Stripe noise is one of the important factors affecting the quality of multispectral remote sensing image, seriously influencing the interpretation of remote sensing data and information extraction. A multispectral remote sensing image quality assessment method aimed at stripe noise based on phase congruency was proposed. This method, which extracts phase congruency point in the frequency domain as the image feature, is not affected by image brightness or contrast, and can objectively evaluate the stripe noise existing in the multispectral remote sensing image by considering the number, length, width, and the intensity of the stripe noise in the image. The experiments show that the proposed algorithm correlates well with subjective quality evaluations.