|
|
Detection of anthropogenic terrain features in the summit area of the Luční and Studniční hora mountains (Krkonoše) with use of airborne laser scanning data
Krusová, Anna ; Lysák, Jakub (advisor) ; Potůčková, Markéta (referee)
The thesis focuses on detection of micrographic features using airborne laser scanning data. The aim of the thesis is to propose a novel method for detecting terrain features and to identify and describe terrain objects corresponding to anthropogenic interventions in the arctic-alpine tundra region in the summit area of the Luční and Studniční hora mountains in the Krkonoše Mountains. Based on existing studies, a new detection method of specific micrographic features was proposed by combining two types of data, i.e. point cloud and its raster representation. Initially, polygon approximations of terrain objects were derived from a digital terrain model, and their spatial delineation was further refined locally using point clouds. The output provides the most complete and accurate polygon delineation of the terrain features to date, considering the available data, the specific characteristics of the area, and in comparison to previously conducted studies. The proposed algorithm has identified hundreds of additional terrain features compared to the reference data. However, this number includes false positive features, which need to be manually eliminated. Key words: airborne laser scanning, micrographic features detection, archaeological prospection, digital terrain model, local relief model, Krkonoše...
|
|
|
Mapping relict arctic-alpine tundra vegetation from multitemporal LiDAR data
Šrollerů, Alex ; Potůčková, Markéta (advisor) ; Lysák, Jakub (referee)
The thesis focuses on metrics of vertical structure of vegetation derived from UAV LiDAR data and their use for multitemporal classification of selected species of arctic-alpine tundra in the Krkonoše Mountains. The metrics are selected based on a literature search focusing on low and shrubby stands. Random Forest algorithm and permutation feature importance, drop column importance and individual predictor performance is used to determine the suitability of metrics for distinguishing tundra vegetation. Subsequently, a fusion with multispectral data is performed and influence of the LiDAR derived variables on the refinement of classification results is determined. The use of metrics derived from a digital surface model obtained by image correlation of multispectral data is also examined. Maximum height followed by minimum height, canopy relief ratio and coefficient of variation yielded the best results, they achieved an overall classification accuracy of 67.3% for Bílá louka meadow and 62.3% for Úpské rašeliniště bog. Fusion with multispectral data led to an increase in overall accuracy up to 2 %. In case of vegetation structure derived from the digital surface model, similar results were achieved apart from higher stands. LiDAR data did not prove to be beneficial in distinguishing grass communities...
|
|
|
Deep learning for tree line ecotone mapping from remote sensing data
Dvořák, Jakub ; Potůčková, Markéta (advisor) ; Lefèvre, Sébastien (referee)
Deep learning is growing in popularity in the remote sensing community, especially as a classification algorithm. First part of this thesis describes deep neural networks commonly used for remote sensing classification and their various applications. Capabilities of selected geospatial software suites in relation to deep models are also discussed in this part. Theoretical findings from the first part of the thesis are validated using two deep convolutional Encoder-Decoder networks - U-Net and its proposed adaptation called KrakonosNet. They are used to perform a sematic segmentation of spruce trees and dwarf pine shrubs in the tree line ecotone of the Krkonoše Mountains, Czechia. A normalised digital surface model is employed for creation of sufficiently large amount of training data, while the classification itself is performed using only optical imagery with very high spatial resolution. Resulting classification is compared to a set of traditional remote sensing classifiers, namely Maximum Likelihood, Random Forest, and a Support Vector Machine. Both U-Net and KrakonosNet significantly outperform the other classifiers on this dataset and will be consequently used in a related research project. Key words deep learning, U-Net, Krkonoše mountains, classification, vegetation mapping, picea abies,...
|
|
|
Using ERT and GPR in polygonal patterned ground analysis
Široký, Jakub ; Křížek, Marek (advisor) ; Hartvich, Filip (referee)
Polygonal cryogenic structures cannot be investigated with conventional methods as they could be harmed during measurement. A real3D GPR and ERT non-destructive surveys were used to examine and prove applicability for topsoil covered ice-wedge pseudomorphs and coarse-grained sorted polygons (patterned ground). A list of processing tools and algorithm suitable for such environments was created and tested. The benefits of 3D measurements are illustrated on horizontal slices and pseudo3D visualisation of 3D Cube. Basic morphometry characteristics of both forms were collected. Abilities of geophysical imaging for advanced shape characterisations are discussed, too. The low-frequency measurements gave better results at both sites. Pseudomorphs, 2 wide and up to 6,5 long, were found penetrating depth bigger than 3,5 . Sorted polygons, 2,5 wide in diameter, were depicted locked by stony ring of width around 1 . Sorting depth extends up to 0,54 depth for sure, perhaps more.
|
|
|
Microevolutionary processes in the Czech endemic Campanula bohemica
Hanušová, Kateřina ; Suda, Jan (advisor) ; Vít, Petr (referee)
The genus Campanula L. - bellflower - is the largest group of the family Campanulaceae with a subcosmopolitan distribution and poorly resolved infrageneric classification. The evolutionary history of the genus has been shaped by a number of microevolutionary processes, including interspecific hybridization, genome duplication and geographical isolation, that resulted in the genesis of several endemic or geographically restricted species. The centre of endemism in the Czech Republic lies in subalpine altitudes of the Jeseníky and the Krkonoše (Giant) mountains, where three endemic taxa can be found: C. gelida Kovanda, C. rotundifolia L. subsp. sudetica (Hruby) Soó and C. bohemica Hruby. Despite their evolutionary and biogeographical value, there is a lack of information about their phenotypic variation, population structure, evolutionary history and processes acting in their populations. A critical assessment of these topics would require application of modern biosystematics tools. Campanula bohemica is an endangered neoendemics of higher altitudes in the Krkonoše Mts., closely related to C. scheuchzeri Vill., native to the Alps. The endemic species often grows in sympatry with related and morphologically similar C. rotundifolia. The incidence of intermediate morphotypes suggest that both species can...
|
|
|
Demand for the services of ski schools in Krkonoše mountains
Čápová, Hana ; Macáková, Libuše (advisor) ; Nečadová, Marta (referee)
The main aim is to describe and analyze a key determinants of the demand for services of ski schools in Krkonoše mountains. At the beginning I focus on a segment of ski schools itself. After, the hypotesis are set up, based on the defined determinants, and verified by a survey. The survey was held in a winter season 2008/ 2009 in Dolní Dvůr, Ski school SKi Baron. In addition the survey focusing on visit rate of Krkonoše mountains is mentioned. Next point is to compare two ski schools according to a different clients (SKOL MAX in Špindlerův Mlýn and Ski Baron in Dolní Dvůr).
|
guest ::
