Earth Sciences / Sciences de la Terre

Permanent URI for this communityhttps://laurentian.scholaris.ca/handle/10219/2111

Browse

Browsing Earth Sciences / Sciences de la Terre by Subject "Aeromagnetic data"

Filter results by typing the first few letters
Now showing 1 - 1 of 1
  • Thumbnail Image
    Item
    Convolutional neural networks applied to the interpretation of lineaments in aeromagnetic data
    (2022-01-01) Naprstek, Tomas; Smith, Richard S.
    Parameter estimation in aeromagnetics is an important tool for geologic interpretation. Due to aeromagnetic data being highly prevalent around the world, it can often be used to assist in understanding the geology of an area as a whole or for locating potential areas of further investigation for mineral exploration. Methods that automatically provide information such as the location and depth to the source of anomalies are useful to the interpretation, particularly in areas where a large number of anomalies exist. Unfortunately, many current methods rely on high-order derivatives and are therefore susceptible to noise in the data. Convolutional neural networks (CNNs) are a subset of machine-learning methods that are well-suited to image processing tasks, and they have been shown to be effective at interpreting other geophysical data, such as seismic sections. Following several similar successful approaches, we have developed a CNN methodology for estimating the location and depth of lineament-type anomalies in aeromagnetic maps. To train the CNN model, we used a synthetic aeromagnetic data modeler to vary the relevant physical parameters, and we developed a representative data set of approximately 1.4 million images. These were then used for training classification CNNs, with each class representing a small range of depth values. We first applied the model to a series of difficult synthetic data sets with varying amounts of noise, comparing the results against the tilt-depth method. We then applied the CNN model to a data set from northeastern Ontario, Canada, that contained a dike with known depth that was correctly estimated. This method is shown to be robust to noise, and it can easily be applied to new data sets using the trained model, which has been made publicly available.