scholarly journals CLOUD DETECTION FOR NIGHT-TIME PANCHROMATIC VISIBLE AND NEAR-INFRARED SATELLITE IMAGERY

2020 ◽  
Vol V-2-2020 ◽  
pp. 853-860
Author(s):  
L. Joachim ◽  
T. Storch
Keyword(s):  
Random Forest ◽  
Satellite Imagery ◽  
Urban Areas ◽  
Near Infrared ◽  
Cloud Detection ◽  
Night Time ◽  
Training Site ◽  
Multi Temporal ◽  
Land Covers ◽  
Scene Features

Abstract. Cloud detection for night-time panchromatic visible and near-infrared (VNIR) satellite imagery is typically performed based on synchronized observations in the thermal infrared (TIR). To be independent of TIR and to improve existing algorithms, we realize and analyze cloud detection based on VNIR only, here NPP/VIIRS/DNB observations. Using Random Forest for classifying cloud vs. clear and focusing on urban areas, we illustrate the importance of features describing a) the scattering by clouds especially over urban areas with their inhomogeneous light emissions and b) the normalized differences between Earth’s surface and cloud albedo especially in presence of Moon illumination. The analyses substantiate the influences of a) the training site and scene selections and b) the consideration of single scene or multi-temporal scene features on the results for the test sites. As test sites, diverse urban areas and the challenging land covers ocean, desert, and snow are considered. Accuracies of up to 85% are achieved for urban test sites.

scholarly journals Estimation of Poverty Using Random Forest Regression with Multi-Source Data: A Case Study in Bangladesh

2019 ◽  
Vol 11 (4) ◽  
pp. 375 ◽  
Author(s):  
Xizhi Zhao ◽  
Bailang Yu ◽  
Yan Liu ◽  
Zuoqi Chen ◽  
Qiaoxuan Li ◽  
...  
Keyword(s):  
Random Forest ◽  
Satellite Imagery ◽  
Urban Areas ◽  
Pearson Correlation ◽  
Explanatory Power ◽  
Household Wealth ◽  
Environmental Data ◽  
Demographic And Health Surveys ◽  
Poverty Level ◽  
Random Forest Regression

Spatially explicit and reliable data on poverty is critical for both policy makers and researchers. However, such data remain scarce particularly in developing countries. Current research is limited in using environmental data from different sources in isolation to estimate poverty despite the fact that poverty is a complex phenomenon which cannot be quantified either theoretically or practically by one single data type. This study proposes a random forest regression (RFR) model to estimate poverty at 10 km × 10 km spatial resolution by combining features extracted from multiple data sources, including the National Polar-orbiting Partnership Visible Infrared Imaging Radiometer Suite (NPP-VIIRS) Day/Night Band (DNB) nighttime light (NTL) data, Google satellite imagery, land cover map, road map and division headquarter location data. The household wealth index (WI) drawn from the Demographic and Health Surveys (DHS) program was used to reflect poverty level. We trained the RFR model using data in Bangladesh and applied the model to both Bangladesh and Nepal to evaluate the model’s accuracy. The results show that the R2 between the actual and estimated WI in Bangladesh is 0.70, indicating a good predictive power of our model in WI estimation. The R2 between actual and estimated WI of 0.61 in Nepal also indicates a good generalization ability of the model. Furthermore, a negative correlation is observed between the district average WI and the poverty head count ratio (HCR) in Bangladesh with the Pearson Correlation Coefficient of -0.6. Using Gini importance, we identify that proximity to urban areas is the most important variable to explain poverty which contribute to 37.9% of the explanatory power. Compared to the study that used NTL and Google satellite imagery in isolation to estimate poverty, our method increases the accuracy of estimation. Given that the data we use are globally and publicly available, the methodology reported in this study would also be applicable in other countries or regions to estimate the extent of poverty.

scholarly journals Utilizing Multilevel Features for Cloud Detection on Satellite Imagery

2018 ◽  
Vol 10 (11) ◽  
pp. 1853 ◽  
Author(s):  
Xi Wu ◽  
Zhenwei Shi
Keyword(s):  
Neural Network ◽  
Satellite Imagery ◽  
Binary Classification ◽  
Image Features ◽  
Image Feature ◽  
Detection Methods ◽  
Wide Field ◽  
Cloud Detection ◽  
Low Level ◽  
Land Covers

Cloud detection, which is defined as the pixel-wise binary classification, is significant in satellite imagery processing. In current remote sensing literature, cloud detection methods are linked to the relationships of imagery bands or based on simple image feature analysis. These methods, which only focus on low-level features, are not robust enough on the images with difficult land covers, for clouds share similar image features such as color and texture with the land covers. To solve the problem, in this paper, we propose a novel deep learning method for cloud detection on satellite imagery by utilizing multilevel image features with two major processes. The first process is to obtain the cloud probability map from the designed deep convolutional neural network, which concatenates deep neural network features from low-level to high-level. The second part of the method is to get refined cloud masks through a composite image filter technique, where the specific filter captures multilevel features of cloud structures and the surroundings of the input imagery. In the experiments, the proposed method achieves 85.38% intersection over union of cloud in the testing set which contains 100 Gaofen-1 wide field of view images and obtains satisfactory visual cloud masks, especially for those hard images. The experimental results show that utilizing multilevel features by the combination of the network with feature concatenation and the particular filter tackles the cloud detection problem with improved cloud masks.

scholarly journals Optimized Performance Parameters for Nighttime Multispectral Satellite Imagery to Analyze Lightings in Urban Areas

Sensors ◽  
2020 ◽  
Vol 20 (11) ◽  
pp. 3313 ◽  
Author(s):  
Jasper de Meester ◽  
Tobias Storch
Keyword(s):  
Spatial Resolution ◽  
Satellite Imagery ◽  
Urban Areas ◽  
Near Infrared ◽  
Atmospheric Effects ◽  
Spectral Bands ◽  
Multispectral Satellite Imagery ◽  
Radiation Sources ◽  
Yellow Orange ◽  
Band Combinations

Contrary to its daytime counterpart, nighttime visible and near infrared (VIS/NIR) satellite imagery is limited in both spectral and spatial resolution. Nevertheless, the relevance of such systems is unquestioned with applications to, e.g., examine urban areas, derive light pollution, and estimate energy consumption. To determine optimal spectral bands together with required radiometric and spatial resolution, at-sensor radiances are simulated based on combinations of lamp spectra with typical luminances according to lighting standards, surface reflectances, and radiative transfers for the consideration of atmospheric effects. Various band combinations are evaluated for their ability to differentiate between lighting types and to estimate the important lighting parameters: efficacy to produce visible light, percentage of emissions attributable to the blue part of the spectrum, and assessment of the perceived color of radiation sources. The selected bands are located in the green, blue, yellow-orange, near infrared, and red parts of the spectrum and include one panchromatic band. However, these nighttime bands tailored to artificial light emissions differ significantly from the typical daytime bands focusing on surface reflectances. Compared to existing or proposed nighttime or daytime satellites, the recommended characteristics improve, e.g., classification of lighting types by >10%. The simulations illustrate the feasible improvements in nocturnal VIS/NIR remote sensing which will lead to advanced applications.

scholarly journals Machine Learning Methods for Classification of the Green Infrastructure in City Areas

2019 ◽  
Vol 8 (10) ◽  
pp. 463 ◽  
Author(s):  
Nikola Kranjčić ◽  
Damir Medak ◽  
Robert Župan ◽  
Milan Rezo
Keyword(s):  
Machine Learning ◽  
Random Forest ◽  
Satellite Imagery ◽  
Urban Areas ◽  
Urban Infrastructure ◽  
Support Vector ◽  
Bayes Classifier ◽  
Learning Methods ◽  
Machine Learning Methods ◽  
Artificial Neural

Rapid urbanization in cities can result in a decrease in green urban areas. Reductions in green urban infrastructure pose a threat to the sustainability of cities. Up-to-date maps are important for the effective planning of urban development and the maintenance of green urban infrastructure. There are many possible ways to map vegetation; however, the most effective way is to apply machine learning methods to satellite imagery. In this study, we analyze four machine learning methods (support vector machine, random forest, artificial neural network, and the naïve Bayes classifier) for mapping green urban areas using satellite imagery from the Sentinel-2 multispectral instrument. The methods are tested on two cities in Croatia (Varaždin and Osijek). Support vector machines outperform random forest, artificial neural networks, and the naïve Bayes classifier in terms of classification accuracy (a Kappa value of 0.87 for Varaždin and 0.89 for Osijek) and performance time.

scholarly journals Monitoring Dynamics of Vegetation Cover with the Integration of OBIA and Random Forest Classifier Using Sentinel-2 Multitemporal Satellite Imagery

2022 ◽  
Vol 8 (2) ◽  
pp. 75-84
Author(s):  
Nurwita Mustika Sari ◽  
R. Rokhmatuloh ◽  
Masita Dwi Mandini Manessa
Keyword(s):  
Random Forest ◽  
Satellite Imagery ◽  
Vegetation Cover ◽  
Anthropogenic Activities ◽  
Research Area ◽  
Random Forest Classifier ◽  
Sustainable Environment ◽  
Multi Temporal ◽  
Sentinel 2 ◽  
Dynamics Of Vegetation

The existence of vegetation in an area has an important role to maintain the carrying capacity of the environment and create a comfortable environment as a place to live. In an effort to create a sustainable environment, there are various pressures on vegetation that cause a decrease in vegetation area. Economic activity, population growth and other anthropogenic activities trigger the dynamics of vegetation cover in an area that causes land cover changes from vegetation to non-vegetation. Majalengka Regency as one of the areas with intensive regional physical development in line with the operation of BIJB Kertajati and the Cipali toll road became the study area in this research. This study aims to monitor the dynamics of vegetation cover with the proposed method namely the integration of the OBIA and Random Forest classifier using multi temporal Sentinel-2 satellite imagery. The results show that there is a decrease in the area of vegetation in the research area as much as 4,329.6 hectares to non-vegetation areas in the period 2016-2020. The vegetation area in 2020 is 84,716.07 hectares and non-vegetation area is 35,708 hectares. Thus, there has been a decrease in the percentage of vegetation area from 73.94% in 2016 to 70.35% in 2020, meanwhile for non-vegetation areas there has been an increase from 26.06% in 2016 to 29.65% in 2020.

scholarly journals Spatial-Temporal Land Use and Land Cover Changes in Urban Areas Using Remote Sensing Images and GIS Analysis: The Case Study of Opole, Poland

Geosciences ◽  
2021 ◽  
Vol 11 (8) ◽  
pp. 312
Author(s):  
Barbara Wiatkowska ◽  
Janusz Słodczyk ◽  
Aleksandra Stokowska
Keyword(s):  
Remote Sensing ◽  
Land Use ◽  
Land Cover ◽  
Satellite Imagery ◽  
Urban Areas ◽  
Agricultural Land ◽  
Urban Expansion ◽  
Landsat 8 ◽  
Rate Of Increase ◽  
Management Procedures

Urban expansion is a dynamic and complex phenomenon, often involving adverse changes in land use and land cover (LULC). This paper uses satellite imagery from Landsat-5 TM, Landsat-8 OLI, Sentinel-2 MSI, and GIS technology to analyse LULC changes in 2000, 2005, 2010, 2015, and 2020. The research was carried out in Opole, the capital of the Opole Agglomeration (south-western Poland). Maps produced from supervised spectral classification of remote sensing data revealed that in 20 years, built-up areas have increased about 40%, mainly at the expense of agricultural land. Detection of changes in the spatial pattern of LULC showed that the highest average rate of increase in built-up areas occurred in the zone 3–6 km (11.7%) and above 6 km (10.4%) from the centre of Opole. The analysis of the increase of built-up land in relation to the decreasing population (SDG 11.3.1) has confirmed the ongoing process of demographic suburbanisation. The paper shows that satellite imagery and GIS can be a valuable tool for local authorities and planners to monitor the scale of urbanisation processes for the purpose of adapting space management procedures to the changing environment.

scholarly journals Extraction of Arecanut Planting Distribution Based on the Feature Space Optimization of PlanetScope Imagery

Agriculture ◽  
2021 ◽  
Vol 11 (4) ◽  
pp. 371
Author(s):  
Yu Jin ◽  
Jiawei Guo ◽  
Huichun Ye ◽  
Jinling Zhao ◽  
Wenjiang Huang ◽  
...  
Keyword(s):  
Random Forest ◽  
Satellite Imagery ◽  
Feature Space ◽  
Kappa Coefficient ◽  
Classification Model ◽  
Support Vector ◽  
Textural Feature ◽  
Monitoring Accuracy ◽  
Areca Catechu ◽  
High Level

The remote sensing extraction of large areas of arecanut (Areca catechu L.) planting plays an important role in investigating the distribution of arecanut planting area and the subsequent adjustment and optimization of regional planting structures. Satellite imagery has previously been used to investigate and monitor the agricultural and forestry vegetation in Hainan. However, the monitoring accuracy is affected by the cloudy and rainy climate of this region, as well as the high level of land fragmentation. In this paper, we used PlanetScope imagery at a 3 m spatial resolution over the Hainan arecanut planting area to investigate the high-precision extraction of the arecanut planting distribution based on feature space optimization. First, spectral and textural feature variables were selected to form the initial feature space, followed by the implementation of the random forest algorithm to optimize the feature space. Arecanut planting area extraction models based on the support vector machine (SVM), BP neural network (BPNN), and random forest (RF) classification algorithms were then constructed. The overall classification accuracies of the SVM, BPNN, and RF models optimized by the RF features were determined as 74.82%, 83.67%, and 88.30%, with Kappa coefficients of 0.680, 0.795, and 0.853, respectively. The RF model with optimized features exhibited the highest overall classification accuracy and kappa coefficient. The overall accuracy of the SVM, BPNN, and RF models following feature optimization was improved by 3.90%, 7.77%, and 7.45%, respectively, compared with the corresponding unoptimized classification model. The kappa coefficient also improved. The results demonstrate the ability of PlanetScope satellite imagery to extract the planting distribution of arecanut. Furthermore, the RF is proven to effectively optimize the initial feature space, composed of spectral and textural feature variables, further improving the extraction accuracy of the arecanut planting distribution. This work can act as a theoretical and technical reference for the agricultural and forestry industries.

Affective labour and class distinction in the night-time economy

2021 ◽  
pp. 003802612110063
Author(s):  
Steven Threadgold ◽  
David Farrugia ◽  
Julia Coffey
Keyword(s):  
Symbolic Dynamics ◽  
Urban Areas ◽  
Service Work ◽  
Night Time ◽  
Feminist Analysis ◽  
Consumption Experience ◽  
Gendered Politics ◽  
The Aesthetic ◽  
The Relationship ◽  
Class Distinction

This article contributes to recent debates about the relationship between affective labour and class by exploring the classed distinctions enacted through affective labour in the urban night-time economy. Bringing theories of affective labour into a dialogue with Bourdieusian feminist analysis, the article explores the affective and symbolic dynamics of hospitality labour in a gentrified inner-urban neighbourhood of Melbourne, Australia. It shows how the practice of hospitality labour enacts classed distinctions and tensions emerging from the gentrification of inner-urban areas, and how the aesthetic and symbolic dimensions of class contribute to the valorisation of affect in hospitality venues. The valorisation of affect are processes in which the value attributed to an atmosphere or consumption experience is based on the forms of distinction practised within the venue, enacted in aesthetics, tastes and modes of embodiment. The article also shows how practices of class distinction – both ‘punching up’ and ‘managing down’ – are connected to the gendered politics of service work in the way that workers manage the threat of violence or sexual harassment in venues. In general, the article shows how the classed dynamics of gentrification are enacted in affective economies, and therefore how Bourdieusian analysis of class can be usefully deployed in theoretical debates about affective labour.

Sign in / Sign up

Export Citation Format

Share Document