VHR Remote Sensing for Impervious Surface Estimation

Tobias Leichtle from DLR investigates how WorldView-2 very high resolution satellite imagery sourced from European Space Imaging can be used for impervious surface estimation.

Urbanization and population growth have led to an increasing demand for land resources worldwide. This global process of land transformation is not only persistent in highly dynamic urban areas in developing countries, but also in cities of the Western world. The exponential increase of impervious surfaces for the establishment of settlements and transport infrastructure yields negative consequences in many domains, such as increased surface runoff and flood risk, decreasing groundwater recharge, or intensification of the urban heat island effect. Thus, exact and area-wide estimation of impervious surfaces is of high value and must be repeated regularly in order to provide up-to-date information. In this regard, very high resolution (VHR) remote sensing data offers a cost-effective solution for area-wide surveying and monitoring of impervious surfaces.

Traditional approaches for estimation of imperviousness are mostly based on aerial surveys, which are collected on demand at a high data cost. This data is commonly analysed by visual interpretation of the imagery which is also very time and labor consuming.

In contrast, modern space-borne earth observation systems provide data with very high spatial resolution of up to 30 centimeters and offer aerial-like capabilities for mapping of the land surface.

State of the art techniques for semi-automatic image interpretation can be employed on these data in order to reduce temporal efforts for image analysis.

Against this background, the Company for Remote Sensing and Environmental Research (SLU) together with the German Aerospace Center (DLR) developed an innovative product on impervious surfaces based on VHR remote sensing imagery. This was accomplished within the BMBF-funded project EO-CITI (Earth Observation for Monitoring Impervious Surfaces and Urban Structure in Cities, grant no. 033LK001A and 033LK001B). Suitable remote sensing imagery with very high spatial resolution was provided by European Space Imaging. The utilised WorldView-2 imagery offers 8 multi-spectral bands in the visible and near infrared region of the electromagnetic spectrum and a ground resolution of 50 centimeters.

Figure 1: WorldView-2 imagery from the city of Munich.

In urban environments, image analysis is challenging since different surface materials are concentrated in a heterogeneous and highly complex manner. In this context, the VHR imagery with 50 centimeters spatial resolution enables proper recognition and identification of individual objects. This allows the classification of pure pixels with regard to imperviousness, which is realized in an object-based image analysis (OBIA) approach. Subsequent to georeferencing, atmospheric correction as well as radiometric enhancement, spectral, textural and contextual features are calculated based on a multiresolution segmentation in the Trimble® eCognition® Developer software. These features are fed into a transferable knowledge-based classification tree in order to identify land cover with respect to different surface materials and subsequent binary classification of pervious and impervious surfaces.

Figure 2: Impervious surfaces product for the city of Munich.

The derived product illustrates that the OBIA classification of the WorldView-2 imagery with 50 centimeters spatial resolution offers a wealth of detail with regard to the characterization of impervious surfaces. This innovative product of imperviousness was validated against a traditional product based on a visual interpretation of an aerial survey provided by the city municipality of Munich. The VHR OBIA classification provides estimations of imperviousness with a Root Mean Square Error (RMSE) of 8.2 % compared to official data.

Figure 3: Comparison of aerial imagery and the derived product on imperviousness based on WorldView-2 imagery.

In practice, this product on imperviousness can be utilised by users from different domains. For example, administrations of cities, municipalities, or countries can employ this information for an objective documentation and monitoring of land consumption in order to support land management and political decision-making. In addition, it supports urban planning of public authorities as well as private companies like construction companies or planning offices. For example, rainwater treatment in terms of the proper dimensioning of the sewer network in general or flood risk management in case of heavy rain is a common field of application. Furthermore, the degree of imperviousness is an important driver of the urban heat island effect and thus, accurate spatial information provides assistance for the avoidance or the introduction of countermeasures of hotspots of urban heat. In order to ensure high usability and flexibility of the derived information product, the impervious surface estimates are provided as raster data in standard formats that can be used with any commercial or open source geographic information system (GIS) or any other image processing platform.

Another asset of this product on impervious surfaces, is that the cost can be reduced by half when using VHR remote sensing imagery with up to 30 centimeters compared to traditional aerial surveys.

In addition, semi-automated OBIA methods for the analysis of VHR data offer great potential for savings through the reduced effort of data processing.

About the Author

Tobias Leichtle received his B.Sc. degree in geography from the Ludwig-Maximilians-University Munich (Germany) and his M.Sc. degree in geospatial technologies from the University of Graz (Austria) in 2009 and 2013, respectively. He is a research scientist at the German Remote Sensing Data Center (DFD) within the German Aerospace Center (DLR) and pursues his Ph.D. at the Geography Institute of the Humboldt-University (HU) Berlin. Currently, he is working in several practice-oriented projects in close cooperation with the Company for Remote Sensing and Environmental Research (SLU). His research interests include the development of algorithms for change detection using very high resolution (VHR) remote sensing imagery as well as analysis of urban structure with focus on German and Chinese cities.

Share on facebook
Share on twitter
Share on linkedin

Related Stories

Scroll to Top

Architecture of ResNet34-UNet model

UNet architecture for semantic segmentation with ResNet34 as encoder or feature extraction part. ResNet34 is used as an encoder or feature extractor in the contracting path and the corresponding symmetric expanding path predicts the dense segmentation output.

Architecture of VGG16-UNet model

UNet architecture for semantic segmentation with VGG16 as the encoder or feature extractor. VGG16 is used as an encoder or feature extractor in the contracting path and the corresponding symmetric expanding path predicts the dense segmentation output.

Architecture of ResNet34-FCN model

In this model, ResNet34 is used for feature extraction and the FCN operation remains as is. The feature of ResNet architecture is exploited where just like VGG, as the number of filters double, the feature map size gets halved. This gives a similarity to VGG and ResNet architecture while supporting deeper architecture and addressing the issue of vanishing gradients while also being faster. The fully connected layer at the output of ResNet34 is not used and instead converted to fully convolutional layer by means of 1×1 convolution.

Architecture of VGG16-FCN model

In this model, VGG16 is used for feature extraction which also performs the function of an encoder. The fully connected layer of the VGG16 is not used and instead converted to fully convolutional layer by means of 1×1 convolution.