Wavelet-based physically guided normalization network for real-time traffic dehazing

Shengdong Zhang

doi:10.1016/j.patcog.2025.112451

Wavelet-based physically guided normalization network for real-time traffic dehazing

2025-09-27

0

PRE

AI

DOI：10.1016/j.patcog.2025.112451

Original

Original request for help

Save

Abstract

En 中文

• We have designed an efficient Dehazing model. Traditional Dehazing models are complex and slow. We improve the speed and quality of Dehazing via frequency domain learning. • We propose a Physically Guided Normalization to restore the feature similarity as in the clean image domain. • We propose a wavelet-based module to enhance the representational capability of the Dehazing model in the feature domain.

Wavelet-based physically guided normalization network for real-time traffic dehazing

Abstract

Journal

Organization