SIMULATION MODEL FOR THE ASSESSMENT OF DIRECT AND INDIRECT GEOREFERENCING TECHNIQUES IN ANALYTICAL PHOTOGRAMMETRY

Abstract

This paper compares between the direct and indirect georeferencing techniques in Photogrammetry bases on a simulation model. A flight plan is designed which consists of three strips with nine overlapped images for each strip by a (Canon 500D) digital camera with a resolution of 15 Mega Pixels.

The triangulation computations are carried out by using (ERDAS LPS) software, and the direct measurements are taken directly on the simulated model to substitute using GPS/INS in real case. Two computational tests have been implemented to evaluate the positional accuracy for the whole model and the Root Mean Square Error (RMSE) relating to (30) check points show that the indirect georeferencing is more accurate than the direct georeferencing. The computed RMSE of indirect georeferencing technique are (±0.0686 m., ± 0.0402 m.±, 0.3583 m.) in (X, Y, Z) respectively, while by direct georeferencing technique are (±0.1274 m., ± 0.1220 m., ± 0.5132 m.) in (X, Y, Z) respectively. Finally, the aim of this paper can be summarized as investigating the possibilities of using a simulation model to evaluate the applicability of direct and indirect georeferencing and analyzing the accuracy of both techniques.