Next Article in Journal
A Rapid and High-Precision Mountain Vertex Extraction Method Based on Hotspot Analysis Clustering and Improved Eight-Connected Extraction Algorithms for Digital Elevation Models
Previous Article in Journal
Development and Performance Evaluation of a Very Low-Cost UAV-Lidar System for Forestry Applications
Previous Article in Special Issue
RSCNN: A CNN-Based Method to Enhance Low-Light Remote-Sensing Images
Open AccessArticle

Deep Learning for Land Cover Change Detection

Institute of Photogrammetry and Remote Sensing, Karlsruhe Institute of Technology, 76131 Karlsruhe, Germany
*
Author to whom correspondence should be addressed.
These authors contributed equally to this work.
Remote Sens. 2021, 13(1), 78; https://doi.org/10.3390/rs13010078
Received: 24 November 2020 / Revised: 20 December 2020 / Accepted: 23 December 2020 / Published: 28 December 2020
Land cover and its change are crucial for many environmental applications. This study focuses on the land cover classification and change detection with multitemporal and multispectral Sentinel-2 satellite data. To address the challenging land cover change detection task, we rely on two different deep learning architectures and selected pre-processing steps. For example, we define an excluded class and deal with temporal water shoreline changes in the pre-processing. We employ a fully convolutional neural network (FCN), and we combine the FCN with long short-term memory (LSTM) networks. The FCN can only handle monotemporal input data, while the FCN combined with LSTM can use sequential information (multitemporal). Besides, we provided fixed and variable sequences as training sequences for the combined FCN and LSTM approach. The former refers to using six defined satellite images, while the latter consists of image sequences from an extended training pool of ten images. Further, we propose measures for the robustness concerning the selection of Sentinel-2 image data as evaluation metrics. We can distinguish between actual land cover changes and misclassifications of the deep learning approaches with these metrics. According to the provided metrics, both multitemporal LSTM approaches outperform the monotemporal FCN approach, about 3 to 5 percentage points (p.p.). The LSTM approach trained on the variable sequences detects 3 p.p. more land cover changes than the LSTM approach trained on the fixed sequences. Besides, applying our selected pre-processing improves the water classification and avoids reducing the dataset effectively by 17.6%. The presented LSTM approaches can be modified to provide applicability for a variable number of image sequences since we published the code of the deep learning models. The Sentinel-2 data and the ground truth are also freely available. View Full-Text
Keywords: machine learning; multi-class classification; long short-term memory network (LSTM); fully convolutional neural network (FCN); multitemporal; time series; Sentinel-2 machine learning; multi-class classification; long short-term memory network (LSTM); fully convolutional neural network (FCN); multitemporal; time series; Sentinel-2
Show Figures

Graphical abstract

MDPI and ACS Style

Sefrin, O.; Riese, F.M.; Keller, S. Deep Learning for Land Cover Change Detection. Remote Sens. 2021, 13, 78.

AMA Style

Sefrin O, Riese FM, Keller S. Deep Learning for Land Cover Change Detection. Remote Sensing. 2021; 13(1):78.

Chicago/Turabian Style

Sefrin, Oliver; Riese, Felix M.; Keller, Sina. 2021. "Deep Learning for Land Cover Change Detection" Remote Sens. 13, no. 1: 78.

Find Other Styles
Note that from the first issue of 2016, MDPI journals use article numbers instead of page numbers. See further details here.

Article Access Map by Country/Region

1
Search more from Scilit
 
Search
Back to TopTop