|Rango, Albert - Al|
Submitted to: American Society for Photogrammetry and Remote Sensing Proceedings
Publication Type: Abstract Only
Publication Acceptance Date: 11/1/2011
Publication Date: 5/1/2011
Citation: Laliberte, A.S., Rango, A. 2011. Unmanned Aircraft Systems (UAS) for Vegetation Mapping: Very High Resolution Multispectral Imagery and Terrain Extraction. American Society for Photogrammetry and Remote Sensing 2011 Annual Meeting [abstract]. May 1-5, 2011, Milwaukee, Wisconsin, CDROM. Interpretive Summary:
Technical Abstract: In recent years, the interest in using unmanned aircraft systems (UAS) for remote sensing of natural resources has been growing considerably. Over the last few years, we have used a small UAS equipped with a low-cost digital camera to acquire thousands of images (6-8 cm GSD), which have been orthorectified and mosaicked using a custom processing procedure. The mosaics were used to create vegetation maps of arid rangelands. While the classification results have been positive, the low radiometric and spectral resolution imagery has its limits, and we are investigating 1) the fusion of the imagery with digital surface models (DSM) extracted from the images, and 2) a small multispectral camera. DSM extraction was done at two resolutions: a general terrain model of a 30 ha area at 1 m resolution, and a dense DSM at 6 cm resolution for deriving vegetation heights. The dense DSM extraction results in a lidar-like point cloud, which permits visualizing the data in 3D format and fuses the RGB values and the elevation data. This allows for differentiation of vegetation based on height information. Data derived from the DSM will be used to estimate parameters for hydrologic and erosion models. The lightweight multispectral camera integrated into the UAS acquires 10-bit data in 6 narrow bands ranging from blue to near infrared, with band centers at 450, 550, 650, 720, 750, and 850 nm. At a flying height of 210 m, the nominal GSD is 13 cm. We performed a camera calibration to determine the camera’s interior orientation parameters. Processing steps were developed to convert the proprietary image format into a format compatible for processing with photogrammetric software. Preliminary results for the UAS-acquired multispectral imagery and DSMs are promising. We report on the accuracy assessment of the DSMs and classification results of the multispectral imagery.