Earth observation (EO) sensors deliver data at daily or weekly intervals. Most land use and land cover classification (LULC) approaches, however, are designed for cloud-free and mono-temporal ...observations. The increasing temporal capabilities of today’s sensors enable the use of temporal, along with spectral and spatial features.Domains such as speech recognition or neural machine translation, work with inherently temporal data and, today, achieve impressive results by using sequential encoder-decoder structures. Inspired by these sequence-to-sequence models, we adapt an encoder structure with convolutional recurrent layers in order to approximate a phenological model for vegetation classes based on a temporal sequence of Sentinel 2 (S2) images. In our experiments, we visualize internal activations over a sequence of cloudy and non-cloudy images and find several recurrent cells that reduce the input activity for cloudy observations. Hence, we assume that our network has learned cloud-filtering schemes solely from input data, which could alleviate the need for tedious cloud-filtering as a preprocessing step for many EO approaches. Moreover, using unfiltered temporal series of top-of-atmosphere (TOA) reflectance data, our experiments achieved state-of-the-art classification accuracies on a large number of crop classes with minimal preprocessing, compared to other classification approaches.
Better spatial information on the global distribution of croplands and pastures is urgently needed. Without reliable cropland–pasture separation it will be impossible to retrieve high-quality ...information on agricultural expansion or land use intensification, and on related ecosystem service provision. In this context, the savanna biome is critically important, but information on land use and land cover (LULC) is notoriously inaccurate in these areas. This is due to pronounced spatial–temporal dynamics of agricultural land use and spectral similarities between cropland, pasture, and natural savanna vegetation. In this study, we investigated the potential to reliably separate cropland, pasture, natural savanna vegetation, and other relevant land cover classes employing Landsat-derived spectral–temporal variability metrics for a savanna landscape in the Brazilian Cerrado. In order to better understand the surplus value and limitations of spectral–temporal variability metrics for classification purposes, we analyzed four datasets of different temporal depth, using 344 Landsat scenes across four footprints between 2009 and 2012. Our results showed a reliable separation between cropland, pasture, and natural savanna vegetation achieving an adjusted overall accuracy of 93%. A similar accuracy and spatial consistency of LULC classification could not be achieved based on spectral information alone, indicating the high additional value of temporal information for identifying LULC classes in the complex land use systems of savanna landscapes. There is great potential for transferring our approach to other savanna systems which still suffer from inaccurate LULC information.
•We classified LULC in a savanna system using spectral–temporal variability metrics.•The classification reached high accuracy (93%) and spatial consistency of LULC pattern.•For separating LULC, phenological timing was more important than observation density.•Our findings offer great potential to classify LULC in other savanna regions.
Sustainable natural resources management relies on effective and timely assessment of conservation and land management practices. Using satellite imagery for Earth observation has become essential ...for monitoring land cover/land use (LCLU) changes and identifying critical areas for conserving biodiversity. Remote Sensing (RS) datasets are often quite large and require tremendous computing power to process. The emergence of cloud-based computing techniques presents a powerful avenue to overcome computing limitations by allowing machine-learning algorithms to process and analyze large RS datasets on the cloud. Our study aimed to classify LCLU for the Talassemtane National Park (TNP) using a Deep Neural Network (DNN) model incorporating five spectral indices to differentiate six land use classes using Sentinel-2 satellite imagery. Optimization of the DNN model was conducted using a comparative analysis of three optimization algorithms: Random Search, Hyperband, and Bayesian optimization. Results indicated that the spectral indices improved classification between classes with similar reflectance. The Hyperband method had the best performance, improving the classification accuracy by 12.5% and achieving an overall accuracy of 94.5% with a kappa coefficient of 93.4%. The dropout regularization method prevented overfitting and mitigated over-activation of hidden nodes. Our initial results show that machine learning (ML) applications can be effective tools for improving natural resources management.
•DNN model optimized with Hyperband achieved 94.5% accuracy classifying LCLU from Sentinel-2 imagery in Morocco park•The use of 5 indices aided distinguishing classes with similar spectral signatures•Comparable 94.5% accuracy validated effectiveness of DL approach for detailed LCLU classification•Provided updated LCLU maps supporting evidence-based conservation and management strategies•Demonstrated potential of cloud-based deep learning with Sentinel-2 for automated environmental monitoring
Land-use and land-cover (LULC) classification using remote sensing imagery plays a vital role in many environment modeling and land-use inventories. In this study, a hybrid feature optimization ...algorithm along with a deep learning classifier is proposed to improve the performance of LULC classification, helping to predict wildlife habitat, deteriorating environmental quality, haphazard elements, etc. LULC classification is assessed using Sat 4, Sat 6 and Eurosat datasets. After the selection of remote-sensing images, normalization and histogram equalization methods are used to improve the quality of the images. Then, a hybrid optimization is accomplished by using the local Gabor binary pattern histogram sequence (LGBPHS), the histogram of oriented gradient (HOG) and Haralick texture features, for the feature extraction from the selected images. The benefits of this hybrid optimization are a high discriminative power and invariance to color and grayscale images. Next, a human group-based particle swarm optimization (PSO) algorithm is applied to select the optimal features, whose benefits are a fast convergence rate and ease of implementation. After selecting the optimal feature values, a long short-term memory (LSTM) network is utilized to classify the LULC classes. Experimental results showed that the human group-based PSO algorithm with a LSTM classifier effectively well differentiates the LULC classes in terms of classification accuracy, recall and precision. A maximum improvement of 6.03% on Sat 4 and 7.17% on Sat 6 in LULC classification is reached when the proposed human group-based PSO with LSTM is compared to individual LSTM, PSO with LSTM, and Human Group Optimization (HGO) with LSTM. Moreover, an improvement of 2.56% in accuracy is achieved, compared to the existing models, GoogleNet, Visual Geometric Group (VGG), AlexNet, ConvNet, when the proposed method is applied.
Due to rapid urbanization, agriculture drought, and environmental pollution, significant efforts have been focused on land use and land cover (LULC) multi-spectral scene classification. Identifying ...the changes in land use and land cover can facilitate updating the geographical maps. Besides, the technical challenges in multi-spectral images with implicit deep learning models due to the nature of multi-modal, it tackles real-life issues such as the collection of large-scale high-resolution data. The limited training samples are considered a crucial challenge in LULC deep learning classification as requiring a huge number of training samples to ensure the optimal learning procedure. The present work has focused on considering the fraction of multi-spectral data (EuroSAT data) and evaluated the exemplary CNN architectures such as shallow network (VGG16) and deep network (ResNet152V2) with different tuning variants along with the additional layers prior to classification layer to improve the optimal training of the networks to classify the multi-spectral data. The performance of the thirteen spectral bands of EuroSAT dataset that contain ten scene classes of land use and land cover were analyzed band-wise and combination of spectral bands. For the scene class ‘Sea & lake’ the best accuracy obtained was 96.17% with individual band B08A and 95.7% with Color Infra Red (CIR) band combination. The analysis provided in this work enables the remote sensing research community to boost performance even if the multi-spectral dataset size is small.
The potential to perform spatiotemporal analysis of the Earth's surface, fostered by a large amount of Earth Observation (EO) open data provided by space agencies, brings new perspectives to create ...innovative applications. Nevertheless, these big datasets pose some challenges regarding storage and analytical processing capabilities. The organization of these datasets as multidimensional data cubes represents the state-of-the-art in analysis-ready data regarding information extraction. EO data cubes can be defined as a set of time-series images associated with spatially aligned pixels along the temporal dimension. Some key technologies have been developed to take advantage of the data cube power. The Open Data Cube (ODC) framework and the Brazil Data Cube (BDC) platform provide capabilities to access and analyze EO data cubes. This paper introduces two new tools to facilitate the creation of land use and land over (LULC) maps using EO data cubes and Machine Learning techniques, and both built on top of ODC and BDC technologies. The first tool is a module that extends the ODC framework capabilities to lower the barriers to use Machine Learning (ML) algorithms with EO data. The second tool relies on integrating the R package named Satellite Image Time Series (sits) with ODC to enable the use of the data managed by the framework. Finally, water mask classification and LULC mapping applications are presented to demonstrate the processing capabilities of the tools.
We propose AiTLAS—an open-source, state-of-the-art toolbox for exploratory and predictive analysis of satellite imagery. It implements a range of deep-learning architectures and models tailored for ...the EO tasks illustrated in this case. The versatility and applicability of the toolbox are showcased in a variety of EO tasks, including image scene classification, semantic image segmentation, object detection, and crop type prediction. These use cases demonstrate the potential of the toolbox to support the complete data analysis pipeline starting from data preparation and understanding, through learning novel models or fine-tuning existing ones, using models for making predictions on unseen images, and up to analysis and understanding of the predictions and the predictive performance yielded by the models. AiTLAS brings the AI and EO communities together by facilitating the use of EO data in the AI community and accelerating the uptake of (advanced) machine-learning methods and approaches by EO experts. It achieves this by providing: (1) user-friendly, accessible, and interoperable resources for data analysis through easily configurable and readily usable pipelines; (2) standardized, verifiable, and reusable data handling, wrangling, and pre-processing approaches for constructing AI-ready data; (3) modular and configurable modeling approaches and (pre-trained) models; and (4) standardized and reproducible benchmark protocols including data and models.
Land use and land cover (LULC) mapping is a powerful tool for monitoring large areas. For the Amazon rainforest, automated mapping is of critical importance, as land cover is changing rapidly due to ...forest degradation and deforestation. Several research groups have addressed this challenge by conducting local surveys and producing maps using freely available remote sensing data. However, automating the process of large-scale land cover mapping remains one of the biggest challenges in the remote sensing community. One issue when using supervised learning is the scarcity of labeled training data. One way to address this problem is to make use of already available maps produced with (semi-) automated classifiers. This is also known as weakly supervised learning. The present study aims to develop novel methods for automated LULC classification in the cloud-prone Amazon basin (Brazil) based on the labels from the MapBiomas project, which include twelve classes. We investigate different fusion techniques for multi-spectral Sentinel-2 data and synthetic aperture radar Sentinel-1 time-series from 2018. The newly designed deep learning architectures—DeepForest-1 and DeepForest-2—utilize spatiotemporal characteristics, as well as multi-scale representations of the data. In several data scenarios, the models are compared to state-of-the-art (SotA) models, such as U-Net and DeepLab. The proposed networks reach an overall accuracy of up to 75.0%, similar to the SotA models. However, the novel approaches outperform the SotA models with respect to underrepresented classes. Forest, savanna and crop were mapped best, with F1 scores up to 85.0% when combining multi-modal data, compared to 81.6% reached by DeepLab. Furthermore, in a qualitative analysis, we highlight that the classifiers sometimes outperform the inaccurate labels.
Information regarding land use and land cover (LULC) is essential for regional land and forest management. The contribution of reliable LULC information remains a challenge depending on the use of ...remote sensing data and classification methods. This study conducted a multiclass LULC classification of an intricate mangrove ecosystem using the U-Net model with PlanetScope and Sentinel-2 imagery and compared it with an artificial neural network model. We mainly used the blue, green, red, and near-infrared bands, normalized difference vegetation index (NDVI), and normalized difference water index (NDWI) of each satellite image. The Digital Elevation Model (DEM) and Canopy Height Model (CHM) were also integrated to leverage the model performance in mixed ecosystems of mangrove and non-mangrove forest areas. Through a labeled image created from field ground truth points, the models were trained and evaluated using the metrics of overall accuracy, Intersection over Union, F1 score, precision, and recall of each class. The results demonstrated that the combination of PlanetScope bands, spectral indices, DEM, and CHM yielded superior performance for both the U-Net and ANN models, achieving a higher overall accuracy (94.05% and 92.82%), mean IoU (0.82 and 0.79), mean F1 scores (0.94 and 0.93), recall (0.94 and 0.93), and precision (0.94). In contrast, models utilizing the Sentinel-2 dataset showed lower overall accuracy (86.94% and 82.08%), mean IoU (0.71 and 0.63), mean F1 scores (0.87 and 0.81), recall (0.87 and 0.82), and precision (0.87 and 0.81). The best-classified image, which was produced by U-Net using the PlanetScope dataset, was exported to create an LULC map of the Wunbaik Mangrove Area in Myanmar.
The classification of land use and land cover (LULC) is a well-studied task within the domain of remote sensing and geographic information science. It traditionally relies on remotely sensed imagery ...and therefore models land cover classes with respect to their electromagnetic reflectances, aggregated in pixels. This paper introduces a methodology which enables the inclusion of geographical object semantics (from vector data) into the LULC classification procedure. As such, information on the types of geographic objects (e.g., Shop, Church, Peak, etc.) can improve LULC classification accuracy. In this paper, we demonstrate how semantics can be fused with imagery to classify LULC. Three experiments were performed to explore and highlight the impact and potential of semantics for this task. In each experiment CORINE LULC data was used as a ground truth and predicted using imagery from Sentinel-2 and semantics from LinkedGeoData using deep learning. Our results reveal that LULC can be classified from semantics only and that fusing semantics with imagery—Semantic Boosting—improved the classification with significantly higher LULC accuracies. The results show that some LULC classes are better predicted using only semantics, others with just imagery, and importantly much of the improvement was due to the ability to separate similar land use classes. A number of key considerations are discussed.
PDF
