Research Progress in Image Dehazing Methods

Abstract

Haze in the environment, caused by airborne particles, reduces image clarity and poses challenges for subsequent in-depth analysis and processing of images. Therefore, it is imperative to evaluate the advantages and disadvantages of various image dehazing methods, identify the challenges they face, and explore prospects. This paper focuses on image dehazing techniques, discussing the latest research advancements with an emphasis on comparing deep learning-based and traditional methods. It delves into the principles and application domains of different dehazing approaches, including image enhancement methods, image restoration methods, and deep learning-related techniques. Research indicates that various image dehazing methods face challenges such as detail processing, colour distortion, and computational complexity. Future developments should integrate the strengths of traditional and deep learning approaches, and improve network architectures, evaluation systems, and generalization capabilities to further advance image dehazing technology in practical applications.