基于CS-GPSR的電容層析成像圖像重建算法
發(fā)布時間：2019-01-05 02:26
【摘要】：提出將基于壓縮感知(CS)理論的稀疏梯度投影(GPSR)算法應(yīng)用于電容層析成像(ECT)圖像重建過程中。采用離散Fourier變換(DFT)基將原始圖像灰度信號進行稀疏化處理;將ECT靈敏度矩陣的各行按隨機順序進行排列,得到ECT系統(tǒng)觀測矩陣,同時將測量電容向量的各行按相同順序進行排列,得到觀測投影向量;使用GPSR算法進行圖像重建。仿真實驗結(jié)果表明:基于CS理論的GPSR(CS-GPSR)算法重建圖像質(zhì)量明顯優(yōu)于LBP算法和Landweber迭代算法。本文所述算法可實現(xiàn)較高精度的圖像重建,為ECT圖像重建的研究提供了一種新的手段。
[Abstract]:A sparse gradient projection (GPSR) algorithm based on compressed perceptual (CS) theory is proposed for the reconstruction of (ECT) images in electrical capacitance tomography. The original image gray signal is sparse based on discrete Fourier transform (DFT) basis. The lines of the ECT sensitivity matrix are arranged in random order to obtain the observation matrix of the ECT system. At the same time, the lines of the capacitance vector are arranged in the same order to obtain the observation projection vector. The GPSR algorithm is used to reconstruct the image. Simulation results show that the reconstructed image quality of GPSR (CS-GPSR) algorithm based on CS theory is obviously better than that of LBP algorithm and Landweber iterative algorithm. The algorithm presented in this paper can achieve high precision image reconstruction, which provides a new method for the research of ECT image reconstruction.
【作者單位】： 河北省發(fā)電過程仿真與優(yōu)化控制工程技術(shù)研究中心;華北電力大學(xué)自動化系;
【基金】：國家自然科學(xué)基金資助項目(51306058) 中央高校基本科研業(yè)務(wù)費專項項目(2014MS142)
【分類號】：TP391.41
[Abstract]:A sparse gradient projection (GPSR) algorithm based on compressed perceptual (CS) theory is proposed for the reconstruction of (ECT) images in electrical capacitance tomography. The original image gray signal is sparse based on discrete Fourier transform (DFT) basis. The lines of the ECT sensitivity matrix are arranged in random order to obtain the observation matrix of the ECT system. At the same time, the lines of the capacitance vector are arranged in the same order to obtain the observation projection vector. The GPSR algorithm is used to reconstruct the image. Simulation results show that the reconstructed image quality of GPSR (CS-GPSR) algorithm based on CS theory is obviously better than that of LBP algorithm and Landweber iterative algorithm. The algorithm presented in this paper can achieve high precision image reconstruction, which provides a new method for the research of ECT image reconstruction.
【作者單位】： 河北省發(fā)電過程仿真與優(yōu)化控制工程技術(shù)研究中心;華北電力大學(xué)自動化系;
【基金】：國家自然科學(xué)基金資助項目(51306058) 中央高校基本科研業(yè)務(wù)費專項項目(2014MS142)
【分類號】：TP391.41
【參考文獻】
相關(guān)期刊論文 前5條
1 王丕濤;王化祥;孫r囋，
本文編號：2401137
本文鏈接：http://sikaile.net/kejilunwen/ruanjiangongchenglunwen/2401137.html
最近更新
教材專著
