Predicting the effect of street environment on residents' mood states in large urban areas using machine learning and street view images

AbstractUrban area mapping is an important application of remote sensing which aims at both estimation and change in land cover under the urban area. A major challenge being faced while analyzing Synthetic Aperture Radar (SAR) based remote sensing data is that there is a lot of similarity between highly vegetated urban areas and oriented urban targets with that of actual vegetation. This similarity between some urban areas and vegetation leads to misclassification of the urban area into forest cover. The present work is a precursor study for the dual-frequency L and S-band NASA-ISRO Synthetic Aperture Radar (NISAR) mission and aims at minimizing the misclassification of such highly vegetated and oriented urban targets into vegetation class with the help of deep learning. In this study, three machine learning algorithms Random Forest (RF), K-Nearest Neighbour (KNN), and Support Vector Machine (SVM) have been implemented along with a deep learning model DeepLabv3+ for semantic segmentation of Polarimetric SAR (PolSAR) data. It is a general perception that a large dataset is required for the successful implementation of any deep learning model but in the field of SAR based remote sensing, a major issue is the unavailability of a large benchmark labeled dataset for the implementation of deep learning algorithms from scratch. In current work, it has been shown that a pre-trained deep learning model DeepLabv3+ outperforms the machine learning algorithms for land use and land cover (LULC) classification task even with a small dataset using transfer learning. The highest pixel accuracy of 87.78% and overall pixel accuracy of 85.65% have been achieved with DeepLabv3+ and Random Forest performs best among the machine learning algorithms with overall pixel accuracy of 77.91% while SVM and KNN trail with an overall accuracy of 77.01% and 76.47% respectively. The highest precision of 0.9228 is recorded for the urban class for semantic segmentation task with DeepLabv3+ while machine learning algorithms SVM and RF gave comparable results with a precision of 0.8977 and 0.8958 respectively.

Download Full-text

Assessing the Impact of Street-View Greenery on Fear of Neighborhood Crime in Guangzhou, China

International Journal of Environmental Research and Public Health ◽

10.3390/ijerph18010311 ◽

2021 ◽

Vol 18 (1) ◽

pp. 311

Author(s):

Fengrui Jing ◽

Lin Liu ◽

Suhong Zhou ◽

Jiangyu Song ◽

Linsen Wang ◽

...

Keyword(s):

Social Factors ◽

Urban Areas ◽

Fear Of Crime ◽

Green Space ◽

Learning Approach ◽

Perceived Safety ◽

Street View ◽

The Impact ◽

The Relationship ◽

Physical Incivilities

Previous literature has examined the relationship between the amount of green space and perceived safety in urban areas, but little is known about the effect of street-view neighborhood greenery on perceived neighborhood safety. Using a deep learning approach, we derived greenery from a massive set of street view images in central Guangzhou. We further tested the relationships and mechanisms between street-view greenery and fear of crime in the neighborhood. Results demonstrated that a higher level of neighborhood street-view greenery was associated with a lower fear of crime, and its relationship was mediated by perceived physical incivilities. While increasing street greenery of the micro-environment may reduce fear of crime, this paper also suggests that social factors should be considered when designing ameliorative programs.

Download Full-text

Facilitating fine-grained intra-urban dengue forecasting by integrating urban environments measured from street-view images

Infectious Diseases of Poverty ◽

10.1186/s40249-021-00824-5 ◽

2021 ◽

Vol 10 (1) ◽

Author(s):

Kang Liu ◽

Ling Yin ◽

Meng Zhang ◽

Min Kang ◽

Ai-Ping Deng ◽

...

Keyword(s):

Machine Learning ◽

High Risk ◽

Feature Vector ◽

Urban Environments ◽

Environmental Feature ◽

Fine Grained ◽

Environmental Features ◽

Street View ◽

The City ◽

Dengue Epidemic

Abstract Background Dengue fever (DF) is a mosquito-borne infectious disease that has threatened tropical and subtropical regions in recent decades. An early and targeted warning of a dengue epidemic is important for vector control. Current studies have primarily determined weather conditions to be the main factor for dengue forecasting, thereby neglecting that environmental suitability for mosquito breeding is also an important factor, especially in fine-grained intra-urban settings. Considering that street-view images are promising for depicting physical environments, this study proposes a framework for facilitating fine-grained intra-urban dengue forecasting by integrating the urban environments measured from street-view images. Methods The dengue epidemic that occurred in 167 townships of Guangzhou City, China, between 2015 and 2019 was taken as a study case. First, feature vectors of street-view images acquired inside each township were extracted by a pre-trained convolutional neural network, and then aggregated as an environmental feature vector of the township. Thus, townships with similar physical settings would exhibit similar environmental features. Second, the environmental feature vector is combined with commonly used features (e.g., temperature, rainfall, and past case count) as inputs to machine-learning models for weekly dengue forecasting. Results The performance of machine-learning forecasting models (i.e., MLP and SVM) integrated with and without environmental features were compared. This indicates that models integrating environmental features can identify high-risk urban units across the city more precisely than those using common features alone. In addition, the top 30% of high-risk townships predicted by our proposed methods can capture approximately 50–60% of dengue cases across the city. Conclusions Incorporating local environments measured from street view images is effective in facilitating fine-grained intra-urban dengue forecasting, which is beneficial for conducting spatially precise dengue prevention and control.

Download Full-text

Prediction of Hourly Effect of Land Use on Crime

ISPRS International Journal of Geo-Information ◽

10.3390/ijgi8010016 ◽

2018 ◽

Vol 8 (1) ◽

pp. 16 ◽

Cited By ~ 4

Author(s):

Irina Matijosaitiene ◽

Peng Zhao ◽

Sylvain Jaume ◽

Joseph Gilkey Jr

Keyword(s):

Machine Learning ◽

Land Use ◽

Logistic Regression ◽

Hot Spots ◽

Urban Areas ◽

Prediction Models ◽

Hot Spot ◽

Time Range ◽

Land Uses ◽

Police Departments

Predicting the exact urban places where crime is most likely to occur is one of the greatest interests for Police Departments. Therefore, the goal of the research presented in this paper is to identify specific urban areas where a crime could happen in Manhattan, NY for every hour of a day. The outputs from this research are the following: (i) predicted land uses that generates the top three most committed crimes in Manhattan, by using machine learning (random forest and logistic regression), (ii) identifying the exact hours when most of the assaults are committed, together with hot spots during these hours, by applying time series and hot spot analysis, (iii) built hourly prediction models for assaults based on the land use, by deploying logistic regression. Assault, as a physical attack on someone, according to criminal law, is identified as the third most committed crime in Manhattan. Land use (residential, commercial, recreational, mixed use etc.) is assigned to every area or lot in Manhattan, determining the actual use or activities within each particular lot. While plotting assaults on the map for every hour, this investigation has identified that the hot spots where assaults occur were ‘moving’ and not confined to specific lots within Manhattan. This raises a number of questions: Why are hot spots of assaults not static in an urban environment? What makes them ‘move’—is it a particular urban pattern? Is the ‘movement’ of hot spots related to human activities during the day and night? Answering these questions helps to build the initial frame for assault prediction within every hour of a day. Knowing a specific land use vulnerability to assault during each exact hour can assist the police departments to allocate forces during those hours in risky areas. For the analysis, the study is using two datasets: a crime dataset with geographical locations of crime, date and time, and a geographic dataset about land uses with land use codes for every lot, each obtained from open databases. The study joins two datasets based on the spatial location and classifies data into 24 classes, based on the time range when the assault occurred. Machine learning methods reveal the effect of land uses on larceny, harassment and assault, the three most committed crimes in Manhattan. Finally, logistic regression provides hourly prediction models and unveils the type of land use where assaults could occur during each hour for both day and night.

Download Full-text

AN IMPROVED AUTOMATIC POINTWISE SEMANTIC SEGMENTATION OF A 3D URBAN SCENE FROM MOBILE TERRESTRIAL AND AIRBORNE LIDAR POINT CLOUDS: A MACHINE LEARNING APPROACH

ISPRS Annals of Photogrammetry Remote Sensing and Spatial Information Sciences ◽

10.5194/isprs-annals-iv-4-w8-139-2019 ◽

2019 ◽

Vol IV-4/W8 ◽

pp. 139-146

Author(s):

X.-F. Xing ◽

M. A. Mostafavi ◽

G. Edwards ◽

N. Sabo

Keyword(s):

Machine Learning ◽

Urban Areas ◽

Learning Algorithm ◽

Semantic Segmentation ◽

Point Clouds ◽

Airborne Lidar ◽

Urban Scenes ◽

Machine Learning Approach ◽

Urban Scene ◽

Point Level

Abstract. Automatic semantic segmentation of point clouds observed in a 3D complex urban scene is a challenging issue. Semantic segmentation of urban scenes based on machine learning algorithm requires appropriate features to distinguish objects from mobile terrestrial and airborne LiDAR point clouds in point level. In this paper, we propose a pointwise semantic segmentation method based on our proposed features derived from Difference of Normal and the features “directional height above” that compare height difference between a given point and neighbors in eight directions in addition to the features based on normal estimation. Random forest classifier is chosen to classify points in mobile terrestrial and airborne LiDAR point clouds. The results obtained from our experiments show that the proposed features are effective for semantic segmentation of mobile terrestrial and airborne LiDAR point clouds, especially for vegetation, building and ground classes in an airborne LiDAR point clouds in urban areas.

Download Full-text

FLOOD MAPPING USING RANDOM FOREST AND IDENTIFYING THE ESSENTIAL CONDITIONING FACTORS; A CASE STUDY IN FREDERICTON, NEW BRUNSWICK, CANADA

ISPRS Annals of Photogrammetry Remote Sensing and Spatial Information Sciences ◽

10.5194/isprs-annals-v-3-2020-609-2020 ◽

2020 ◽

Vol V-3-2020 ◽

pp. 609-615 ◽

Cited By ~ 1

Author(s):

M. Esfandiari ◽

S. Jabari ◽

H. McGrath ◽

D. Coleman

Keyword(s):

Machine Learning ◽

Random Forest ◽

New Brunswick ◽

Urban Areas ◽

Learning Algorithm ◽

Satellite Image ◽

Machine Learning Algorithms ◽

Slope Aspect ◽

Flood Peak ◽

Conditioning Factors

Abstract. Flood is one of the most damaging natural hazards in urban areas in many places around the world as well as the city of Fredericton, New Brunswick, Canada. Recently, Fredericton has been flooded in two consecutive years in 2018 and 2019. Due to the complicated behaviour of water when a river overflows its bank, estimating the flood extent is challenging. The issue gets even more challenging when several different factors are affecting the water flow, like the land texture or the surface flatness, with varying degrees of intensity. Recently, machine learning algorithms and statistical methods are being used in many research studies for generating flood susceptibility maps using topographical, hydrological, and geological conditioning factors. One of the major issues that researchers have been facing is the complexity and the number of features required to input in a machine-learning algorithm to produce acceptable results. In this research, we used Random Forest to model the 2018 flood in Fredericton and analyzed the effect of several combinations of 12 different flood conditioning factors. The factors were tested against a Sentinel-2 optical satellite image available around the flood peak day. The highest accuracy was obtained using only 5 factors namely, altitude, slope, aspect, distance from the river, and land-use/cover with 97.57% overall accuracy and 95.14% kappa coefficient.

Download Full-text

Leveraging machine learning to produce a cost-effective national building height map of Ireland

10.5194/ems2021-48 ◽

2021 ◽

Author(s):

Eoghan Keany ◽

Geoffrey Bessardon ◽

Emily Gleeson

Keyword(s):

Machine Learning ◽

Remote Sensing ◽

High Resolution ◽

Land Cover ◽

Urban Areas ◽

Cost Effective ◽

Land Cover Classification ◽

National Scale ◽

Building Height ◽

Height Data

To represent surface thermal, turbulent and humidity exchanges, Numerical Weather Prediction (NWP) systems require a land-cover classification map to calculate sur-face parameters used in surface flux estimation. The latest land-cover classification map used in the HARMONIE-AROME configuration of the shared ALADIN-HIRLAMNWP system for operational weather forecasting is ECOCLIMAP-SG (ECO-SG). The first evaluation of ECO-SG over Ireland suggested that sparse urban areas are underestimated and instead appear as vegetation areas (1). While the work of (2) on land-cover classification helps to correct the horizontal extent of urban areas, the method does not provide information on the vertical characteristics of urban areas. ECO-SG urban classification implicitly includes building heights (3), and any improvement to ECO-SG urban area extent requires a complementary building height dataset.Openly accessible building height data at a national scale does not exist for the island of Ireland. This work seeks to address this gap in availability by extrapolating the preexisting localised building height data across the entire island. The study utilises information from both the temporal and spatial dimensions by creating band-wise temporal aggregation statistics from morphological operations, for both the Sentinel-1A/B and Sentinel-2A/B constellations (4). The extrapolation uses building height information from the Copernicus Urban Atlas, which contains regional coverage for Dublin at 10 m x10 m resolution (5). Various regression models were then trained on these aggregated statistics to make pixel-wise building height estimates. These model estimates were then evaluated with an adjusted RMSE metric, with the most accurate model chosen to map the entire country. This method relies solely on freely available satellite imagery and open-source software, providing a cost-effective mapping service at a national scale that can be updated more frequently, unlike expensive once-off private mapping services. Furthermore, this process could be applied by these services to reduce costs by taking a small representative sample and extrapolating the rest of the area. This method can be applied beyond national borders providing a uniform map that does not depends on the different private service practices facilitating the updates of global or continental land-cover information used in NWP.&#160;(1) G. Bessardon and E. Gleeson, &#8220;Using the best available physiography to improve weather forecasts for Ireland,&#8221; in Challenges in High-Resolution Short Range NWP at European level including forecaster-developer cooperation, European Meteorological Society, 2019.(2) E. Walsh, et al., &#8220;Using machine learning to produce a very high-resolution land-cover map for Ireland, &#8221; Advances in Science and Research,&#160; (accepted for publication).(3) CNRM, "Wiki - ECOCLIMAP-SG" https://opensource.umr-cnrm.fr/projects/ecoclimap-sg/wiki(4) D. Frantz, et al., &#8220;National-scale mapping of building height using sentinel-1 and sentinel-2 time series,&#8221; Remote Sensing of Environment, vol. 252, 2021.(5) M. Fitrzyk, et al., &#8220;Esa Copernicus sentinel-1 exploitation activities,&#8221; in IGARSS 2019-2019 IEEE International Geoscience and Remote Sensing Symposium, IEEE, 2019.

Download Full-text

Using machine learning to produce a very high-resolution land-cover map for Ireland and beyond

10.5194/ems2021-53 ◽

2021 ◽

Author(s):

Geoffrey Bessardon ◽

Emily Gleeson ◽

Eoin Walsh

Keyword(s):

Machine Learning ◽

High Resolution ◽

Land Cover ◽

Satellite Imagery ◽

Urban Areas ◽

Weather Forecasting ◽

Land Cover Classification ◽

Surface Processes ◽

Land Cover Map ◽

Very High

An accurate representation of surface processes is essential for weather forecasting as it is where most of the thermal, turbulent and humidity exchanges occur. The Numerical Weather Prediction (NWP) system, to represent these exchanges, requires a land-cover classification map to calculate the surface parameters used in the turbulent, radiative, heat, and moisture fluxes estimations.The land-cover classification map used in the HARMONIE-AROME configuration of the shared ALADIN-HIRLAM NWP system for operational weather forecasting is ECOCLIMAP. ECOCLIMAP-SG (ECO-SG), the latest version of ECOCLIMAP, was evaluated over Ireland to prepare ECO-SG implementation in HARMONIE-AROME. This evaluation suggested that sparse urban areas are underestimated and instead appear as vegetation areas in ECO-SG [1], with an over-classification of grassland in place of sparse urban areas and other vegetation covers (Met &#201;ireann internal communication). Some limitations in the performance of the current HARMONIE-AROME configuration attributed to surface processes and physiography issues are well-known [2]. This motivated work at Met &#201;ireann to evaluate solutions to improve the land-cover map in HARMONIE-AROME.In terms of accuracy, resolution, and the future production of time-varying land-cover map, the use of a convolutional neural network (CNN) to create a land-cover map using Sentinel-2 satellite imagery [3] over Estonia [4] presented better potential outcomes than the use of local datasets [5]. Consequently, this method was tested over Ireland and proven to be more accurate than ECO-SG for representing CORINE Primary and Secondary labels and at a higher resolution [5]. This work is a continuity of [5] focusing on 1. increasing the number of labels, 2. optimising the training procedure, 3. expanding the method for application to other HIRLAM countries and 4. implementation of the new land-cover map in HARMONIE-AROME.&#160;[1] Bessardon, G., Gleeson, E., (2019) Using the best available physiography to improve weather forecasts for Ireland. In EMS Annual Meeting.Retrieved fromhttps://presentations.copernicus.org/EMS2019-702_presentation.pdf[2] Bengtsson, L., Andrae, U., Aspelien, T., Batrak, Y., Calvo, J., de Rooy, W.,. . . K&#248;ltzow, M. &#216;. (2017). The HARMONIE&#8211;AROME Model Configurationin the ALADIN&#8211;HIRLAM NWP System. Monthly Weather Review, 145(5),1919&#8211;1935.https://doi.org/10.1175/mwr-d-16-0417.1[3] Bertini, F., Brand, O., Carlier, S., Del Bello, U., Drusch, M., Duca, R., Fernandez, V., Ferrario, C., Ferreira, M., Isola, C., Kirschner, V.,Laberinti, P., Lambert, M., Mandorlo, G., Marcos, P., Martimort, P., Moon, S., Oldeman,P., Palomba, M., and Pineiro, J.: Sentinel-2ESA&#8217;s Optical High-ResolutionMission for GMES Operational Services, ESA bulletin. Bulletin ASE. Euro-pean Space Agency, SP-1322,2012[4] Ulmas, P. and Liiv, I. (2020). Segmentation of Satellite Imagery using U-Net Models for Land Cover Classification, pp. 1&#8211;11,http://arxiv.org/abs/2003.02899, 2020[5] Walsh, E., Bessardon, G., Gleeson, E., and Ulmas, P. (2021). Using machine learning to produce a very high-resolution land-cover map for Ireland. Advances in Science and Research, (accepted for publication)

Download Full-text