Edge detection is one of the most important steps in the map interpretation of potential field data. In such a dataset, it is difficult to distinguish adjacent anomalous sources due to their field superposition. In particular, the presence of overlain shallow and
deep magnetic/gravity sources leads to strong and weak anomalies. In this paper, we present an improved filter, STDR, which utilises the ratio of the second-order vertical derivative to the second-order total horizontal derivative at the tilt angle equation. The maximum
and minimum values of this filter delineate the positive and negative anomalies, respectively. This novel filtering approach normalises the intensity of strong and weak anomalies, as well as anomalies with different depths and properties. Moreover, to better
illustrate the edges, its total horizontal derivative (THD_STDR) is also used. For positive and negative anomalies, the maximum value of the THD_STDR filter shows the edges of the anomalies. The potentiality of the proposed method is examined through both
synthetic and real case scenarios and the results are compared with a number of existing edge detector filters, namely TDR, THD_TDR, Theta and TDX. Due to substantial improvements in the filtering, STDR and its total horizontal derivative allow for
more accurate estimation of anomaly edges in comparison with the other filtering techniques. As a consequence, the interpretation of the potential field data is more feasible using the STDR filtering method.
Potential field methods produce anomaly maps with different magnitudes and depths that are typically contaminated by noise, making them hard to interpret. In order to highlight edges of the anomalies with different depths and magnitudes, data filtering techniques have received a great attention, in particular for mineral explorations. Filtering approaches render to explore more details from potential field data maps. In this respect, high pass filters are commonly used for enhancing the anomaly edges all of which utilize gradients of the potential field. In order to apply different filters on the potential field data, major attempts have been made to make a balance between noise and the signal obtained from a filtered image (Cooper & Cowan, 2006).