Automated object-based classification of topography from SRTM data

Abstract

We introduce an object-based method to automatically classify topography from SRTM data. The new method relies on the concept of decomposing land-surface complexity into more homogeneous domains. An elevation layer is automatically segmented and classified at three scale levels that represent domains of complexity by using self-adaptive, data-driven techniques. For each domain, scales in the data are detected with the help of local variance and segmentation is performed at these appropriate scales. Objects resulting from segmentation are partitioned into sub-domains based on thresholds given by the mean values of elevation and standard deviation of elevation respectively. Results resemble reasonably patterns of existing global and regional classifications, displaying a level of detail close to manually drawn maps. Statistical evaluation indicates that most of classes satisfy the regionalization requirements of maximizing internal homogeneity while minimizing external homogeneity. Most objects have boundaries matching natural discontinuities at regional level. The method is simple and fully automated. The input data consist of only one layer, which does not need any pre-processing. Both segmentation and classification rely on only two parameters: elevation and standard deviation of elevation. The methodology is implemented as a customized process for the eCognition® software, available as online download. The results are embedded in a web application with functionalities of visualization and download.

Semi-Automated Classification of Landform Elements in Armenia Based on SRTM DEM using K-Means Unsupervised Classification

“…SRTM DEM with 90 m spatial resolution reduces the possibilities for study of small elements of relief, large elements are represented properly (Wright et al 2006, Dragut, Eisank 2012, which in our opinion is sufficient for a terrain analysis covering the whole territory of Armenia. For this reason our analysis is based on the SRTM elevation dataset.…”

Morphological characterisation of reef types in Torres Strait and an assessment of their carbonate production

“…Geospatial procedures for classifying coral reef geomorphology lag behind terrestrial applications where semi-automated geomorphological mapping approaches are gradually replacing conventional manual interpretations due to improved geospatial techniques and increasing availability of high-quality digital elevation data (Drăguţ and Eisank, 2012;Zieger et al, 2009). Classification trees have been successfully used in geomorphological research because of their relative simplicity for the exploration of complex datasets and their predictive capability (Luoto and Hjort, 2005).…”

Quantitative studies of the morphology of the south Poland using Relief Index (RI)

“…Jasiewicz and Stepinski [32,33] presented novel geomorphons method for classification and mapping of landforms based on the idea that the earth surface can be described by the two complementary measures: relief-independent, local spatial pattern and the magnitude of the relief itself. To sum up one can conclude that issues of morphometric studies in geomorphology are still important and present (see more [34][35][36][37][38][39]). …”