DTM extraction of lidar returns via adaptive processing

Airborne light detection and ranging is emerging as a tool to provide accurate digital terrain models (DTMs) of forest areas, since it can penetrate beneath the canopy. Although traditional techniques, such as linear prediction, have shown to be robust type methods for the extraction of DTMs, they fail to effectively model terrain with steep slopes and large variability. In this paper, a modified linear prediction technique, followed by adaptive processing and refinement, is developed. A comparison with the traditional linear prediction method is provided along with statistical measures to ascertain the validity of the foregoing technique.