An ECT system Based on Improved RBF Network and Adaptive Wavelet Image Enhancement for Solid/gas Two­Phase Flow

Xia Chen, Hong­li Hu* and Juan Zhang

State Key Laboratory of Power Equipment and Electrical Insulation Xi’an Jiaotong University, Xi’an710049, China

ABSTRACT

Electrical capacitance tomography (ECT) is a non­invasive imaging technique that aims at visualizing the cross­sectional permittivity distribution of a dielectric object based on the measured capacitance in solid/gas two­phase flow measurement. To solve the nonlinear and ill­posed inverse problem for ECT, an improved radial basis function (RBF) neural network image reconstruction algorithm was put forward in the paper to fulfill the requirement of flow regime identification. And for better image quality, adaptive wavelet image enhancement technique was emphatically analyzed and studied, which belongs to a space­frequency analysis method and is suitable for image feature­enhanced. Through multi­level wavelet decomposition, edge points of the image can be determined based on the neighborhood property of each sub­band; noise distribution in the space­frequency domain can be estimated based on statistical characteristics; and then a self­adaptive edge enhancement gain can be constructed. Finally, the image is reconstructed with the adjusting wavelet coefficients. Experiment results demonstrated that adaptive wavelet image enhancement technique effectively implemented edge detection and image enhancement. The algorithm based on impoved RBF network and adaptive wavelet image enhancement greatly improved the quality of reconstructed image of solid/gas two­ phase flow (pulverized coal (PC)/air).

Keywords ECT; Image reconstruction; Radial basis function network; wavelet image enhancement