Object of shadow area suffers from low contrast and noise, causing difficulties for object detection and recognition. However, conventional image enhancement methods perform inefficiently in severer shadow conditions, where the contrast enhancement is not obvious, or the noise amplification is huge. In fact, Polarization information can be preserved even in very heavy shadows. Combining the polarization information with the intensity and chromaticity information of the original image can overcome the above problems. This paper proposes a method of object enhancement in shadow based on polarization imaging. Firstly, this paper puts forward degree of polarization (DOP) Invariance Prior and DOP-Chromatic Consistency Prior. Then, based on these two priors, a feedback image enhancement method based on polarization imaging is proposed. Experiments show that the proposed method is superior to the other conventional image enhancement methods.