研究期間：10208~10307;Full Waveform LIDAR is an emerging active remote sensing technology. In comparison with conventional laser scanning systems, Full Waveform LIDAR not only can generate point clouds with higher density, it can also record the complete waveform of the backscattered signal echo. Therefore, in addition to range information, Full Waveform LIDAR data also provide a high potential for retrieving physical properties of objects and thus may be used for more sophisticated applications. However, existing LIDAR data processing and analysis are all designed for dealing with range measurements and are not adequate for processing and analyzing waveform information. Consequently, it is necessary to develop innovative data processing and analysis methodology in order to take full advantages of Full Waveform LIDAR remote sensing for sophisticated applications. This project proposes to undertake the advanced research and development of systematic data processing and feature extraction algorithms specifically designed for analyzing Full Waveform LIDAR data. Important topics of the research include: understanding of basic Full Waveform LIDAR data characteristics, waveform modeling and fitting, derivative waveform analysis and volumetric feature extraction. In addition, this research will also integrate developed data processing and analysis algorithms and procedures to perform related applications in vegetated and urban areas. These applications include object reconstruction, target identification, segmentation and classification, and vegetationcoved slope terrain reconstruction.
