Raster was Yesterday: Using Vector Engines to Process Geophysical Data

Tools for the processing of raster data are well developed, but noisy data still pose considerable challenges. If anomalies are broken up into isolated individual readings, for example due to high noise levels, it may still be possible for a human interpreter to recognize the isolated readings as being part of a single anomaly. However, such a concept of neighbourhood is difficult to implement with raster tools and an alternative, vector-based approach is presented here. By converting the measured raster data into polygons, it is possible to undertake shape and neighbourhood analysis to process the data. This allows discriminating, reshaping and merging of the anomalies based on their spatial location relative to each other (neighbourhood) and with respect to the size of each anomaly. The added advantage of this approach is the possibility to use the processed vector data as a basis for interpretation and visualization diagrams in two- and three dimensions. This method is applied to the GPR survey of a necropolis at Pessinus, showing several types of grave monuments.