| 摘要 |
<P>PROBLEM TO BE SOLVED: To provide less computationally intensive and more accurate method for generating accurate high-resolution MCG images from a low-resolution 2D map. <P>SOLUTION: An MCG apparatus includes magnetic sensors with usually having few planar arrays, each of the sensors providing very low-resolution 2D MCG maps. Algorithm based on model learning is used to create the high-resolution MCG images from the measurement of rarefaction. A model is constructed using multiple resolution MCG images that are randomly created based on the Biot-Savart law. The high-resolution MCG images are created by adapting the model to the measurement of rare faction. Subsequently, a 2D current position is positionally measured by localizing a peak in a tangent component of the high-resolution MCG images. Finally, the 2D current position measurement is elaborated by a non-linear optimization algorithm, which simultaneously restores the depth, size and orientation of the current from the sensor. <P>COPYRIGHT: (C)2012,JPO&INPIT |
