To carry out a computationally fast and accurate atmospheric correction based on polarized image, an algorithm of atmospheric correction was proposed by combining the intensity and polarization degree information of polarized light. Firstly, based on the vectorized linearized discrete ordinate radiative transfer(VLIDORT) model, a look-up tables (LUTs) which is composed of observation geometry, atmospheric parameters, aerosol model, and surface polarization characteristic were set up. Then the atmospheric correction was carried out through the look-up tables by interpolation method. The experimental results show that the algorithm has a good effect on foggydays, which is a good way to remove the negative impact of fog. After the correction, the image information entropy increased by 0.44 and the contrast increased by 41.84%. And then it is proved that the algorithm can restore the target’s intensity and degree of polarization. It also has a good performance on enhancing the image amount of information and contrast.