Spatially consistent and up-to-date maps of human settlements are crucial for addressing policies related to urbanization and sustainability, especially in the era of an increasingly urbanized world. ...The availability of open and free Sentinel-2 data of the Copernicus Earth Observation program offers a new opportunity for wall-to-wall mapping of human settlements at a global scale. This paper presents a deep-learning-based framework for a fully automated extraction of built-up areas at a spatial resolution of 10 m from a global composite of Sentinel-2 imagery. A multi-neuro modeling methodology building on a simple Convolution Neural Networks architecture for pixel-wise image classification of built-up areas is developed. The core features of the proposed model are the image patch of size 5 × 5 pixels adequate for describing built-up areas from Sentinel-2 imagery and the lightweight topology with a total number of 1,448,578 trainable parameters and 4 2D convolutional layers and 2 flattened layers. The deployment of the model on the global Sentinel-2 image composite provides the most detailed and complete map reporting about built-up areas for reference year 2018. The validation of the results with an independent reference dataset of building footprints covering 277 sites across the world establishes the reliability of the built-up layer produced by the proposed framework and the model robustness. The results of this study contribute to cutting-edge research in the field of automated built-up areas mapping from remote sensing data and establish a new reference layer for the analysis of the spatial distribution of human settlements across the rural–urban continuum.
This paper presents the analysis of Earth Observation data records collected between 1975 and 2014 for assessing the extent and temporal evolution of the built-up surface in the frame of the Global ...Human Settlement Layer project. The scale of the information produced by the study enables the assessment of the whole continuum of human settlements from rural hamlets to megacities. The study applies enhanced processing methods as compared to the first production of the GHSL baseline data. The major improvements include the use of a more refined learning set on built-up areas derived from Sentinel-1 data which allowed testing the added-value of incremental learning in big data analytics. Herein, the new features of the GHSL built-up grids and the methods are described and compared with the previous ones using a reference set of building footprints for 277 areas of interest. The results show a gradual improvement in the accuracy measures with a gain of 3.6% in the balanced accuracy, between the first production of the GHSL baseline and the latest GHSL multitemporal built-up grids. A validation of the multitemporal component is also conducted at the global scale establishing the reliability of the built-up layer across time.
Continuous global-scale mapping of human settlements in the service of international agreements calls for massive volume of multi-source, multi-temporal, and multi-scale earth observation data. In ...this paper, the latest developments in terms of processing big earth observation data for the purpose of improving the Global Human Settlement Layer (GHSL) data are presented. Two experiments with Sentinel-1 and Landsat data collections were run leveraging on the Joint Research Centre Earth Observation Data and Processing Platform. A comparative analysis of the results of built-up areas extraction from different remote sensing data and processing workflows shows how the information production supported by data-intensive computing infrastructure for optimization and multiple testing can improve the output information reliability and consistency within the GHSL scope. The paper presents the processing workflows and the results of the two main experiments, giving insights into the enhanced mapping capabilities gained by analyzing Sentinel-1 and Landsat data-sets, and the lessons learnt in terms of handling and processing big earth observation data.
Built-up areas extraction and characterization from remote sensing images is essential for monitoring urbanization and the associated challenges. This work presents a novel integrated classification ...framework building on the symbolic machine learning classifier and fully polarimetric Phased Array type L-band Synthetic Aperture Radar-2 (PALSAR-2) to derive both the extent and vertical components of built-up areas from the same scene. It also explores the complementarity between ascending and descending orbits of PALSAR-2 for built-up areas detection. The experimental results in Chicago and Tokyo cities with different landscape and characteristics of built-up areas demonstrate that the proposed generic method can achieve three main challenges of urban remote sensing: 1) enabling automated delineation of built-up areas at a spatial resolution of 5 m with a balanced accuracy of 85% using globally available low-resolution training data, 2) assessing the density of building height class with a root mean square error of 0.25, 0.034, and 0.032 for the low-rise, mid-rise, and high-rise building density class, respectively, and 3) dealing with the scattering components of buildings with different orientation angles by combining data from ascending and descending orbits for enhanced mapping of built-up areas.
The aim of the paper is to compare Moderate Resolution Imaging Spectroradiometer (MODIS) Normalized Difference Vegetation Index (
) products with different compositing periods (8-day, 16-day and ...8-day-dual) at 250 m spatial resolution for early corn yield estimation. In order to achieve this objective, several regression models were used where the average yield was the dependent variable and the different values of
were independent variables. The various inputs in the regression models included: (i) maximum
(peak value) during the season or heading date value, (ii)
values from the first date after heading date, (iii),
values from the second date after heading date, (iv) Seasonally integrated
values. Results showed that the 16-day composite was better yield predictor than the 8-day composite when using maximum
value during the season, which is the value from the most significant earliest period for yield estimation, which is called the heading date. The 8-day composites were more useful than 16-day composites later in the season for yield estimation when
values from first date after heading date and values from second date after heading were used. However, the 8-day-dual was not useful for yield prediction. In order to validate the results, the authors used the leave-one-year-out approach, which trains the remaining years for the left out year and is used for yield prediction for missing year. It was found that the inverse regression model produced the best yield estimates. After excluding the anomalous 2012 year, the R
values for the regression model were > 0.5 for all remaining years and products, with statistical significant at 0.05. The smallest difference between predicted and actual corn yield when using 8-day composite was 0.05% while the largest difference was 34.47%, whilst in the case of 16-day composite the smallest difference between predicted and actual yield was 1.67% and the largest difference was 44.12%.
Change detection is a process of detecting differences with the objects or phenomena which are observed in the different time intervals. In this study different methods of analyzing satellite images ...are presented, with the aim to identify changes in land cover in a certain period of time (1985 - 2013). The area observed in this study is the region of mountain Zlatibor (Serbia) with its surroundings. The methods represented in this study are vegetation indices differencing, Supervised classification and Object based classification. These methods gave different results in term of land cover area, and it is generally concluded that supervised classification gave the most accurate results with the images of medium spatial resolution. The results of this study can be used for urban and environmental planning. All information lead to conclusion that the surface under the forests is reduced for about 4% (or about 1000 ha) while the built up area has doubled (grown about 600 ha) during the examined period. The results also highlights the importance of change detection techniques in land cover for the areas that are developing rapidly, such as Zlatibor study area.
Over the recent years, networks of coupled oscillators or oscillatory neural networks (ONNs) emerged as an alternative computing paradigm with information encoded in phase. Such networks are ...intrinsically attractive for associative memory applications such as pattern retrieval. Thus far, there are few works focusing on image classification using ONNs, as there is no straightforward way to do it. This paper investigates the performance of a neuromorphic phase-based classification model using a fully connected single layer ONNs. For benchmarking, we deploy the ONN on the full set of 28\times 28 binary MNIST handwritten digits and achieve around 70% accuracy on both training and test set. To the best of our knowledge, this is the first effort classifying such large images utilizing ONNs.
Predicting crop yields, and especially anomalously low yields, is of special importance for food insecure countries. In this study, we investigate a flexible deep learning approach to forecast crop ...yield at the provincial administrative level based on deep 1D and 2D convolutional neural networks using limited data. This approach meets the operational requirements—public and global records of satellite data in an application ready format with near real time updates—and can be transferred to any country with reliable yield statistics. Three-dimensional histograms of normalized difference vegetation index (NDVI) and climate data are used as input to the 2D model, while simple administrative-level time series averages of NDVI and climate data to the 1D model. The best model architecture is automatically identified during efficient and extensive hyperparameter optimization. To demonstrate the relevance of this approach, we hindcast (2002–2018) the yields of Algeria’s three main crops (barley, durum and soft wheat) and contrast the model’s performance with machine learning algorithms and conventional benchmark models used in a previous study. Simple benchmarks such as peak NDVI remained challenging to outperform while machine learning models were superior to deep learning models for all forecasting months and all tested crops. We attribute the poor performance of deep learning to the small size of the dataset available.
Monitoring phenology of crops and yield estimate based on vegetation indices as well as other parameters such as temperature or amount of rainfall were largely reported in literature. In this ...research, MODIS Normalized Difference Vegetation Index (NDVI) was used as an indicator of specific crop condition; the other parameter was Land Surface Temperature (LST) which can indicate the amount of crop moisture. Trial years were 2011, 2012, and 2013. For those years sowing structure was acquired from agricultural organizations Nova Budućnost from Žarkovac and Sava Kovačević from Vrbas, both in Serbia. Also, satellite images with high and medium resolution for these areas and years were available. Multiple linear regression was used for crop yield estimate for Vojvodina Province, Serbia where the NDVI and LST were independent variables and the average yield for specific crop was the dependent variable. The results of crop yield estimate two months before harvest are presented (excluding wheat).
The validation of built‐up areas derived from different sensors is crucial for gaining a deeper understanding of the consistency and interoperability between them. This article presents the ...methodology and results of an inter‐sensor comparison of built‐up area data derived from Landsat, Sentinel‐1, Sentinel‐2, and SPOT5/SPOT6. The assessment was performed for 13 cities across the world for which cartographic reference building footprints were available. Several validation approaches were used: cumulative built‐up curve analysis, pixel‐by‐pixel performance metrics, and regression analysis. The results indicate that Sentinel‐1 and Sentinel‐2 contribute greatly to improved built‐up area detection compared to Landsat, within the global human settlement framework. However, Sentinel‐2 tends to show high omission errors while Landsat tends to have the lowest omission error. The built‐up area obtained from SPOT5/SPOT6 shows high consistency with the reference data for all European cities, and hence can potentially be considered as a reference dataset for wall‐to‐wall validation in Europe.