Title :
A threshold sensitivity method for image reconstruction of electromagnetic tomography
Author :
Li, Liu ; Shao, Fuqun ; Li, Liu
Author_Institution :
Coll. of Inf. Sci. & Eng., Norheastern Univ., Shenyang, China
Abstract :
Sensitivity matrix generation in electromagnetic tomography was an important step to reconstruct images. Threshold sensitivity matrix calculation method based on electrical field centre lines was applied to speed up generation of a sensitivity matrix, and improve reconstructed images quality. Firstly, the necessity and complexity about sensitivity matrix were introduced. According to the characteristic of sensitivity distribution in object field, electrical field centre lines and their relation with sensitivity distribution were led to the calculation. Secondly, the method of threshold sensitivity matrix generation was elaborated, and every threshold when each coil acted as the exciter and the others acted as receivers were selected. Formulas and threshold of calculating sensitivity matrix were given. Finally, comparing calculation speed of sensitivity matrix and the quality of reconstructed images with normal method, the conclusion was drawn: the reconstructed images (evaluating every image by the relative error and correlation coefficient) with the threshold sensitivity method have superior image quality than those with normal method, and time spent on calculating sensitivity is no more than 50 percent of normal method.
Keywords :
correlation methods; electric impedance imaging; image reconstruction; matrix algebra; tomography; correlation coefficient; electrical field centre lines; electromagnetic tomography; image quality; image reconstruction; relative error; sensitivity distribution; speed up generation; threshold sensitivity matrix calculation method; threshold sensitivity matrix generation; threshold sensitivity method; Coils; Electromagnetics; Image reconstruction; Magnetic resonance imaging; Receivers; Sensitivity; Tomography; electromagnetic tomography; image reconstruction; sensitivity;
Conference_Titel :
Image and Signal Processing (CISP), 2010 3rd International Congress on
Conference_Location :
Yantai
Print_ISBN :
978-1-4244-6513-2
DOI :
10.1109/CISP.2010.5647870