In order to solve the problem of image defocus caused by the spacevariant errors after traditional phase correction methods, a highresolution algorithm for inverse synthetic aperture radar imaging based on complex Bayesian compressed sensing (CBCS) using Laplace priors is proposed. Assume that each pixel of the target image follows a Laplace prior to establish the prior models, then take the phase errors as the model errors. Target images and phase errors are solved using alternating iteration based on BCS theory. The algorithm is directly solved in the complex domain by Bayesian inference so as to avoid increasing the computational complexity of converting the complex to the real. In addition, the distributed computing method improves the computational efficiency compared with the traditional matrix vectorization method. Simulation experiments verify the effectiveness of the algorithm.