Images have a wide significance these days for detailed observation in different fields like remote sensing, and navigation. When images got captured by the image acquisition equipment, they got affected by environmental effects. These environmental effects are haze, smog, fog which cannot be ignored while acquisition. Such type of effects should be reduced so that images can be accurately observed. These effects can be considered as noise in our work. These effects are difficult to reduce efficiently while acquisition. Therefore, there arises a need for an algorithm to process the noisy image after capturing. In our work, Dark Channel Prior (DCP) technique is used for the de-noising purpose and our focus is on the refining of the estimated transmission map so that a noise-less image can be recovered. The efficient noise reduction is determined by the Peak Signal to Noise Ratio (PSNR) value. PSNR value is the evaluation parameter which is used to determine that the proposed work has improved the performance compared to recently introduced approaches. An analysis of our proposed method and other research work is shown.