Understanding deep learning in land use classification based on Sentinel-2 time series

Campos-Taberner, Manuel; Garcı́a-Haro, Francisco Javier; Martínez, Beatriz; Izquierdo‐Verdiguier, Emma; Atzberger, Clement; Camps‐Valls, Gustau; Gilabert, María Amparo

doi:10.1038/s41598-020-74215-5

Cited by 143 publications

(76 citation statements)

References 52 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Even though in terms of overall accuracy, the performance of ML algorithms generally outperforms -shallow-machine learning techniques (Campos-Taberner et al, 2016;Liu et al, 2018), the understanding of these algorithms and their interpretation is typically limited (Campos-Taberner et al, 2020;Montavon et al, 2018).…”

Section: Resultsmentioning

confidence: 99%

Rocky Shoreline Extraction Using a Deep Learning Model and Object-Based Image Analysis

Bengoufa

Niculescu

Mihoubi

et al. 2021

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

View full text Add to dashboard Cite

Abstract. In the context of the increasing anthropogenic influence on the coastal areas that are subject to high climate variability, the main challenge is to understand its current dynamics and to predict its future evolution. Therefore, monitoring of the shoreline kinematics is a key factor for the coastal erosion assessment and an essential feature for the sustainable management of these naturally vulnerable areas.This work focuses on the detection and extraction of the shoreline, basing on a specific remote sensing methodology using Very High Resolution (VHR) optical images. Indeed, an integrated approach based on a Deep Learning model, which is the Convolutional Neural Network (CNN) and Object Based Image Analysis (OBIA) has been developed. This study aims to evaluate the methodological contribution of this integrated approach for the (semi)-automatic extraction of the rocky shoreline, for which the botanical indicator has been chosen. Therefore the upper limit of black marine lichen has been detected and extracted as the target shoreline. It is the first indication of a (semi)-automatic detection of such a complex type of shoreline.The classification results derived from the combined CNN model and OBIA methods had achieved a high overall accuracy of 0.94. The extracted shoreline have been compared to a shoreline of reference derived from a traditional method that is a manual digitizing. The distances between the two shorelines has been calculated in order to assess the accuracy of the extraction method. This comparison revealed that 76% of the extracted shoreline lies within 1 m, and 35% lies within 0.5 m of reference one. Therefore, the CNN model integrated to OBIA was successfully shown to be a good method for shoreline extraction and could offer an immediate insight regarding rocky shoreline position, providing an alternative to its monitoring.

show abstract

Section: Resultsmentioning

confidence: 99%

Rocky Shoreline Extraction Using a Deep Learning Model and Object-Based Image Analysis

Bengoufa

Niculescu

Mihoubi

et al. 2021

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

View full text Add to dashboard Cite

show abstract

“…One common replacement for this approach is using the available databases as the training input, including existing land classification maps or the survey data. In a recent study by Campos-Taberner et al (2020), the classification samples originate from in situ surveys provided by the regional government of Valencia in Spain (Campos-Taberner et al, 2020). Similarly, Zhu et al (2016), used the Land Cover Trends (LCT) data produced through the most recent Geological Survey of the US as the training sample (Zhu et al, 2016).…”

Section: Methods For Developing the Training Set And The Application Of Open Datamentioning

confidence: 99%

Using open data to detect the structure and pattern of informal settlements: an outset to support inclusive SDGs’ achievement

2021

View full text Add to dashboard Cite

The detection of informal settlements is the first step in planning and upgrading deprived areas in order to leave no one behind in SDGs. Very High-Resolution satellite images (VHR), have been extensively used for this purpose. However, as a cost-prohibitive data source, VHR might not be available to all, particularly nations that are home to many informal settlements. This study examines the application of open and freely available data sources to detect the structure and pattern of informal settlements. Here, in a case study of Jakarta, Indonesia, Medium Resolution satellite imagery (MR) derived from Landsat 8 (2020) was classified to detect these settlements. The classification was done using Random Forest (RF) classifier through two complementary approaches to develop the training set. In the first approach, available survey data sets (Jakarta's informal settlements map for 2015) and visual interpretation using High-Resolution Google Map imagery have been used to build the training set. Throughout the second round of classification, OpenStreetMap (OSM) layers were used as the complementary approach for training. Results from the validation test for the second round revealed better accuracy and precision in classification. The proposed method provides an opportunity to use open data for informal settlements detection, when: 1) more expensive high resolution data sources are not accessible; 2) the area of interest is not larger than a city; and 3) the physical characteristics of the settlements differ significantly from their surrounding formal area. The method presents the application of globally accessible data to help the achievement of resilience and SDGs in informal settlements.

show abstract

“…It is noted that besides SVM, deep neural networks (DNNs) have also been successfully applied in remote sensing-based land-use classification [32][33][34]. DNNs are highly appropriate for image categorization due to its convolution operator based autonomous feature extraction phase [35].…”

Section: Research Background and Motivationmentioning

confidence: 99%

Remote Sensing–Based Urban Green Space Detection Using Marine Predators Algorithm Optimized Machine Learning Approach

Hoang

Tran

2021

Mathematical Problems in Engineering

View full text Add to dashboard Cite

Information regarding the current status of urban green space is crucial for urban land-use planning and management. This study proposes a remote sensing and data-driven solution for urban green space detection at regional scale via employment of state-of-the-art metaheuristic and machine learning approaches. Remotely sensed data obtained from Sentinel 2 satellite in the study area of Da Nang city (Vietnam) are used to construct and verify an intelligent model that hybridizes Marine Predators Algorithm (MPA) and support vector machines (SVM). SVM are employed to generalize a decision boundary that separates features characterizing statistical measurements of remote sensing data into two categories of “green space” and “nongreen space”. The MPA metaheuristic is used to optimize the SVM training phase by identifying an appropriate set of the SVM’s hyperparameters including the penalty coefficient and the kernel function parameter. Experimental results show that the proposed model which processes information provided by all of the Sentinel 2 satellite’s spectral bands can deliver a better performance than those obtained from the model based on vegetation indices. With a good classification accuracy rate of roughly 93%, an F1 score = 0.93, and an area under the receiver operating characteristic = 0.98, the newly developed model is a promising tool to assist local authority to obtain up-to-date information on urban green space and develop plans of sustainable urban land use.

show abstract

Understanding deep learning in land use classification based on Sentinel-2 time series

Cited by 143 publications

References 52 publications

Rocky Shoreline Extraction Using a Deep Learning Model and Object-Based Image Analysis

Rocky Shoreline Extraction Using a Deep Learning Model and Object-Based Image Analysis

Using open data to detect the structure and pattern of informal settlements: an outset to support inclusive SDGs’ achievement

Remote Sensing–Based Urban Green Space Detection Using Marine Predators Algorithm Optimized Machine Learning Approach

Contact Info

Product

Resources

About