An ECT Image Reconstruction Algorithm Based On Dual Sensitivity Field And Artificial Bee Colony

The electrical capacitance tomography technology is a new kind of processTomography Technology that based on the electrical properties of the object. Itobtains the detection capacitance by using the capacitance sensor that arrangedaround the pipe evenly, and reconstructs the distribution of the medium in the pipe byusing a certain image reconstruction technology. This kind of technology has widelyused in the field of industrial two phase flow inspection. Due to its superior safety,economy and extensive adaptability, the electrical capacitance tomographytechnology has got rapid development in visualization monitoring of two phase flow.Now the ECT technology has become a hot topic of topographic imaging technique.Based on a vast amount of domestic and international literature, and based onthe research of capacitance tomography technology, this paper mainly does thefollowing works:1. For the problems of the traditional artificial colony algorithm’s lowconvergence speed and easy to fall into local optimal result, an improved artificialcolony algorithm has been proposed. According to the basic algorithm of artificialcolony algorithm, it only carries on the comprehensive evaluation of the result thatthe bee colony has found. At this time, there may be a better variable in a certainsearch results while other variables are better in other search results. If this conditionappears, the basic artificial colony algorithm will be premature convergence. In thelate operation process it also will cause the algorithm easy to fall into local minimum,and the accuracy of the will never high. Aiming at this problem, the algorithm wasimproved as follows: recombine the result after the iteration to find the optimalresults, then according to the basic algorithm to deal with the results. The simulationexperiments have shown the effectiveness of this improved algorithm.2. The basic principle of the capacitance tomography technology has beenanalyzed. The mathematical model and the working principle of the capacitancetomography technology have been summarized. Because of the low accuracy of thetraditional ECT reconstruction algorithm, this paper has put forward a new kind of algorithm that based on a new type of sensitivity field (that is the dual sensitivityfield). This sensitivity field is complementary with the traditional sensitivity field, itwas obtained by hollowing out the high dielectric object in each pixels. In thisalgorithm, we calculate the dual sensitivity field and traditional sensitivity field atfirst, and then using the dual sensitivity and the traditional sensitivity to reconstructthe images respectively. At last, fuse the two kinds of image to get the final image.The improved landweber algorithm that based on the dual sensitivity considered theinteraction between the adjacent pixels during the procession of obtaining thesensitivity field. By using the complementary relationship between the traditionalsensitivity field and the dual sensitivity field, a better reconstruction image will beobtained. The simulate results have proved that this method can overcome the “softfield” effect to some extent, and achieve the goal of improve the quality ofreconstruction image.3. Although the image reconstruction algorithm that based on the dual sensitivitycan reduce the error caused by the “soft field” effect to some extent, it still has a largeerror compared with the original image. Especially, there’s a transitional zone on theborder of different media, this area has large error. This paper has introduced theimproved artificial colony algorithm into the calculation of the ECT imagereconstruction by using the artificial bee colony algorithm to post-process the graymatrix of ECT reconstruction image. The specific approach is as follows: using anappropriate fitness function, the artificial bee colony algorithm can determine athreshold during the iteration, and by using the threshold to improve the image thatobtained by the improved image reconstruction algorithm that based on the dualsensitivity field, so as to improve the quality of the final reconstructed image. Finallysimulate experiment results also have shown that the method is feasible and effective.