Maximizing grid-on-grid transformation performance with regularized regression techniques for integrating multi-source geospatial data

ABSTRACT

Three-dimensional coordinate transformations are required to harmonise different types of geospatial data accurately. Developing a mathematical model for data fusion relies on ground control points that produce discrepancies between the physical reality and depicted elements. The disparities between the two coordinate systems are known as the grid-to-ground issue that can be minimised by grid-to-grid or map-to-map transformation. This study develops simplified and rapid models for map-matching with global coordinates using regularised regression approaches to improve the accuracy and reliability of geospatial data. The results indicated that the proposed approach provides superior performance and employs any area with high accuracy.

KEYWORDS:

• 2D similarity transformation
• conformal polynomial
• GNSS
• datum shifts
• regularised regression approaches

Disclosure statement

No potential conflict of interest was reported by the author(s).