Synergistic Use of LiDAR and APEX Hyperspectral Data for High-Resolution Urban Land Cover Mapping

Abstract

Land cover mapping of the urban environment by means of remote sensing remains a distinct challenge due to the strong spectral heterogeneity and geometric complexity of urban scenes. Airborne imaging spectroscopy and laser altimetry have each made remarkable contributions to urban mapping but synergistic use of these relatively recent data sources in an urban context is still largely underexplored. In this study a synergistic workflow is presented to cope with the strong diversity of materials in urban areas, as well as with the presence of shadow. A high-resolution APEX hyperspectral image and a discrete waveform LiDAR dataset covering the Eastern part of Brussels were made available for this research. Firstly, a novel shadow detection method based on LiDAR intensity-APEX brightness thresholding is proposed and compared to commonly used approaches for shadow detection. A combination of intensity-brightness thresholding with DSM model-based shadow detection is shown to be an efficient approach for shadow mask delineation. To deal with spectral similarity of different types of urban materials and spectral distortion induced by shadow cover, supervised classification of shaded and sunlit areas is combined with iterative LiDAR post-classification correction. Results indicate that height, slope and roughness features contribute to improved classification accuracies in descending order of importance. Results of this study illustrate the potential of synergistic application of hyperspectral imagery and LiDAR for urban land cover mapping.

A Novel Spectral Library Pruning Technique for Spectral Unmixing of Urban Land Cover

“…In general, the accuracy values of our spectral unmixing results, obtained using a combination of AMUSES and MESMA, are in line with typical values found by other studies in the urban environment, using a variety of different image data, library pruning approaches and spectral unmixing techniques [16,47,[55][56][57]. In particular, overall RMSE values from our first experiment on the Berlin data closely match the performance reported by a study conducted on the same data using machine learning-based approaches [48].…”

“…However, significant mapping errors were still observed, particularly the overestimation of roof and underestimation of pavement and soil (Figure 8). The same problems are observed by other urban mapping studies (e.g., [16,48]) and are caused by the well-known phenomenon of spectral confusion between soil, pavement and roof materials, which are typically made of the same materials (e.g., asphalt roads and bitumen roofs [3,11]). In response to these limitations, many studies have already suggested and successfully proven the added benefits of fusing hyperspectral data with various other data, especially LiDAR data, for urban land cover mapping (e.g., [16,[56][57][58]).…”

“…New strategies are required to optimally make use of the wealth of spectral information already collected. In recent years, multiple urban spectral libraries have been collected using various sensors, at different spatial resolutions, for various cities and on different moments throughout the year (e.g., [11] and all references mentioned therein, [12][13][14][15][16]). All this information merged together in one vast generic urban spectral library would theoretically capture all possible endmember variability present within an urban image, and could consequently be used to process imagery from different study sites, sensors and timings, including those for which previously no endmember information has been extracted.…”

Mapping urban impervious surfaces from an airborne hyperspectral imagery using the object-oriented classification approach

“…This information derived from remotely sensed datasets leads to an improved observation and monitoring that can benefit applied urban planning and management [2]. However, land cover mapping of the urban environment by means of the remote sensing remains a distinct challenge due to the strong diversity of materials and objects, leading to spectral heterogeneity and geometric complexity of the urban areas as well as the presence of the shadow scenes [3]. In fact, several studies have showed that current satellite data provide limited information to deal with the richness, heterogeneity and complexity of the materials and forms in the urban areas.…”

Kernel Sparse Subspace Clustering with a Spatial Max Pooling Operation for Hyperspectral Remote Sensing Data Interpretation

“…Hyperspectral sensors acquire nearly continuous spectral bands with hundreds of spectral channels to capture the diagnostic information of land-cover materials, opening new possibilities for remote sensing applications such as mineral exploration, precision agriculture, and disaster monitoring [1][2][3]. As an unsupervised information extraction technique, clustering is a very useful tool for hyperspectral image (HSI) interpretation.…”