Revolutionizing spatial data analysis: Unveiling a cutting-edge approach for batch coordinate transformation

• Published in: Data science and management Vol. 6; no. 4; pp. 214 - 226
• Main Authors: Buddhika, Waruna, Premawansha, Kumesha, Bandara, Thushara R., Samaranayake, Lakdinu, Dayananda, Viraj, Mudannayaka, Chameera, Priyadarshani, Shyama
• Format: Journal Article
• Language: English
• Published: Elsevier B.V 01.12.2023
• Subjects: Batch coordinate transformation; Coordinate transformation software; Datum; GNSS; Map projection; Batch coordinate transformation; Datum; Map projection; GNSS; Coordinate transformation software
• ISSN: 2666-7649; 2666-7649
• DOI: 10.1016/j.dsm.2023.07.001

Spatial data have become indispensable across various disciplines and provide crucial insights. These data are associated with coordinates and different coordinate systems. However, the diversity of geospatial data formats and disparate coordinate systems present challenges in harmonizing them for analysis. This study addresses the pressing need for an improved approach to the batch transformation of commonly used coordinate systems in Sri Lanka. First, we examine different coordinate transformation systems and identify their limitations. Subsequently, we present a comprehensive procedure for seamless coordinate transformations between various systems. To demonstrate the practical applications of our approach, we have developed a user-friendly desktop application capable of simultaneously converting input coordinates into multiple systems. This application streamlines the process for users unfamiliar with sophisticated geographic information system (GIS) applications and datum transformations. We validate the output coordinates transformed using our application by comparing them with those obtained from established applications such as ArcGIS and epsg.io. The results, which have been assessed based on the root mean squared error (RMSE) and mean absolute error (MAE), indicate high levels of accuracy, with a maximum RMSE of approximately 0.013 and a maximum MAE of approximately 0.008. A performance evaluation reveals that our approach is exceptionally efficient, outperforming ArcGIS and epsg.io by 40x and 60x, respectively. Moreover, the proposed pipeline holds potential as an infrastructure for developing web applications, mobile applications, and plugins for popular GIS platforms such as ArcGIS and QGIS.