The invention belongs to the field of radar signal processing, and particularly relates to a sparse aperture ISAR self-focusing and lateral scaling method based on Bayesian learning. The method comprises the following steps: firstly performing sparse prior modeling on an ISAR image by using a Laplace layered model, and performing sparse reconstruction on the ISAR image by using a variational Bayesian method, wherein during the reconstruction of the ISAR image, a modified Newton iterative method is used to simultaneously estimate a phase error and a target rotational speed to achieve self-focusing and lateral scaling of the ISAR under the sparse aperture condition. The invention obtains the beneficial effects that: due to the sparse aperture ISAR self-focusing and lateral scaling method based on Bayesian learning, the self-focusing and lateral scaling of the ISAR under the sparse aperture condition can be realized, and the ISAR image with good focusing effect, high resolution and exactlateral scaling can still be obtained under the condition of sparse radar echo data aperture caused by the factors such as low SNR, strong interference and insufficient effective aperture, which has important engineering application value and can provide theoretical support for the design of a compressed sensing radar.