Image/Video Dehazing Techniques & Challenges

Image and video dehazing are essential techniques in computer vision to restore visual clarity and contrast in outdoor scenes affected by fog, haze, or atmospheric particles. This paper comprehensively reviews model-based, filtering-based, and learning-driven dehazing techniques, highlighting their effectiveness and limitations. Particular attention is given to hybrid approaches that integrate physical priors with data-driven models to improve accuracy, edge preservation, and computational efficiency. The objective of this study is to evaluate these techniques across complexity, performance, and adaptability dimensions. Quantitative results using PSNR, SSIM, and processing time are discussed alongside practical implementation challenges. The paper concludes with future directions for real-time, scalable, and adaptive dehazing suitable for dynamic environments and constrained devices.