Vegetation management of utility corridors using high-resolution hyperspectral imaging and LiDAR

This study examines the use of high spatial resolution hyperspectral imagery in combination with light detection and ranging (LiDAR) data and digital aerial imagery for vegetation management of utility corridors. Two different classification methods, i.e. the support vector machines (SVM) and the spectral angle mapper (SAM) were applied on the datasets to test their ability for discrimination of various vegetation species. The SVM classifier performed best with an overall accuracy of 83% applied on the hyperspectral imagery. With inclusion of the LiDAR data the accuracy could be increased to 92%. Power lines were extracted from the LiDAR data and the conductor clearance was calculated. The results were merged with the SVM classification and a species map of vegetation that could cause potential damage to the power lines was generated. The results of this study show that an improved approach for vegetation management of utility corridors can be achieved by combining the spatial and spectral information of multi-source datasets.