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：2006
卷号：XXXVI Part 4
出版社：Copernicus Publications
摘要：In the present study CARTOSAT-1 data has been evaluated for retrieval of biophysical parameters like tree height, crown diameter, canopy density and canopy gap, which are critical variables for estimation of stand volume, biomass and generation of site index (indicator of site quality of growing stock). Based on this, empirical relationship were established between Cartosat derived estimated height & crown diameter and ground measured parameters. Cartosat-1 launched on 5 th may 2005 is the first Indian operational remote sensing satellite capable of providing in –orbit stereo images with 2.5 m nadir resolution and 27 km swath. This satellite carries two panchromatic cameras mounted on board to take pictures in two different angles so that stereo pairs are produced, using which the required digital elevation models (DEM), and Ortho images can be produced. Investigations have been carried out at a site located in Forest Research Institute Campus, Dehradun. The site has major dominating forest species growing in the country in different age plots such as Pinus roxbhurghii, Pinus caribea, Tectona gradis, Shorea robusta, Terminalia sps. and Eucalyptus spps. CARTOSAT- 1 panchromatic data was corrected for geometric distortion using various GPS points collected at different altitudinal location. Based on this, orthoimage and Digital Elevation Model (DEM) were derived which have been used for facilitating large-scale cartographic forest stock information /plot boundaries and retrieval of various biophysical parameters. Analysis using Gray Level Co-occurrence Matrix and image segregation was performed for separation of forest species, homogeneity of the forest stand classes, stratification and measurement of crown area, canopy gap and stand density. Field measurement for each forest species i.e. tree height, diameter at breast height (DBH), stand density, age, crown diameter and other relevant parameter were made from each stand/plot. Statistically high correlation coefficient (r 2 = 0.81) was found between DEM derived average tree stand height and actual measured tree stand height for all the forest species studied. Empirical models for prediction of DBH were developed and validated with Cartosat drived height and crown diameter. Measured and predicted values were found in close proximity to each other as is shown by the values of Pearson's coefficient. A local contrast co-occurrence measure (contrast, mean and variance) calculated for 3x3 and 5X5 pixel window has been found suitable for estimation of average crown diameter, stand density and canopy gap area for each forest stand class/plot. Texture image is found to be strongly related to structural attributes i.e. stand density, crown diameter and canopy gap of the forest tree stands. The results presented in this study show that CARTOSAT-1 stereo derived DEM and orthoimage have good potential for the estimation of stem volume, growing stock conditions, biomass and site index for important timber species in reasonable accuracy and cost effective manner
关键词：Biophysical parameters; Digital Elevation Model (DEM); Gray- Level Co-occurrence Matrix (GLCM)
