Adverse weather conditions such as haze can deteriorate the performance of autonomous driving and intelligent transport systems. As a potential remedy, we propose an enhanced multi-scale network, dubbed GridDehazeNet+, for single image dehazing. The proposed dehazing method does not rely on the Atmosphere Scattering Model (ASM), and an explanation as to why it is not necessarily performing the dimension reduction offered by this model is provided. GridDehazeNet+ consists of three modules: pre-processing, backbone, and post-processing. The trainable pre-processing module can generate learned inputs with better diversity and more pertinent features as compared to those derived inputs produced by hand-selected pre-processing methods. The backbone module implements multi-scale estimation with two major enhancements: 1) a novel grid structure that effectively alleviates the bottleneck issue via dense connections across different scales; 2) a spatial-channel attention block that can facilitate adaptive fusion by consolidating dehazing-relevant features. The post-processing module helps to reduce the artifacts in the final output. Due to domain shift, the model trained on synthetic data may not generalize well on real data. To address this issue, we shape the distribution of synthetic data to match that of real data, and use the resulting translated data to finetune our network. We also propose a novel intra-task knowledge transfer mechanism that can memorize and take advantage of synthetic domain knowledge to assist the learning process on the translated data. Experimental results demonstrate that the proposed method outperforms the state-of-the-art on several synthetic dehazing datasets, and achieves the superior performance on real-world hazy images after finetuning.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    GridDehazeNet+: An Enhanced Multi-Scale Network With Intra-Task Knowledge Transfer for Single Image Dehazing


    Contributors:
    Liu, Xiaohong (author) / Shi, Zhihao (author) / Wu, Zijun (author) / Chen, Jun (author) / Zhai, Guangtao (author)


    Publication date :

    2023-01-01


    Size :

    8285195 byte




    Type of media :

    Article (Journal)


    Type of material :

    Electronic Resource


    Language :

    English



    Single-Image Dehazing

    Tian, Jiandong | Springer Verlag | 2021


    Image Dehazing Network Based on Multi-scale Feature Extraction

    Feng, Ting / Zhang, Fuquan / Yu, Zhaochai et al. | TIBKAT | 2022


    Image Dehazing Network Based on Multi-scale Feature Extraction

    Feng, Ting / Zhang, Fuquan / Yu, Zhaochai et al. | Springer Verlag | 2021


    Single Image Dehazing of Railway Images via Multi-scale Residual Networks

    Cao, Zhi wei / Qin, Yong / Xie, Zheng yu | TIBKAT | 2020


    Single Image Dehazing of Railway Images via Multi-scale Residual Networks

    Cao, Zhi wei / Qin, Yong / Xie, Zheng yu | Springer Verlag | 2020