- Lin and Mills paper in the current PE & RS identifies surface roughness as the variable affecting LiDAR pulse width.
- Other factors such as slope and scan angle, amplitude and footprint size were fairly stable for different land cover types.
- This points to the promise of small footprint waveform LiDAR systems leading to more automated feature extraction capabilities.
In a very important paper just published in the January 2010 ASPRS PE&RS Journal, Yu-Ching Lin and Jon P. Mills identify surface roughness has having the most effect on the shape of the returned LiDAR-pulse waveform. When combined with the geometry this can lead to impressive automated feature extraction results.
In addition to surface roughness Lin and Mills looked at slope angle, scan angle and amplitude and footprint size. Of these surface roughness was the most significant factor affecting the return pulse width. Over relatively smooth surfaces they found there was no significant variation in pulse width behavior resulting from different footprint sizes. Pulse width was also relatively stable when amplitude, range distance, or scan angle were varied substantially.
They were able to achieve greater than 85% accuracy of classification for a number of different land-cover types, with greater than 94% for open vegetation areas. These results are very impressive and point to the promise of improved feature extraction automation using small footprint waveform systems.
This is a must read.
