When collecting magnetic data from UAV-platforms, the data set often contains “white spots”, i.e. part of survey lines where sensors for some reason did not successfully collect data. This projects aim to investigate and compare different methods for interpolating between data points, across an area of no data.
a. Estimate missing survey data using a series of methods
b. Examples of methods: fit functions (along track), use along-track gradient as a predictor, use the across-track horizontal gradient from neighboring lines to estimate value, expand/fit Fourier series, use single-point iterative fit
c. Compare with “truth” (e.g. remove some data, estimate it, and compare the removed data with the estimate)
d. Quantify the results from different schemes based on the relative error of the result and total computation time/computation time per datapoint.