COMPARISON OF PRINCIPAL GEODETIC DISTANCE CALCULATION METHODS FOR AUTOMATED PROVINCE ASSIGNMENT IN TURKEY


ESENBUGA O. G., Akoguz A., COLAK E., VAROL B., Erol B.

16th International Multidisciplinary Scientific Geoconference (SGEM 2016), Albena, Bulgaristan, 30 Haziran - 06 Temmuz 2016, ss.141-148 identifier

  • Yayın Türü: Bildiri / Tam Metin Bildiri
  • Basıldığı Şehir: Albena
  • Basıldığı Ülke: Bulgaristan
  • Sayfa Sayıları: ss.141-148
  • İstanbul Teknik Üniversitesi Adresli: Evet

Özet

The distance between two points on the Earth or a reference surface, defined by their latitude and longitude, can be calculated by using various methods. Regarding their computational complexity, calculation time and errors stem from the assumptions in their formulations; these methods reveal advantages or drawbacks in practice. In this paper, a number of spheroidal and spherical methods as well as a proposed method are investigated. Comparisons relative to their error percentage and duration time of calculation have been performed via C Programming Language and GNU C Compiler under Linux Operating System. In addition, Box-Whisker Analyses have been performed for each method. The experiments have shown that, the increment of the baseline length results in a rise of error percentage. However, most accurate results have been accomplished with Vincenty's Formulas. The formulations based on spherical or planar assumptions for computing the baselines do not reveal very accurate results; however, they are more desirable due to their easier interpretability and calculation speed for specific type of applications. Based on the findings described in this paper, appropriate calculation methods can be easily chosen depending on the purpose of application exemplary, in navigation, environmental science, geodetic and cartographic applications as well as surveying. The main purpose of the research beyond is to decide upon an optimal distance calculation method for academic project "Automated Province Assignment for Satellite Images Based Cascaded K Nearest Neighbor Algorithm: A Case Study of Turkey".