Classification of Herbaceous Vegetation Using Airborne Hyperspectral Imagery

Abstract

Abstract:Alkali landscapes hold an extremely fine-scale mosaic of several vegetation types, thus it seems challenging to separate these classes by remote sensing. Our aim was to test the applicability of different image classification methods of hyperspectral data in this complex situation. To reach the highest classification accuracy, we tested traditional image classifiers (maximum likelihood classifier-MLC), machine learning algorithms (support vector machine-SVM, random forest-RF) and feature extraction (minimum noise fraction (MNF)-transformation) on training datasets of different sizes. Digital images were acquired from an AISA EAGLE II hyperspectral sensor of 128 contiguous bands (400-1000 nm), a spectral sampling of 5 nm bandwidth and a ground pixel size of 1 m. For the classification, we established twenty vegetation classes based on the dominant species, canopy height, and total vegetation cover. Image classification was applied to the original and MNF (minimum noise fraction) transformed dataset with various training sample sizes between 10 and 30 pixels. In order to select the optimal number of the transformed features, we applied SVM, RF and MLC classification to 2-15 MNF transformed bands. In the case of the original bands, SVM and RF classifiers provided high accuracy irrespective of the number of the training pixels. We found that SVM and RF produced the best accuracy when using the first nine MNF transformed bands; involving further features did not increase classification accuracy. SVM and RF provided high accuracies with the transformed bands, especially in the case of the OPEN ACCESS Remote Sens. 2015, 7 2047 aggregated groups. Even MLC provided high accuracy with 30 training pixels (80.78%), but the use of a smaller training dataset (10 training pixels) significantly reduced the accuracy of classification (52.56%). Our results suggest that in alkali landscapes, the application of SVM is a feasible solution, as it provided the highest accuracies compared to RF and MLC. SVM was not sensitive in the training sample size, which makes it an adequate tool when only a limited number of training pixels are available for some classes.

Post-War Land Cover Changes and Fragmentation in Halgurd Sakran National Park (HSNP), Kurdistan Region of Iraq

“…Pixel-based MLC algorithm relies on a statistical model [56]. The MLC algorithm is categorizing objects into classes with the highest probability that a pixel belongs to a particular class [29][30][31]. Furthermore, it assumes that the image data for each class is normally distributed according to their spatial and spectral characteristics [56].…”

“…Unsupervised classification does not require training data, unlike supervised classification [27]. In object-based classification, an object starts with the combination of neighboring pixels into similar areas or depends on groups of pixels that represent the shapes and sizes, while in pixel-based classification method each pixel has a class that bases exclusively on the digital number of the pixel itself [29][30][31]. In contrast to traditional classification methods newer non-parametric algorithms such as random forest (RF) have been developed, which is least sensitive to the parameters and applies multiple decision trees to a set of training data [32].…”

An interactive tool for semi-automatic feature extraction of hyperspectral data

“…Seven LULC categories were analysed (asphalt, building, forest, grassland, synthetic grass, tennis court and shadow); we collected 100 points for the train and 1000 points for the ground truth datasets by LULC classes in ENVI (Exelis Visual Solutions, 2014) according to Burai et al (2015) [26]. Spectral band data were exported as a ROI file and processed in MS Excel with the HypDA add-in.…”

Assessing the Utility of Uav-Borne Hyperspectral Image and Photogrammetry Derived 3d Data for Wetland Species Distribution Quick Mapping

“…In terms of classification, probability-based Maximum Likelihood Classifier (MLC) is a widely used traditional supervised classification algorithm that offers reasonable classification accuracy (Binaghi, Gallo, Boschetti, & Brivio, 2005;Burai et al, 2015). Machine-learning classifiers such as decision tree, artificial neural network and support vector machine (SVM) are becoming popular due to their relatively accurate and robust performance in classification exercise (Dalponte, Orka, Gobakken, Gianelle, & Naesset, 2013;Wong & Fung, 2014).…”

“…Feature extraction methods reduce the number of band data through data transformation with the aim to extract maximal information from the original data. Minimum noise fraction (MNF) transformation and principal component analysis (PCA) are widely applied in hyperspectral image (Burai, Deá k, Valkó, & Tomor, 2015;Marcinkowska-Ochtyra et al, 2017). MNF is reported yielding higher accuracy than feature selection result or full band used result (Fabian E. Fassnacht et al, 2014).…”