A saliency guided remote sensing image dehazing network model

Shi, Zhongchao; Shao, Shuai; Zhou, Zhaorun

doi:10.1049/ipr2.12502

Cited by 4 publications

(2 citation statements)

References 46 publications

(214 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For instance, Huang et al [26] integrated the dense residual network with the squeeze and excitation block as the basic module to design a dual-step cascaded dense residual network, which can extract multi-scale local and global features of the RS hazy image more effectively. Similarly, Shi et al [27] also selected the dense residual network to capture the RS image's fine details and used the corresponding saliency map generated from global contrast to guide network training. However, the proposed dense networks by Huang et al [26] and Shi et al [27] are too bulky to be implemented in real-time applications.…”

Section: Learning-based Dehazing Methods With Training Datasetsmentioning

confidence: 99%

“…Similarly, Shi et al [27] also selected the dense residual network to capture the RS image's fine details and used the corresponding saliency map generated from global contrast to guide network training. However, the proposed dense networks by Huang et al [26] and Shi et al [27] are too bulky to be implemented in real-time applications. Based on the encoder and decoder architecture, Jiang et al [28] proposed a deep dehazing network for RS images with non-uniform haze and used the wavelet transform as an additional channel to retain image textures.…”

Section: Learning-based Dehazing Methods With Training Datasetsmentioning

confidence: 99%

See 1 more Smart Citation

Self-Supervised Remote Sensing Image Dehazing Network Based on Zero-Shot Learning

Wei,

Cao,

Yang

et al. 2023

Remote Sensing

View full text Add to dashboard Cite

Traditional dehazing approaches that rely on prior knowledge exhibit limited efficacy when confronted with the intricacies of real-world hazy environments. While learning-based dehazing techniques necessitate large-scale datasets for effective model training, the acquisition of these datasets is time-consuming and laborious, and the resulting models may encounter a domain shift when processing real-world hazy images. To overcome the limitations of prior-based and learning-based dehazing methods, we propose a self-supervised remote sensing (RS) image-dehazing network based on zero-shot learning, where the self-supervised process avoids dense dataset requirements and the learning-based structures refine the artifacts in extracted image priors caused by complex real-world environments. The proposed method has three stages. The first stage involves pre-processing the input hazy image by utilizing a prior-based dehazing module; in this study, we employed the widely recognized dark channel prior (DCP) to obtain atmospheric light, a transmission map, and the preliminary dehazed image. In the second stage, we devised two convolutional neural networks, known as RefineNets, dedicated to enhancing the transmission map and the initial dehazed image. In the final stage, we generated a hazy image using the atmospheric light, the refined transmission map, and the refined dehazed image by following the haze imaging model. The meticulously crafted loss function encourages cycle-consistency between the regenerated hazy image and the input hazy image, thereby facilitating a self-supervised dehazing model. During the inference phase, the model undergoes training in a zero-shot manner to yield the haze-free image. These thorough experiments validate the substantial improvement of our method over the prior-based dehazing module and the zero-shot training efficiency. Furthermore, assessments conducted on both uniform and non-uniform RS hazy images demonstrate the superiority of our proposed dehazing technique.

show abstract