Search

The information on the topography of the seabed included in data from airborne lidar bathymetry can be used for the detection of changes occurring at the bottom of the basin and for the detection of objects deposited on it. Processing of full waveform data enables obtaining data on the water surface and the identification of the underwater situation. The classification process based on the Random Forest (RF) algorithm is presented using data from lidar bathymetry. The classification was performed in two independent approaches using input vector consisting of 16 features. In the first approach, the entire point cloud was classified, in the second the point cloud did not contain points of the water surface. In the classification, traits based on the full waveform and resulting from point cloud geometry were used. The quantitative efficiency of classification was verified through error matrices. The obtained efficiency (100% water surface, 99.9% seabed and 60% objects) of the point classification of the objects enables the possibility for using the RF algorithm for detection of objects on the seabed. In comparison to Support Vector Machines, the RF algorithm has better results in the detection of points on the objects in point cloud with water surface. [ABSTRACT FROM AUTHOR]

During recent years, many studies have been undertaken to investigate how spectral characteristics of forests can provide information on spatial patterns of tree species diversity (TSD). Important advances have been made, and significant relationships between TSD and remotely sensed indicators of net primary productivity and environmental heterogeneity have been reported. However, the season-dependence of these relationships has not yet been fully investigated, and the influence of phenology remains poorly understood. In this study, we aim to assess how the relationships between remote sensing indicators and TSD depend on the season of the year. TSD measures, including species richness, Shannon’s diversity and Simpson’s diversity, were determined for 82 field plots in the Afromontane cloud forests of Taita Hills, Kenya. A time series of 15 Landsat images were used to calculate a set of spectral and heterogeneity metrics. The relationship between remote-sensing metrics and TSD measures was analysed by simple and multivariate regression analysis. We conclude that the relationships between remote-sensing metrics and TSD are season-dependent. Hence, it is demonstrated the date of image acquisition is an important aspect to be considered in biodiversity studies. Given that the dependence of the relationships is closely linked to climate seasonality defining vegetation phenology, the relationships may also vary according to geographical conditions. [ABSTRACT FROM AUTHOR]

The aim of this study was to investigate to which degree the accuracy of vegetation classification could be improved by combining optical satellite data and airborne laser scanner (ALS) data, compared with using satellite data only. A Satellite Pour l'Observation de la Terre (SPOT) 5 scene and Leica ALS 50-II data from 2009, covering a test area in the mid-Sweden (latitude 60° 43' N, longitude 16° 43' E), were used in maximum likelihood and decision tree classifications. Training and validation data were obtained from the interpretation of digital aerial photo stereo models. Combination of SPOT and ALS data gave classification accuracies up to 72%, compared with 56% using only SPOT data. This indicates that integrating features from large area laser scanning may lead to significant improvements in satellite data-based vegetation classifications. [ABSTRACT FROM AUTHOR]