Cutblock Update With High Resolution Satellite Images

摘要

Increasing demands on the forest resource and the high costs of re-inventory is driving the need for methods by which cost-effective update can be accomplished. The India IRS-1C/D panchromatic sensors with 5.8 m pixel resolution is the satellite with the highest spatial resolution next to the recent launch of IKONOS. The objective of this study was to compare photogrammetrically-derived cutblocks with those interpreted from IRS-1C/D images that were geometrically rectified using two methods: orthorectification and rubber sheeting. Aerial photographs at a scale of 1:20,000 were digitally aero-triangulated and displayed on a stereo photogrammetric workstation (DiAP viewer) from which cutblock information was derived. Although cutblock accuracy from the orthorectified IRS image was higher than from the rubber sheeted image, they were not statistically different. Cutblock areas were on average, within 2 ha and boundary placements ranged from 16 to 20 m of photogrammetric ground truth values. An assessment of individual cutblock delineations identified factors such as terrain shadow, residual snow cover and low density trees and vegetation will influence the spatial integrity of cutblock delineations.