstrapping language-image pre-training with frozen image encoders and large language models. Proceedings of the International Conference on Machine Learning, Honolulu, HI, USA.
30. Radford, A., Kim, J.W., Hallacy, C., Ramesh, A., Goh, G., Agarwal, S., Sastry, G., Askell, A., Mishkin, P., and Clark, J. (2021, January 18–24). Learning transferable visual models from natural language supervision. Proceedings of the International Conference on Machine Learning, Online.
31. Perez, J., and Fusco, G. (2025). Streetscape Analysis with Generative AI (SAGAI): Vision-Language Assessment and Mapping of Urban Scenes. arXiv.10.2139/ssrn.5226191 ↗
32. Blečić, I., Saiu, V., and Trunfio, G.A. (2024, January 1–4). Enhancing urban walkability assessment with multimodal Large Language models. Proceedings of the International Conference on Computational Science and Its Applications, Hanoi, Vietnam.10.1007/978-3-031-65282-0_26 ↗
33. Exploring destination image of dark tourism via analyzing user generated photos: A deep learning approach
34. Chen, T., Borth, D., Darrell, T., and Chang, S.-F. (2014). Deepsentibank: Visual sentiment concept classification with deep convolutional neural networks. arXiv.
35. Zeng, X., Zhong, Y., Yang, L., Wei, J., and Tang, X. (2022). Analysis of Forest landscape preferences and emotional features of Chinese Forest recreationists based on deep learning of geotagged photos. Forests, 13.10.3390/f13060892 ↗
36. Yan, J., Yue, J., Zhang, J., and Qin, P. (2023). Research on spatio-temporal characteristics of tourists’ landscape perception and emotional experience by using photo data mining. Int. J. Environ. Res. Public Health, 20.10.3390/ijerph20053843 ↗
37. Zubić, N., Soldá, F., Sulser, A., and Scaramuzza, D. (2024). Limits of deep learning: Sequence modeling through the lens of complexity theory. arXiv.
38. Revealing disparities in different types of park visits based on cellphone signaling data in Guangzhou, China10.1016/j.jenvman.2023.119969 ↗
39. Examining the scale effect of nearby residential green space on residents’ BMI: A case study of Guangzhou, China10.1016/j.ufug.2024.128329 ↗
40. Muhammad, R., Zhao, Y., and Liu, F. (2019). Spatiotemporal analysis to observe gender based check-in behavior by using social media big data: A case study of Guangzhou, China. Sustainability, 11.10.3390/su11102822 ↗
41. Fine-grained assessment of greenspace satisfaction at regional scale using content analysis of social media and machine learning10.1016/j.scitotenv.2021.145908 ↗
42. Identifying the integrated visual characteristics of greenway landscape: A focus on human perception10.1016/j.scs.2023.104937 ↗
43. Exploring temporal and spatial patterns and nonlinear driving mechanism of park perceptions: A multi-source big data study10.1016/j.scs.2024.106083 ↗
44. Woo, S., Debnath, S., Hu, R., Chen, X., Liu, Z., Kweon, I.S., and Xie, S. (2023, January 17–24). Convnext v2: Co-designing and scaling convnets with masked autoencoders. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, Vancouver, BC, Canada.10.1109/CVPR52729.2023.01548 ↗
45. Yu, W., Zhou, P., Yan, S., and Wang, X. (2024, January 16–22). Inceptionnext: When inception meets convnext. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, Seattle, WA, USA.10.1109/CVPR52733.2024.00542 ↗
46. Min, Z., Ge, Q., and Tai, C. (2022). Why the pseudo label based semi-supervised learning algorithm is effective?. arXiv.
47. Lee, D.-H. (2013, January 28). Pseudo-label: The simple and efficient semi-supervised learning method for deep neural networks. Proceedings of the Workshop on Challenges in Representation Learning, ICML, Atlanta, GA, USA.
48. Liu, Z., Mao, H., Wu, C.-Y., Feichtenhofer, C., Darrell, T., and Xie, S. (2022, January 18–24). A convnet for the 2020s. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, New Orleans, LA, USA.10.1109/CVPR52688.2022.01167 ↗
49. Mikolov, T., Chen, K., Corrado, G., and Dean, J. (2013). Efficient estimation of word representations in vector space. arXiv.
50. An example of the use of factor analysis and cluster analysis in groundwater chemistry interpretation10.1016/0022-1694(78)90011-2 ↗
51. A beginner’s guide to factor analysis: Focusing on exploratory factor analysis10.20982/tqmp.09.2.p079 ↗
52. Logistic regression10.1161/CIRCULATIONAHA.106.682658 ↗
53. Unveiling the Positivity Bias on Social Media: A Registered Experimental Study on Facebook, Instagram, and X10.1525/collabra.132410 ↗
54. Exposure to the positivity bias and adolescents’ differential longitudinal links with social comparison, inspiration and envy depending on social media literacy10.1007/s12144-022-03893-3 ↗
55. Urban planning in the age of large language models: Assessing OpenAI o1’s performance and capabilities across 556 tasks10.1016/j.compenvurbsys.2025.102332 ↗
56. Urban AI: Understanding the emerging role of artificial intelligence in smart cities10.1007/s00146-022-01537-5 ↗
57. Demographic groups’ differences in visual preference for vegetated landscapes in urban green space10.1016/j.scs.2016.10.010 ↗
58. Perceptions of parks and urban derelict land by landscape planners and residents10.1016/j.ufug.2012.04.001 ↗
59. Associations between park features, park satisfaction and park use in a multi-ethnic deprived urban area10.1016/j.ufug.2019.126485 ↗
60. Empirical study of landscape types, landscape elements and landscape components of the urban park promoting physiological and psychological restoration10.1016/j.ufug.2019.126488 ↗
61. Which affects park satisfaction more, environmental features or spatial pattern?10.1007/s10980-025-02076-2 ↗
62. Quantifying emotional differences in urban green spaces extracted from photos on social networking sites: A study of 34 parks in three cities in northern China10.1016/j.ufug.2021.127133 ↗
63. Knez, I., Ode Sang, Å., Gunnarsson, B., and Hedblom, M. (2018). Wellbeing in urban greenery: The role of naturalness and place identity. Front. Psychol., 9.10.3389/fpsyg.2018.00491 ↗
64. Bressane, A., Silva, M.B., Goulart, A.P.G., and Medeiros, L.C.d.C. (2024). Understanding how green space naturalness impacts public well-being: Prospects for designing healthier cities. Int. J. Environ. Res. Public Health, 21.10.3390/ijerph21050585 ↗
65. Liu, R., and Xiao, J. (2021). Factors affecting users’ satisfaction with urban parks through online comments data: Evidence from Shenzhen, China. Int. J. Environ. Res. Public Health, 18.10.3390/ijerph18010253 ↗
66. Eliciting users’ preferences and values in urban parks: Evidence from analyzing social media data from Hong Kong10.1016/j.ufug.2021.127172 ↗
67. Effects of flowering, foliation, and autumn colors on preference and restorative potential for designed digital landscape models10.1177/0013916518811424 ↗
68. Consensus in visual preferences: The effects of aesthetic quality and landscape types10.1016/j.ufug.2016.09.005 ↗
69. Exposure to restorative environments helps restore attentional capacity10.1016/j.jenvp.2005.07.001 ↗
70. Social interactions in urban parks: Stimulating social cohesion?10.1016/j.ufug.2009.11.003 ↗
71. Interracial interaction, park ownership, belonging, community asset, and perceived provision of cultural ecosystem services10.1016/j.ufug.2024.128551 ↗
72. Welch, D., Shepherd, D., Dirks, K., Tan, M.Y., and Coad, G. (2019). Use of creative writing to develop a semantic differential tool for assessing soundscapes. Front. Psychol., 9.10.3389/fpsyg.2018.02698 ↗
73. Herranz-Pascual, K., Aspuru, I., Iraurgi, I., Santander, Á., Eguiguren, J.L., and García, I. (2019). Going beyond quietness: Determining the emotionally restorative effect of acoustic environments in urban open public spaces. Int. J. Environ. Res. Public Health, 16.10.3390/ijerph16071284 ↗
74. The effect of visibility on green space recovery, perception and preference10.1016/j.tfp.2024.100538 ↗
75. Using social media data in understanding site-scale landscape architecture design: Taking Seattle Freeway Park as an example10.1080/01426397.2020.1736994 ↗
76. Users of the world, unite! The challenges and opportunities of Social Media10.1016/j.bushor.2009.09.003 ↗
77. An integrated deep learning approach for assessing the visual qualities of built environments utilizing street view images10.1016/j.engappai.2023.107805 ↗