Research On Electrical Impedance Imaging Method Based On Compressed Sensing And Structured Sparse Prior Information

Electrical impedance tomography(EIT)has been investigated extensively during the past decades as a noninvasive visualization and measurement technique.The EIT technique has the advantages of portable,safety,low cost,non-invasive and rapid measurement.EIT has wide application prospects in biomedical imaging and industrial process imaging.Due to the ill-posedness of EIT imaging problem and the diffusive nature of EIT,there are still challenges to develop an effective imaging reconstruction algorithm to improve the accuracy and stability of the calculation and reduce the consuming time of reconstruction.Algebraic Reconstruction Technique(ART)is an important iterative method used for EIT reconstruction.However,large iterative steps are required for ART method,which leads to low convergence speed.In order to overcome the disadvangtage of ART,the Symkaczmarz iterative method is innovalively applied for EIT imaging in this paper.The Symkaczmarz method can reduce the number of iterations by optimizing of the projection order.However,the Symkaczmarz method is susceptible to system noise and it is difficult to guarantee the quality of EIT image.In order to improve the quality of reconstructured image,a structured sparse representation model based on wavelet tree structure model is established.An EIT image reconstruction method based on structured sparse prior information is proposed in this paper.Feature information corresponding to the structure information of the image is extracted.As a result,the degree of sparseness and the reconstructued image quality can be further improved.However,the real-time performance of EIT imaging method is considered in this paper.In order to further reduce consuming time and guarantee the quality of image reconstruction without scarifying the quality of reconstructed image,an EIT method based on compressed sensing sampling is proposed in this paper.Random sampling method based on the theory of compressed sensing is used in the EIT system.The structured sparse representation method,which is suitable to reconstruct images through random sampling data,is used to solve the inverse problem.The proposed method can effectively shorten the reconstruction time without affecting the reconstruction quality.Finally,the EIT method based on compressed sensing sampling and structured sparse representation is vertified by the simulation and system experiments.The experimental results show that the proposed method can improve the imaging speed while enhancing the robustness to noise.As a result,this method is more suitable for the occasions that require both imaging accuracy and speed in biomedical imaging.