Three-dimensional image reconstruction in electrical tomography

Three-dimensional reconstruction is one of the most important tasks in computer vision. There are many efficient and robust three-dimensional reconstruction algorithms in medicine, chemistry, biology and etc. But so far there is no efficient method for the three-dimensional reconstruction in electrical tomography since there are uncertainties, incompleteness, and “soft field” effects in ET measured data. In this paper, a novel three-dimensional reconstruction method is proposed in terms of two-dimensional image sequences from a single sensor. The fuzzy clustering first is applied to segment any original two-dimensional image in the image sequence, the clustering prototypes in any pair of adjacent two-dimensional images are matched to realize the matching of all pixels in the two frames of images, and finally the three-dimensional reconstruction of the investigated objectives is realized in terms of the interpolation algorithms. This proposed method can greatly overcome the problems of low efficiency in the existing algorithms. A group of experiments are performed to validate the proposed method.