Advanced Techniques for Image Dehazing Using Optimized Gaussian Weighting Algorithms

(To address the halo effects and color distortion caused by the dark channel prior algorithm, this paper proposes an image dehazing method that combines Gaussian function weighting with transmission map optimization. First, an initial transmission map is obtained through superpixel segmentation and color attenuation prior, and a Gaussian weighting function is constructed by integrating pixel values to enhance the transmission map. Then, the transmission map is further refined using weighted total variation, and the dehazed image is obtained through a restoration formula. Additionally, the bright channel prior is introduced to further enhance the brightness of the image. Experimental results show that the proposed algorithm achieves a 49.7% improvement in signal-to-noise ratio and a 29.4% improvement in structural similarity, significantly outperforming traditional methods, particularly in terms of subjective visual quality and detail preservation.)