|
|
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
|
|
|
Classification of UAV hyperspectral images using deep learning methods
Řádová, Martina ; Potůčková, Markéta (advisor) ; Kupková, Lucie (referee)
Diploma thesis "Classification of UAV hyperspectral images using deep learning methods" focuses on the classification methods, namely convolutional neural networks (CNN), of hyperspectral (HS) images. Based on a thorough literature review, a comprehensive overview on CNN utilisation in remote sensing is assembled as a basis for identifying suitable methods for the specific task of this thesis. Two methods with an open solution in programming language Python were selected - Capsule Network and U-Net. The main aim of this work is to verify the suitability of chosen methods for the classification of hyperspestral images. The images were acquired by sensors with high spatial resolution carried by a UAV over Krkonoše Mts. tundra. Important step was to prepare input HS data (54 bands, 9cm) to have suitable form for entering the network. Not all the required results were achieved due to the complexity of the Capsule Network architecture. The U-Net method was used in purpose of comparing and verifying the results. Accuracies retrieved from the U-Net overcome results achieved by traditionally used machine learning methods (SVM, ML, RF, etc). Overall accuracy for U-Net was higher than 90% where other mentioned methods did not get over 88%. Especially classes block fields and dwarf pine achieved higher...
|
|
|
SBU_1802
Fabiánek, Tomáš ; Hanuš, Jan ; Fajmon, Lukáš
Data Acquisition by FLIS (Flying Laboratory of Imaging Systems of Global change research institut CAS) over the localities of the villages of Krasíkovice and Kochánky was carried out on May 27. and 29., 2018. The reason for the acquisition of hyperspectral and LiDAR data was the verification of critical source sites in experimental catchments area IV. order or in parts with high concentrations of nitrates in water (including drainage).
|
|
|
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
|
|
|
FMI_1704
Hanuš, Jan ; Fabiánek, Tomáš ; Fajmon, Lukáš
Hyperspectral images with spatial resolution 1 m2 and spectral range from 400 to 2500 nm were used for analysis of health state of forest stands. The data was collected by the Flying Laboratory of Imaging Systems of Global change research institute CAS on 5. 8. 2017 over Nízký Jeseník mountain. The LiDAR scanner cloud of points was used for definition of forest surface structure.
|
|
|
SBU_1702
Hanuš, Jan
The purpose of aerial hyperspectral campaign with laser scanning of surface over the Dehtare area (Českomoravská vysočina highlands) was to obtain data for the study of water vegetation stress in the agricultural landscape. The data sets were scanned using the Flying Laboratory of Imaging Systems of the Global change research institute CAS in spectral range 380-2450nm, 8000-11500nm and LiDAR cloud density of 0.5 point per m2. Georeferenced and atmospherically corrected hyperspectral data was passed in spatial resolution 2 and 5m.
|
|
|
CUA_1701
Hanuš, Jan ; Fabiánek, Tomáš ; Fajmon, Lukáš
Aerial hyperspectral campaign with laser scanning of surface in Litvínov and Bílina dumps area was carried out on May 18., 2017. Flying Laboratory of Imaging Systems of the Global change research institute CAS was used on two location with area of 100 and 60 km2. Surface scanning took place in the spectral range 380–2450nm (in 46 and 141 bands), spatial resolution 0.5 and 1.25m and LiDAR cloud density of 4 point per m2. The resulting product was orthocertificated raster data in ENVI format (with atmospheric correction) and georeferenced LiDAR for physical geographic diversity of the mining area studies.
|
|
|
CGS_1703
Hanuš, Jan ; Fabiánek, Tomáš ; Fajmon, Lukáš
Data scanned by Flying Laboratory of Imaging Systems of Global change research institut CAS was used for soil degradation modeling. Aerial survey was running on 18. 5. 2018 over Lítov and Silvestr mines. The raster data with spectral range from 380 to 2450nm, thermal data 8000 to 11500nm and 3 point per m2 cloud density of LiDAR sensor. Georeferenced and atmospherically corrected hyperspectral data was passed in spatial resolution 0.8 and 2m.
|
|
|
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
|
|
|
Hyperspectral data for classification of alpine treeless vegetation in the Krkonoše Mts.
Andrštová, Martina ; Kupková, Lucie (advisor) ; Potůčková, Markéta (referee)
Hyperspectral data for classification of vegetation of alpine treeless in the Krkonoše Mts. ABSTRACT The Master Thesis is a part of the HyMountEcos project, which deals with a complex evaluation of mountain's ecosystems in the Giant Mountains National Park using the hyperspectral data. The area of interest is alpine treeless in the Giant Mountains National Park. The main goal of this thesis was to create detailed methodology for classification of vegetation cover using hyperspectral data from AISA DUAL and APEX sensors, to find a classification method, which would improve the accuracy of the results compared to those found in the literature, and to compare the accuracy reached with these two types of the data. Many different classification algorithms (Spectral Angle Mapper, Linear Spectral Unmixing, Support Vector Machine, MESMA a Neural Net) were applied and the classification results were statistically evaluated and compared in the next part of the work. The classification method Neural Net was found as the most accurate one, as it gives the most accurate results for APEX data (the overall accuracy 96 %, Kappa coefficient 0,95) as well as for AISA DUAL data (the overall accuracy 90 %, Kappa coefficient 0,88). The resulting accuracy of the classification (the overall one and also for some classes) reached...
|
guest ::
