Exploration of Glacial Landforms by Object-Based Image Analysis and Spectral Parameters of Digital Elevation Model

Abstract

Glacial landforms are a significant element of landscape in many regions of Earth. The increasing availability of high-resolution digital elevation models (DEMs) provides an opportunity to develop automated methods of glacial landscape exploration and classification. In this study, we aimed to: 1) identify glacial landforms based on high-resolution DEM datasets; 2) determine relevant geomorphometric and spectral parameters and object-based features for the mapping of glacial landforms; and 3) develop an accurate workflow for glacial landform classification based on DEM. The developed methodology included the extraction of secondary features from DEM, feature selection with the Boruta algorithm, object-based image analysis, and random forest supervised classification. We applied the workflow for three study sites: one in Svalbard and two in Poland. It allowed the identification of six categories of glacial landforms: till plains, end moraines, hummocky moraines, outwash/glaciolacustrine plains, valleys, and kettle holes. The majority of relevant secondary features represented DEM spectral parameters calculated from 2-D Fourier analysis. The supervised classification models with the highest performance exhibited up to 96% overall accuracy with regard to a groundtruth dataset. This study showed that glacial landforms can be identified using novel image-processing methodology and spectral parameters of high-resolution DEM. The complete classification workflow developed herein provides a solution for the transparent generation of thematic maps of glacial landforms that may be reproducible and transferrable to various glacial regions worldwide.

Application of Airborne LiDAR Data to the Archaeology of Agrarian Land Use: The Case Study of the Early Medieval Microregion of Bled (Slovenia)

“…In addition, climate (temperature, precipitation) and human impact are also very important for the distribution of plant species, as are many other factors. Alternative types of similar predictor variables include airborne LiDAR-derived feature detection used to identify landslides [112], spectral parameters of airborne LiDAR data applied for detection of glacial landforms [113], and object-based image analysis applied for volcanic and glacial landforms mapping [114]. TPI and mSWI methods in no way intend to compete with verified and established methods of environmental archaeology, such as archaeopalynology, archaeobotany, or archaeozoology, e.g., [115][116][117][118][119][120].…”

Overview of Research Progress of Digital Image Processing Technology

“…Through the preliminary processing of the original information, many kinds of characteristic information from the image can be analyzed. The main types of analysis are landform analysis [5], vegetation analysis [6] and cloud analysis [7], which are derived from the wide use of high-resolution digital elevation model (DEM). At the same time, through shape correction and improving color tolerance, these high-resolution images can still retain most of the useful information after reprocessing.…”

Quantification of Loess Landforms from Three-Dimensional Landscape Pattern Perspective by Using DEMs

“…The principle of MRS is to satisfy the conditions of minimum average heterogeneity of pixels and maximum homogeneity between pixels in the object; this objective is realized by combining adjacent pixels or small segmentation objects [52]. Multiresolution segmentation is a bottom-up region merging technology starting from a pixel object [53,54]. Small image objects can be merged into larger objects.…”