This paper introduces a novel Adaptive Dark Channel Prior (ADCP) enhancement algorithm designed to improve image clarity in autonomous vehicles by effectively addressing haze and halation. The algorithm utilizes the Dark Channel Prior to estimate the initial transmittance map, which is then adaptively refined based on the intensity of halation. To further enhance image quality while preserving resolution and reducing noise, the ADCP model incorporates an edge-preserving filter. Additionally, the initial transmittance is determined using Otsu’s thresholding algorithm, which optimally segments halation and non-halation regions by maximizing between-class variance. Simulation results demonstrate that the ADCP technique, combined with the edge-preserving filter and Otsu’s thresholding, significantly improves visual quality, efficiently mitigates halation, and reveals details in dark regions of the image. This approach outperforms conventional methods, making it particularly suitable for applications in autonomous vehicles.