Journal
SPATIAL INFORMATION RESEARCH
Volume 28, Issue 1, Pages 113-127Publisher
SPRINGER SINGAPORE PTE LTD
DOI: 10.1007/s41324-019-00274-0
Keywords
Topographic correction; Tandem-X DEM; Illumination condition (IL); Land use land cover (LULC) classification; Support vector machine (SVM)
Categories
Funding
- German Aerospace Center (DLR) [DEM_HYDR1925]
Ask authors/readers for more resources
Topographic shadows of irregular mountains obstruct the analysis of satellite images in hilly areas. Due to this effect, there is high variability in the reflectance response of similar vegetation types, i.e. sunny areas show more than actual reflectance, whereas shaded areas show less than expected reflectance. In this study, we have evaluated the performance of five topographic correction methods, namely Cosine, C-Huang Wei, semi empirical C, SCS + C and Variable Empirical Coefficient Algorithm (VECA) depending on the solar incidence angle and exitance angle. The two well-known digital elevation models (DEM) i.e. Shuttle Radar Topography Mission (SRTM) and TanDEM-X have been used for the study. The efficiency of the correction methods is assessed on Landsat-8 satellite image using three criteria: visual interpretation, statistical assessment and classification accuracy assessment. As seen from the statistical analysis, VECA and C-correction method provides good correction of topography for both SRTM and TanDEM-X elevation models. We have used support vector machine (SVM) classifier for classification of topographically corrected images. Our results show that VECA and C-correction method increased classification accuracy from 65.60% (for uncorrected image) to 82.40% for SRTM and 64.00% to 80.00% for TanDEM-X respectively. The highest accuracy of classification is obtained using VECA/C method with SRTM DEM. However, the SCS + C-correction method impressively reduced the visual topography effects.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available