This paper presents a novel single-image haze removal algorithm that utilizes dark channel prior and adaptive transmission rate. The method aims to address the issues of low clarity, brightness, and color distortion in restored images obtained by traditional haze removal algorithms. The process involves partitioning the image into local regions to calculate the dark channel and minimum map, estimating atmospheric light, and determining the atmospheric dissipation value for each pixel. The median guide filtering algorithm is then used to distinguish between close-range and remote areas, correcting the atmospheric dissipation function to achieve accurate transmittance estimation. The restored image is generated based on the atmospheric scattering model. Experimental results show that the proposed algorithm reduces the fuzzy coefficient by 23.09%, increases structural similarity by 14.63%, and reduces tone reduction by 12.29%, significantly improving the quality of the restored images.This paper presents a novel single-image haze removal algorithm that utilizes dark channel prior and adaptive transmission rate. The method aims to address the issues of low clarity, brightness, and color distortion in restored images obtained by traditional haze removal algorithms. The process involves partitioning the image into local regions to calculate the dark channel and minimum map, estimating atmospheric light, and determining the atmospheric dissipation value for each pixel. The median guide filtering algorithm is then used to distinguish between close-range and remote areas, correcting the atmospheric dissipation function to achieve accurate transmittance estimation. The restored image is generated based on the atmospheric scattering model. Experimental results show that the proposed algorithm reduces the fuzzy coefficient by 23.09%, increases structural similarity by 14.63%, and reduces tone reduction by 12.29%, significantly improving the quality of the restored images.