To solve the “soft-field” effect and ill-posed problem in electrical capacitance tomography, an image reconstruction algorithm based on weighted Singular Value Decomposition(SVD) truncation conjugate gradient is presented for electrical capacitance tomography.The working principle of electrical capacitance tomography is introduced and a measurement method for ECT system with 12 electrodes is proposed.On analysis of the sensitive matrix based on the SVD theory, a weighted conjugate gradient truncated SVD mathematical model is derived,and it is weighted normally by Tikhonov regularization method.Finally, the convergence of the algorithm is analyzed and applied to the image reconstruction for electrical capacitance tomography.Experimental results and simulation data indicate that for laminar flow, the average error can reach 27.54%, and the average number of iterative steps for all flow regimes can achieve 13 by the proposed algorithm.Compared with LBP,Landweber and CG algorithms, the algorithm has advantages in good image quality,high image speed and is a feasible and effective method for image reconstruction.