Mapping human perception of urban landscape from street-view images: A deep-learning approach

Wei, Jing-Xian; Yue, Wenze; Li, Mengmeng; Gao, Jiabin

doi:10.1016/j.jag.2022.102886

Cited by 33 publications

(18 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Interviews and surveys are powerful tools for sound decision-making because they give clear insights into users' preferences. Nevertheless, they are time-consuming to administer and often susceptible to biased interpretation due to insufficient samples that may prevent their generalization [63]. In this sense, a significant limitation of this study was the limited size and the lack of representativeness of the study in terms of foreigners and regular users of the marina.…”

Section: Limitations and Future Researchmentioning

confidence: 97%

See 1 more Smart Citation

Landscape Values in a Marina in Granada (Spain): Enhancing Landscape Management through Public Participation

Martín

Piqueras

2023

Land

View full text Add to dashboard Cite

Landscape values are related to the attributes that people assign to a perceived landscape. They reflect marina user perceptions, thus representing a feedback tool for marina managers to use to verify the degree of user satisfaction. This study focused on identifying and assessing a marina’s landscape values. We took Marina del Este (Granada, Spain) as a case study. We considered interviews and a questionnaire to devise methods to enhance the participation of stakeholders and users. First, the SWOT analysis from stakeholder interviews enabled us to collect management’s perceptions. Second, the survey gathered the marina’s landscape values, comprising 104 respondents from visitors and users. ANOVA and PCA methods were applied to check the suitability of the values. The results showed that the marina should be in keeping with an atmosphere of tranquility and well-being. Nevertheless, there was a need to improve values related to nautical tourism, such as hospitality and maintenance, dealing with the lack of space and an excess of urbanization in the surroundings. Marina managers should consider these outcomes and analyze the points of improvement to establish the causes of these disagreements and propose solutions for the established management model. The perception of stakeholders and users can enable more consensual policies with greater levels of acceptance and involvement.

show abstract

Section: Limitations and Future Researchmentioning

confidence: 97%

“…The perception of landscape is a liquid demand insofar as it becomes a service when people request it. In addition, this perception is based on personal experiences, preferences, or abilities [36,63]. This variability in perception or lack of definition makes it difficult to generalize landscape values.…”

Section: Learnings From Landscape Valuesmentioning

confidence: 99%

Landscape Values in a Marina in Granada (Spain): Enhancing Landscape Management through Public Participation

Martín

Piqueras

2023

Land

View full text Add to dashboard Cite

show abstract

“…Zhang, Siriaraya, et al (2018) analyzed the correlation between human perception and various elements in urban streetscapes (150 types of objects) using this dataset. Wei et al (2022) investigated the relationship between land use and four human perception attributes: security, depression, vitality, and aesthetics. The current research focuses primarily on creating appealing urban spatial designs.…”

Section: Human Perception Measures Based On Street-view Imagesmentioning

confidence: 99%

“…Based on this, Dubey et al (2016) built a more comprehensive and widely surveyed Place Pulse 2.0 (PP2.0) dataset that covers more cities and users, thus providing a more generalizable data source for measuring urban perceptions. Scholars have used this dataset to map subjective perceptions of urban streets and explore the relationship between spatial land use and human perceptions (Wei et al, 2022). Such research underscores the significance of our contribution to high-quality urban path planning.…”

mentioning

confidence: 99%

Urban high‐quality navigation path planning that integrates human emotion perception learning

Liu,

Long,

Zhang

et al. 2023

Transactions in GIS

View full text Add to dashboard Cite

Geographic information system navigation services are now incorporating psychological well‐being as a factor when devising navigation routes. However, challenges such as limited data, method generalizability, and subjective human perception remain unresolved. Therefore, a general humanized path‐navigation method that effectively quantifies human emotional perception demands is required. In this study, we designed a deep learning model using a large, crowdsourced dataset to predict emotional responses to street‐view images. Our method enhances urban path planning, thus providing comprehensive emotional benefits. Our approach and several goal‐oriented methods were applied in Nanjing, and the findings were compared via comparative analyses and questionnaire surveys. The results confirmed that our proposed method outperforms utilitarian goal‐driven path planning methods in terms of subjective perception. This study provides a widely available technique for high‐quality navigation planning that meets psychological perception needs and offers a valuable guidance for the research on humanized geographic information services.

show abstract

“…Previous studies on human cognition have mainly focused on individuals because human understanding is highly dependent on individual experiences. Likewise, personal experiences play a significant role in landscape perception, and the spatial properties of landscapes can limit the co-production of knowledge operating across scales and sectors (Wei et al, 2022). However, advanced computer vision and deep learning methods enable easy mapping and analysis of the human perception of the urban landscape, helping to understand the relationship between human perception and perception.…”

Section: Computer Vision and Visual Explanations From Deep Learningmentioning

confidence: 99%

Resident's Satisfaction in Street Landscape Using the Immersive Virtual Environment-Based Eye-Tracking Technique and Deep Learning Model

Han

Lee

2022

Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci.

View full text Add to dashboard Cite

Abstract. Virtual reality technology provides a significant clue to understanding the human visual perception process by enabling the interaction between humans and computers. In addition, deep learning techniques in the visual field provide analysis methods for image classification, processing, and segmentation. This study reviewed the applicability of gaze movement and deep learning-based satisfaction evaluation on the landscape using an immersive virtual reality-based eye-tracking device. To this end, the following research procedures were established and analysed. First, the gaze movement of the test taker is measured using an immersive virtual environment-based eye tracker. The relationship between the gaze movement pattern of the test taker and the satisfaction evaluation result for the landscape image is analysed. Second, using the Convolutional Neural Networks (CNN)-based Class Activation Map (CAM) technique, a model for estimating the satisfaction evaluation result is constructed, and the gaze pattern of the test taker is derived. Third, we compare and analyse the similarity between the gaze heat map derived through the immersive virtual environment-based gaze tracker and the heat map generated by CAM. This study suggests the applicability of urban environment technology and deep learning methods to understand landscape planning factors that affect urban landscape satisfaction, resulting from the three-dimensional and immediate visual cognitive activity.

show abstract

Mapping human perception of urban landscape from street-view images: A deep-learning approach

Cited by 33 publications

References 28 publications

Landscape Values in a Marina in Granada (Spain): Enhancing Landscape Management through Public Participation

Landscape Values in a Marina in Granada (Spain): Enhancing Landscape Management through Public Participation

Urban high‐quality navigation path planning that integrates human emotion perception learning

Resident's Satisfaction in Street Landscape Using the Immersive Virtual Environment-Based Eye-Tracking Technique and Deep Learning Model

Contact Info

Product

Resources

About