|
|
The influence of spectral resolution on land cover classification in Krkonoše Mts. tundra
Palúchová, Miroslava ; Červená, Lucie (advisor) ; Kupková, Lucie (referee)
The influence of spectral resolution on land cover classification in Krkonoše Mts. tundra Abstract The aim of this diploma thesis was to specify the spectral resolution requirements for classification and to identify the most important spectral bands to discriminate classes of the predefined legend. Aerial hyperspectral data acquired by AisaDUAL sensor were used. The method applied for the selection of the important bands was discriminant analysis performed in IBM SPSS Statistics. The most discriminative bands were found in intervals 1500-1750 nm (beginning of SWIR), 1100- 1300 nm (longer wavelengths of NIR), 670-760 (red-edge) and 500-600 nm (green light). The classification of the selected bands was realized in ENVI 5.4 using the Support Vector Machine classifier, achieving overall accuracy of 80,54 %, Kappa coefficient 0,7755. The suitability of available satellite data for the classification of tundra vegetation in Krkonoše mountains based on spectral resolution was evaluated as well. Keywords: tundra, Krkonoše, classification, spectral resolution, class separability, discriminant analysis, hyperspectral data
|
|
|
Application of imaging spectroscopy in monitoring of vegetation stress caused by soil pollutants in the Sokolov lignite basin
Mišurec, Jan ; Kupková, Lucie (advisor) ; Pavelka, Karel (referee) ; Homolová, Lucie (referee)
Forests can be considered as one of the most important Earth's ecosystems not only because of oxygen production and carbon sequestration via photosynthesis, but also as a source of many natural resources (such as wood) and as a habitat of many specific plants and animals. Monitoring of forest health status is thus crucial activity for keeping all production and ecosystem functions of forests. The main aim of the thesis is development of an alternative approach for forest health status based on airborne hyperspectral data (HyMap) analysis supported by field sampling. The proposed approach tries to use similar vegetation parameters which are used in case of the current methods of forest health status assessment based on field inspections. It is believed that importance of such new methods will significantly increase in the time when the planned satellite hyperspectral missions (e.g. EnMap) will move into operational phase. The developed forest health monitoring approach is practically demonstrated on mature Norway spruce (Picea abies L. Karst) forests of the Sokolov lignite basin which were affected by long-term coal mining and heavy industry and therefore high variability of forest health status was assumed in this case. Two leaf level radiative transfer models were used for simulating spectral...
|
|
| |
|
|
Forest species determination from satellite data
Launer, Michal ; Kolář, Jan (advisor) ; Kupková, Lucie (referee) ; Brodský, Lukáš (referee)
Forest species determination from satellite data Abstract Examining the species composition of forests from satellite imagery is constantly evolving. The new ways of exploring forests from the satellites make it easier for foresters to maintain a more accurate and up-to-date overview of the state of forests. In this work, the research was made on the forests in the cadastral territories of Osvětimany and Buchlovice in the Chřiby Mountains in the Czech Republic. In this work, data from the Landsat-8 satellite from three seasons and the Maximum Likelihood Classification method were used. The reference maps were used as reference data. The method of work consists in the fact that 6 frames were classified with the help of training sets using Maximum Likehood Classification. Subsequently, the pixels which were at least 4 times out of 6 ranked in the same class after the classification were selected. Based on these pixels, artificial training sets were calculated for each of the 6 frames, and they were used for another classification with the expectation of better results. The accuracy of the individual classification frames was verified by an error matrix on the crop maps. Keywords: remote sensing, forest canopy, forest tree types, forestry map
|
|
|
Validation of global forest change detection databases
Šístek, Petr ; Štych, Přemysl (advisor) ; Kupková, Lucie (referee)
Validation of global forest change detection databases Abstract The main aim of the thesis is to validate selected databases of changes in forest areas based on the analysis of satellite imagery time series in the Czech Republic. For this purpose we are using databases of M. C. Hansen and P. V. Potapov which are mapping the evolution of forest areas internationally. For the purposes of validation, we have proposed a methodology primarily based on historical ortophotographs from 2000-2012, the same time period which is documented in the validated databases. The results obtained were statistically processed, allowing to assess the accuracy of validated databases. At the end of the thesis, we are discussing the causes of identified inaccuracies and presented with recommendations for future improvements of detection of changes in forest areas. Keywords: validation, forest, land cover, change detection, Hansen, Potapov
|
|
|
Classification of selected vegetation land cover categories in the Krkonoše Mts. tundra from Sentinel-2A imagery using multi-temporal data
Roubalová, Markéta ; Kupková, Lucie (advisor) ; Suchá, Renáta (referee)
Classification of selected vegetation land cover categories in the Krkonoše Mts. Tundra from Sentinel-2A imagery using multitemporal data Abstract The aim of this thesis was to evaluate the possibilities of multi-temporal approach to improve classification accuracy of vegetation cover in eastern tundra in the Krkonoše Mts. National Park. Sentinel-2A imagery - 10 spectral bands with spatial resolution 10 and 20 m - was used. The classification legend was created by a botanist of the national park. Maximum likelihood classification for 11 categories of vegetation land cover was executed in software ENVI 5.3. The overall accuracy of the best classification result was 53,4 % which is similar result as in the case of single image classification (overall accuracy was 51,2 %). Key words: multi-temporal classification, vegetation, spectral features, Sentinel-2A, tundra, The Krkonoše Mts. National Park
|
|
|
Spectroscopic and statistical methods for detailed mapping of vegetation in the Krkonoše Mountains National Park
Minárčik, Miroslav ; Kupková, Lucie (advisor) ; Potůčková, Markéta (referee)
Spectroscopic and statistical methods for detailed mapping of vegetation in the Krkonoše Mountains National Park Abstract The diploma thesis is focused on the detailed classification of vegetation in the Krkonoše Mountains National Park using DCA (Detrended Correspondence Analysis) ordination method in combination with PLSR (Partial Least Square Regression) analysis. The resulting regression analysis values were applied to the hyperspectral imagery (APEX). The classification results were compared to the supervised classification SVM (Support Vector Machine). The DCA method was able to explain 16,3 % variation for the first three axes of the ordination analysis. Subsequent correlation with spectral data of vegetation showed that the highest confidence value reached the first axis correlated with field spectral data (R2 = 0,56). The resulting classification map created using RGB composition showed detailed information on the composition of the vegetation. Keywords: The Krkonoše Mountains National Park, classification, APEX, DCA, PLSR, hyperspectral data
|
|
|
Classification of meadow vegetation in the Krkonoše Mts. using aerial hyperspectral data and support vector machines classifier
Hromádková, Lucie ; Kupková, Lucie (advisor) ; Potůčková, Markéta (referee)
Meadow vegetation in the Krkonoše Mountains National Park is classified in this master thesis using aerial hyperspectral data from sensor AISA and Support Vector Machines (SVM) and Neural Networks (NN) classification algorithms. The main goals of the master thesis are to determine the best settings of SVM parameters and to propose an ideal design for a training dataset for this classification algorithm and mapping of the meadows in the Krkonoše mountains. The criterion of the tests will be the result of classification accuracy (confusion matrices and kappa coefficient). The additional goal of the master thesis is to compare performances of both utilized classifiers, especially regarding the amount of training pixels necessary for successful classification of the mountainous meadow vegetation. Classification maps of the area of interest and Python scripts are the main outputs of the master thesis. These outputs will be handed over to the Administration of the Krkonoše Mountains National Park for further utilization in the monitoring and protecting these valuable meadow vegetation communities. Key words: hyperspectral data, AISA, Support Vector Machines, Neural Networks, training dataset, mountainous meadow vegetation
|
|
|
Comparison of NATURA 2000 mapping with Earth Observation mapping (Case study of tundra in the Krkonoše Mts.)
Ondrušková, Kateřina ; Kupková, Lucie (advisor) ; Červená, Lucie (referee)
Comparison of NATURA 2000 mapping with Earth observation mapping (Case study of tundra in the Krkonoše Mts.) Abstract The bachelor thesis is focused on comparing Natura 2000 mapping with results of classification of Landsat 8 and WorldView-2 satellite images with different spatial resolution in the case study of tundra in the Krkonoše Mts. Satellite images were classified using Maximum Likelihood supervised classification and ISODATA unsupervised classification. The aim of the thesis was to find out what categories of Natura 2000 mapping are detectable using satellite images. For all classifications two levels of modified legend of the Natura 2000 mapping were used. The best results for both satellite images were achieved by unsupervised classification on level 1 of the legend - overall accuracy for Landsat 8 image was 64,1 % and for Word-View-2 image 67,16 %. Software ENVI 5.1 was used for all classifications. Keywords: Earth observation, supervised classification, unsupervised classification, legend, classification accuracy, NATURA mapping
|
|
|
Utilization of airborne laser scanning data for the detection of agrarian forms of relief in the Giant Mountains
Jebavá, Lucie ; Kupková, Lucie (advisor) ; Potůčková, Markéta (referee)
Utilization of airborne laser scanning data for the detection of agrarian forms of relief in the Giant Mountains Abstract This thesis deals with the possible use of airborne laser scanning data for the detection of agrarian forms of relief in the Giant Mountains. The main research method is the analy- sis of the digital terrain model and digital surface model in the area of the Giant Mountains in the software ArcGIS 10.0. The analysis is based on specific functions in ArcGIS software (Slope, Curvature) as well as on combinations of rasters of shaded reliefs which led to the detection of further agrarian forms. To verify the precision of agrarian forms detection field verification was used. Based on verification and accuracy assessment the results can be designated as a very satisfactory. The Administration of the Giant Moun- tains National Park can use the research results to quantify the agrarian forms of relief, to improve their protection and, possibly also for further research and management. Keywords: airborne laser scanning, agrarian forms of relief, anthropogenic geo- morphology
|
guest ::
RSS feed.