AbstractSome of the serious factors that deteriorate the reconstructed image in MRI are the static field inhomogeneity and the motion of the object (patient) during data acquisition. To prevent the deterioration of the reconstructed image by such factors,a prioriknowledge about the static field inhomogeneity and the motion of the patient is required.This paper considers the ultrahigh‐speed imaging using the time‐varying gradient magnetic field and proposes a method which estimates the static field inhomogeneity and the patient motion from the measured FID signal and thea prioriknowledge about the region of existence for the patient. When the patient is constrained within a certain region in the field of view, a strong constraint called consistency condition essentially is imposed on the functional form of the measured FID signal.In the proposed method, a simple mathematical model is applied to the static field inhomogeneity and the patient motion and the model parameters are estimated so that the consistency condition is satisfied. By the proposed method of estimation, the artifact due to the static field inhomogeneity and the patient motion can be reduced without performing an additional measurem