Remote sensing image dehazing is challenging because it is massively ill-posed and the haze is dependent on the unknown depth information. Haze removal based on dark channel prior is effective, and refining the transmission map with Gaussian filter will produce a god result. But need to improve the naturalness and sharpness and effectiveness of images and to remove fine haze . So Proposes new approach to dehaze the remote sensing image using guided filter. Guided filter is a type of edge-preserving smoothing operator, which filters the input image under the guidance of another image. Experiments and comparisons show that this method generates satisfactory dehazed results with improved naturalness and sharpness.