Using of remote sensing filters in processing of the potential filed images

Remote sensing is the science and art of the gathering information about an object, area or phenomena by means of a device that is not in contact with the object, area and phenomena under investigation. In remote sensing, the interaction between an energy source (electromagnetic) and surface features (trees, rocks, soil) is recorded. Image processing can be considered as an image-to-information-mapping procedure, which provides the image for future analysis. Image processing techniques usually employ a filter for selecting desired information. Filters are used for spatial image enhancement, for example, to reduce noise or to sharpen blurred images. Such a processing technique works on pixel values and produces a filtered image which the pixel values of which depend on its former neighbors. In this regard, the sunshading, viewshed, edge enhancement, and majority filters are most commonly used as image processing filters. In contrast with remote sensing, geophysical methods employ human agents for data collection and interpretation. Potential geophysical field data, such as aeromagnetic and gravity data are collected by both government Geological Surveys and mineral exploration companies for a wide range of purposes, including geological mapping and mineral exploration. Since the measured data are presented in the form of a contour map, it can be considered as an image and image processing can apply on it. In addition, these images possess a large number of geological features, such as dykes, faults, and folding, which can disappear with disturbing noises. Consequently, the use of image processing techniques for the purposes of noise reduction and edge enhancement of different geological features is necessary. In geophysical literature, derivative filters and local phase filters are the most applicable filters applied on potential field images. The main intent of this study is to develop remote sensing filters, mentioned above, on potential files images.
Sunshading is a process in which the observed data are considered as topography data that is illuminated by a light source, such as sun. The elevation and azimuth of a ‘sun’, or light source, are chosen by the interpreter, and the resulting reflectance from the data surface (the elevations of which are proportional to the values of the data to be interpreted) is calculated. Linear features that lie at 90 degrees of the azimuth are enhanced, while those which lie parallel to it become less apparent. The selection of the azimuth and elevation of the sun is normally done in an interactive manner by the interpreter. There are many different algorithms for determining the reflectance from the surface based on a variety of physical models. In this regard, the Lambertian reflector is most admissible regarding geosciences data is given by
 
where , ,  is the sun elevation (measured from the vertical), and is the azimuth (measured anticlockwise from East). Notably, p and q are any two orthogonal gradients of the data, and may be calculated in either the space or frequency domains.
The viewshed of a dataset is the region around a given observation point that is visible from that location. The viewshed is used here as a data enhancement tool to aid in the interpretation of geophysical potential field data. The fraction of the area of a moving window visible from each point on the dataset (considered as if it were topography) is computed.
Edge enhancement is particularly suited to the analysis of the potential field data. First, edges are naturally what human eyes focus on in the analysis of any kind of image. Second, edges in a gravity/magnetic map mimic the sketches which geophysicists usually draw on potential field maps to guide their interpretation. Edge enhancement filters are divided on two main groups, including high-pass filters and Laplacian filters. With high-pass filters, the high frequency features are passed and those of low frequency suppressed. Since the edge can be considered as a high frequency feature which occupies a portion of the image densely, this kind of filter is admissible. High-pass filters are capable to run either as directional or perpendicular. In the other hand, Laplacian filters enhance the edge in all directions. These filters are logical filters. The main idea of these filers is to replace the central pixel with the pixel having the maximum observed quantity. In fact, the application of these filters led to removing undesirable features.
In this paper, these filters were successfully applied on the synthetic magnetic data from prismatic model. Additionally, filters were applied on gravity and magnetic data from southwest England in which the output results relevant to causative granite bodies have a broad correlation with those from the geological map.