Abstract: A new hybrid algorithm is presented to solve the ill-conditioned inverse problem of magnetic induction tomography (MIT) and improve the quality of reconstructed images. The hybrid algorithm first produces the preliminary image region using the solution well-posedness of the Tikhonov regularization algorithm, and then obtains the final reconstructed image using the edge-preserving and sharpening property of the variation regularization algorithm. Compared with the Tikhonov regularization algorithm and the variation regularization algorithm, the hybrid algorithm overcame the numerical instability of MIT image reconstruction, accelerated the convergence speed of image reconstruction, and improved the resolving power of target conductors and the quality of the reconstructed image. Simulation results verified the effectiveness of the proposed algorithm.