Improving Ecotope Segmentation by Combining Topographic and Spectral Data

Scale-Specific Prediction of Topsoil Organic Carbon Contents Using Terrain Attributes and SCMaP Soil Reflectance Composites

There is a growing need for an area-wide knowledge of SOC contents in agricultural soils at the field scale for food security and monitoring long-term changes related to soil health and climate change. In Germany, SOC maps are mostly available with a spatial resolution of 250 m to 1 km2. The nationwide availability of both digital elevation models at various spatial resolutions and multi-temporal satellite imagery enables the derivation of multi-scale terrain attributes and (here: Landsat-based) multi-temporal soil reflectance composites (SRC) as explanatory variables. In the example of a Bavarian test of about 8000 km2, relations between 220 SOC content samples as well as different aggregation levels of the explanatory variables were analyzed for their scale-specific predictive power. The aggregation levels were generated by applying a region-growing segmentation procedure, and the SOC content prediction was realized by the Random Forest algorithm. In doing so, established approaches of (geographic) object-based image analysis (GEOBIA) and machine learning were combined. The modeling results revealed scale-specific differences. Compared to terrain attributes, the use of SRC parameters leads to a significant model improvement at field-related scale levels. The joint use of both terrain attributes and SRC parameters resulted in further model improvements. The best modeling variant is characterized by an accuracy of R2 = 0.84 and RMSE = 1.99.

DSSM: A Deep Neural Network with Spectrum Separable Module for Multi-Spectral Remote Sensing Image Segmentation

Over the past few years, deep learning algorithms have held immense promise for better multi-spectral (MS) optical remote sensing image (RSI) analysis. Most of the proposed models, based on convolutional neural network (CNN) and fully convolutional network (FCN), have been applied successfully on computer vision images (CVIs). However, there is still a lack of exploration of spectra correlation in MS RSIs. In this study, a deep neural network with a spectrum separable module (DSSM) is proposed for semantic segmentation, which enables the utilization of MS characteristics of RSIs. The experimental results obtained on Zurich and Potsdam datasets prove that the spectrum-separable module (SSM) extracts more informative spectral features, and the proposed approach improves the segmentation accuracy without increasing GPU consumption.

What Factors Shape Spatial Distribution of Biomass in Riparian Forests? Insights from a LiDAR Survey over a Large Area

Riparian ecosystems are home to a remarkable biodiversity, but have been degraded in many regions of the world. Vegetation biomass is central to several key functions of riparian systems. It is influenced by multiple factors, such as soil waterlogging, sediment input, flood, and human disturbance. However, knowledge is lacking on how these factors interact to shape spatial distribution of biomass in riparian forests. In this study, LiDAR data were used in an individual tree approach to map the aboveground biomass in riparian forests along 200 km of rivers in the Meuse catchment, in southern Belgium (Western Europe). Two approaches were tested, relying either on a LiDAR Canopy Height Model alone or in conjunction with a LiDAR point cloud. Cross-validated biomass relative mean square error for 0.3 ha plots were, respectively, 27% and 22% for the two approaches. Spatial distribution of biomass patterns were driven by parcel history (and particularly vegetation age), followed by land use and topographical or geomorphological variables. Overall, anthropogenic factors were dominant over natural factors. However, vegetation patches located in the lower parts of the riparian zone exhibited a lower biomass than those in higher locations at the same age, presumably due to a combination of a more intense disturbance regime and more limiting growing conditions in the lower parts of the riparian zone. Similar approaches to ours could be deployed in other regions in order to better understand how biomass distribution patterns vary according to the climatic, geological or cultural contexts.

About the Pitfall of Erroneous Validation Data in the Estimation of Confusion Matrices

Accuracy assessment of maps relies on the collection of validation data, i.e., a set of trusted points or spatial objects collected independently from the classified map. However, collecting spatially and thematically accurate dataset is often tedious and expensive. Despite good practices, those datasets are rarely error-prone. Errors in the reference dataset propagate to the probabilities estimated in the confusion matrices. Consequently, the estimates of the quality are biased: accuracy indices are overestimated if the errors are correlated and underestimated if the errors are conditionally independent. The first findings of our study highlight the fact that this bias could invalidate statistical tests of map accuracy assessment. Furthermore, correlated errors in the reference dataset induce unfair comparison of classifiers. A maximum entropy method is thus proposed to mitigate the propagation of errors from imperfect reference datasets. The proposed method is based on a theoretical framework which considers a trivariate probability table that links the observed confusion matrix, the confusion matrix of the reference dataset and the “real” confusion matrix. The method was tested with simulated thematic and geo-reference errors. It proved to reduce the bias to the level of the sampling uncertainty. The method was very efficient with geolocation errors because conditional independence of errors can reasonably be assumed. Thematic errors are more difficult to mitigate because they require the estimation of an additional parameter related to the amount of spatial correlation. In any case, while collecting additional trusted labels is usually expensive, our result show that the benefits for accuracy assessment are much larger than collecting a larger number of questionable reference data.

First 1-M Resolution Land Cover Map Labeling the Overlap in the 3rd Dimension: The 2018 Map for Wallonia

Land cover maps contribute to a large diversity of geospatial applications, including but not limited to land management, hydrology, land use planning, climate modeling and biodiversity monitoring. In densely populated and highly fragmented landscapes as observed in the Walloon region (Belgium), very high spatial resolution is required to depict all the infrastructures, buildings and most of the structural elements of the semi-natural landscapes (like hedges and small water bodies). Because of the resolution, the vertical dimension needs explicit handling to avoid discontinuities incompatible with many applications. For example, how to map a river flowing under a bridge? The particularity of our data is to provide a two-digit land cover code to label all the overlapping items. The identification of all the overlaps resulted from the combination of remote sensing image analysis and decision rules involving ancillary data. The final product is therefore semantically precise and accurate in terms of land cover description thanks to the addition of 24 classes on top of the 11 pure land cover classes. The quality of the map has been assessed using a state-of-the-art validation scheme. Its overall accuracy is as high as 91.5%, with an average producer’s accuracy of 86% and an average user’s accuracy of 91%.

Image Segmentation and Object-Based Image Analysis for Environmental Monitoring: Recent Areas of Interest, Researchers’ Views on the Future Priorities

Image segmentation and geographic object-based image analysis (GEOBIA) were proposed around the turn of the century as a means to analyze high-spatial-resolution remote sensing images. Since then, object-based approaches have been used to analyze a wide range of images for numerous applications. In this Editorial, we present some highlights of image segmentation and GEOBIA research from the last two years (2018–2019), including a Special Issue published in the journal Remote Sensing. As a final contribution of this special issue, we have shared the views of 45 other researchers (corresponding authors of published papers on GEOBIA in 2018–2019) on the current state and future priorities of this field, gathered through an online survey. Most researchers surveyed acknowledged that image segmentation/GEOBIA approaches have achieved a high level of maturity, although the need for more free user-friendly software and tools, further automation, better integration with new machine-learning approaches (including deep learning), and more suitable accuracy assessment methods was frequently pointed out.

How Response Designs and Class Proportions Affect the Accuracy of Validation Data

Reference data collected to validate land-cover maps are generally considered free of errors. In practice, however, they contain errors despite best efforts to minimize them. These errors propagate during accuracy assessment and tweak the validation results. For photo-interpreted reference data, the two most widely studied sources of error are systematic incorrect labeling and vigilance drops. How estimation errors, i.e., errors intrinsic to the response design, affect the accuracy of reference data is far less understood. In this paper, we analyzed the impact of estimation errors for two types of classification systems (binary and multiclass) as well as for two common response designs (point-based and partition-based) with a range of sub-sample sizes. Our quantitative results indicate that labeling errors due to proportion estimations should not be neglected. They further confirm that the accuracy of response designs depends on the class proportions within the sampling units, with complex landscapes being more prone to errors. As a result, response designs where the number of sub-samples is predefined and fixed are inefficient. To guarantee high accuracy standards of validation data with minimum data collection effort, we propose a new method to adapt the number of sub-samples for each sample during the validation process. In practice, sub-samples are incrementally selected and labeled until the estimated class proportions reach the desired level of confidence. As a result, less effort is spent on labeling univocal cases and the spared effort can be allocated to more ambiguous cases. This increases the reliability of reference data and of subsequent accuracy assessment. Across our study site, we demonstrated that such an approach could reduce the labeling effort by 50% to 75%, with greater gains in homogeneous landscapes. We contend that adopting this optimization approach will not only increase the efficiency of reference data collection, but will also help deliver more reliable accuracy estimates to the user community.

Region Merging Method for Remote Sensing Spectral Image Aided by Inter-Segment and Boundary Homogeneities

Image segmentation is extensively used in remote sensing spectral image processing. Most of the existing region merging methods assess the heterogeneity or homogeneity using global or pre-defined parameters, which lack the flexibility to further improve the goodness-of-fit. Recently, the local spectral angle (SA) threshold was used to produce promising segmentation results. However, this method falls short of considering the inherent relationship between adjacent segments. In order to overcome this limitation, an adaptive SA thresholds methods, which combines the inter-segment and boundary homogeneities of adjacent segment pairs by their respective weights to refine predetermined SA threshold, is employed in a hybrid segmentation framework to enhance the image segmentation accuracy. The proposed method can effectively improve the segmentation accuracy with different kinds of reference objects compared to the conventional segmentation approaches based on the global SA and local SA thresholds. The results of the visual comparison also reveal that our method can match more accurately with reference polygons of varied sizes and types.

The critical role of abiotic factors and human activities in the supply of ecosystem services in the ES matrix

In Western Europe, ecosystems have been shaped to maximise the supply of one specific biomass provisioning ecosystem service (ES), such as food or timber, with detrimental impacts on other ES. The ES approach has therefore been established to better understand the multiple interactions between human society and ecosystems. A variety of methods have been developed to assess ES and their relationships, for instance the ES matrix model based on land cover classes. This popular, flexible and simple method allows combining different data sources and easily comparing ES. However, in general, this method poorly takes into account landscape heterogeneity while abiotic factors and human activities seem to play an important role in ES supply. The objective of this paper is twofold: (1) to extent the methodology based on the ES matrix model by including abiotic factors and human activities and (2) to test the impacts of these two types of factors on ES supply and their relationships.

The assessment focused on the capacity of the forest to supply six ES depending on six types of soil ranging from productive soils to more constraining or less productive soils (i.e. abiotic factors) and two contrasting forest management strategies (i.e. human activities). This amended ES matrix was applied on one hand, to map the supply of ES and their relationships in four municipalities in the Ardenne ecoregion (Southern Belgium) and on the other hand, to investigate the impacts of three scenarios (i.e. three different management strategies) on ES supply and their relationships.

The amended ES matrix shows large differences in ES supply between the two forest management strategies on the more constraining and less productive soils, creating differences in the spatial pattern of ES. The changes in ES supply amongst the three scenarios and the current supply were quantified to identify the best management options.

In conclusion, one particular forest is not like another in terms of ES supply and their relationships. To capture this heterogeneity, we propose an amended ES matrix including abiotic factors and human activities. The maps, based on this matrix, allow identifying the hotspots (i.e. high capacity to supply different ES) and coldspots (i.e. low capacity to supply different ES or strong trade-offs between provisioning ES and regulating/cultural ES). Forest management should be adapted to the abiotic conditions, in particular in the coldspots, to ensure a more balanced supply of ES.