In hazy weather， the image quality is decreased owing to the scattering effect of atmospheric particles on light， which can affect target detection and key information extraction. In response to the shortcomings of current polarization defogging algorithms， this study proposes a defogging method based on polarization dark channel. First， polarization information is calculated through the polarized light intensity of four different angles， to obtain the polarized light intensity of any angle. Thereafter， based on the dark channel prior principle， the dark channel of the polarized image is defined. The atmospheric light intensity at infinity is obtained， and the atmospheric transmission rate is calculated. Finally， the atmospheric scattering model is constructed to obtain a defogging image. The experimental results demonstrate that the evaluation indicators of grayscale variance， average gradient， and image entropy of the defogged image processed by this method are significantly improved compared to those of the original image. The grayscale variance， average gradient， and image entropy are increased by 25%， 30%， and 5%， respectively. This method can effectively improve the quality and clarity of images in hazy weather and enrich image details.