Segmentation is an important step for the diagnosis of multiple sclerosis. In this paper, a method for segmentation of multiple sclerosis lesions from Magnetic Resonance (MR) brain image is proposed. The proposed method combines the strengths of two existing techniques: fuzzy connectedness and artificial neural networks. From the input MR brain image, the fuzzy connectedness algorithm is used to extract segments which are parts of Cerebrospinal Fluid (CSF), White Matter (WM) or Gray Matter (GM). Segments of the MRI image which are not extracted as part of CSF, WM or GM are processed morphologically, and features are computed for each of them. Then these computed features are fed to a trained artificial neural network, which decides whether a segment is a part of a lesion or not. The results of our method show 90% correlation with the expert's manual work.
Manual fault mapping in 3D seismic interpretation is labor-intensive and time-consuming. Complex fault geometries and the distortion of the seismic signal close to faults complicate full automation of the fault-mapping process. We present a semiautomatic fault-tracking method for 3D seismic data that consists of fault highlighting followed by model-based fault tracking. Fault highlighting uses log-Gabor filters for emphasizing oriented amplitude discontinuities at faults in the presence of noise. Subsequent fault tracking fits an active contour to the highlighted fault voxels. The active contour searches for a connected, smooth curve which fits the data and disambiguates misleading or missing information. The fault tracker requires the interpreter to place the active contour close to a fault on one initial seismic inline (2D pick). The active contour deforms to the closest amplitude dis-continuity highlighted. This tracking result is then projected forward to the next inline, providing an initial fault pick on this section that is again optimized by the active contour. Tracking results on a series of successive seismic sections, finally, constitute a 3D fault surface. User interaction is solely required for an approximate fault pick on the first inline, and in cases where the fault line is lost due to insufficient signal. Use of the autotracker prototype provides a fast solution for the mapping of complete 3D fault surfaces of constant dip, and for the automated tracking of fault portions within distinct dip domains, if fault surfaces are curved (i.e., listric). The method was applied to a series of high-quality reflectivity sections extracted from a 3D seismic volume from shallow-offshore Nigeria, with the tracking results (generated within seconds) comparing well with manually interpreted fault surfaces.
Abstract-While 3D seismic data become widespread and the data-sets get larger, the demand for automation to speed up the seismic interpretation process is increasing as well. However, the development of intelligent tools which can do more to assist interpreters has been difficult due to low information content in seismic data. In this paper, we present an image processing method in which a-priori geological knowledge is incorporated to correlate horizons across faults. Our new method exploits anisotropic spatial correlation of horizons for robustness and aims at developing an interpreter friendly interactive environment. The results of this method are compared with previously proposed methods.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.