期刊名称：ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences
印刷版ISSN：2194-9042
电子版ISSN：2194-9050
出版年度：2004
卷号：XXXV Part B7
页码：423-426
出版社：Copernicus Publications
摘要：The main purpose of this study is comparison of the potential of the Landsat7 ETM+ data and SPOT5-HRG image for forest area mapping at the scale of 1:25000 in northern of Iran. A Landsat7-ETM+ image and SPOT5-HRG in Pan and XS modes, with the spatial resolution of 5 and 10 meters, were analyzed. The images were almost simultaneous. The investigation on the image quality showed that there was a non-systematic misregistration between SPOT5-Pan and SPOT5-XS. There were no other noticeable radiometric and geometric distortions. Orthorectifications of the both satellite data were implemented using ephemeris data and a digital elevation model. The geocoded images were checked for reliability in comparison with the digital topographic map. Diverse suitable spectral transformation such as rationing, PCA, and Tasseled Cap transformation were performed on the images. To generate effective multispectral bands with better spatial resolution, the green, red and near infrared bands, which lie in the spectral range of panchromatic band, were fused with Pan using DIRS method as a radiometric approach. In order to estimate the potential of the satellite data precisely, a ground truth was prepared using aerial photographs. In this relation, 38 black and white aerial photographs at the scale of 1:7000 were taken especially for this project. These airphotos were orthorectified and interpreted. Image classifications were performed using supervised and a new hybrid approach (digital and visual). At first, the images were classified using maximum likelihood classifier without any knowledge of a-priori possibilities. The best band sets were selected using Bhatacharrya distance criterion and the defined training areas. To get advantage of contextual information and expert knowledge, a forest/non-forest classification was also carried out using visual interpretation at computer display. Based on comparison of the resulted maps and the ground truth, a better result had been achieved from the hybrid approach, up to 4%. The Landsat7 and SPOT5 data concluded an overall accuracy of 93% and 97 % respectively. It could be concluded that the SPOT5-HRG data is more appropriate than Landsat7 for forest mapping and updating at the scale of 1:25000. SPOT5 data permits fine and accurate forest/non-forest mapping. Revision of the forest road is also precisely possible trough SPOT5 image
