Terrain Eigenvector Dyad Analysis

摘要

In the study a surface-fitting algorithm, the dyad-spline method, was applied to typical terrain elevation data in order to evaluate the method's worth for incorporation in an All-Weather Mapping System. Various combinations of data spacing and representative amounts of simulated measurement noise were considered. In general the algorithm proved to be relatively impervious to measurement noise, and was capable of reproducing the true terrain with good fidelity. In a typical case, 5-mile spacing was employed for flight-line data, with nine rows of intermediate data per flight line. Normal noise was added to all data, with sigma = 10 ft along the flight lines, and with sigma = 50 ft along the intermediate lines. Under these conditions the dyad-spline method was able to reproduce the original terrain at interpolated points to within 105.3 ft RMS. The application of a weighting matrix technique to the variable quality data was also examined, and it was found that the minimum error was obtained when all data were weighted equally. (Author)