The problem of image reconstruction in additive noise and jitter environment is considered. Bispectrum-based Fourier magnitude and phase spectra recovery from the pre-distorted digital image rows is proposed and inves-tigated. Pre-distortion function of Gaussian shape is introduced in each image row to decrease spectral leak-age, obtain continuous phase bispectrum functions and fix the locations of the bispectrum reconstructed image rows. Computer simulations are provided to demon-strate the performance of the proposed reconstruction technique. Visual inspection of the reconstructed test images illustrates de-jittering and spectral leakage de-creasing in the presence of additive white Gaussian noise (AWGN).