Single-image dehazing via depth-guided deep retinex decomposition

Chen, Hui; Zhang, Tian-Lun; Ma, Long; Li, Nannan

doi:10.1007/s00371-022-02659-z

Cited by 8 publications

(6 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Liu et al 33 proposed a novel dehazing algorithm based on Retinex theory and homomorphic filtering. Chen et al 34 proposed a depth-guided Retinex decomposition network that presents a reverse Retinex-based approach to single-image dehazing. Li et al 25 designed a deep Retinex dehazing network that consists of a multiscale residual dense network for estimating the residual illumination map and a U-Net structure with channel and spatial attention mechanisms for image dehazing.…”

Section: Retinex-based Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Retinex decomposition and fusion dehazing network

et al. 2023

View full text Add to dashboard Cite

Image dehazing has been widely used in vision-based fields, such as detection, segmentation, traffic monitoring, and automated vehicle system. However, most of the existing endto-end dehazing networks are fully data driven without physical constraints or prior information guidance, leading to difficulties in exploring latent structures and statistical characteristics of hazy images. We propose a novel Retinex decomposition-fusion dehazing network consisting of a dual-branch decomposition module and a fusion optimization module. Different from existing solutions, we decompose the clear images in the commonly used RESIDE dataset based on Retinex theory to construct the clear illumination map and reflection map datasets to drive the network training, equivalently imposing reasonable constraints on the network and achieving impressive dehazing performances. The dual-branch decomposition module is developed to estimate the illumination map and the reflection map, respectively. The illumination map mainly contains the global features of the image, whereas the reflection map reflects the inherent color properties of the image and contains rich details, with which we explore the latent structures and statistical characteristics of hazy images. In addition, the dual-branch structure avoids the error accumulation and information cancellation existing in current methods. Subsequently, the estimated illumination map and reflection map are fused and refined via the fusion optimization module to access the dehazed image. Experiments show that the proposed network has better generalization and visual effects than existing fully data-driven methods and can be applied successfully to real-world scenarios.

show abstract

Section: Retinex-based Methodsmentioning

confidence: 99%

“…proposed a novel dehazing algorithm based on Retinex theory and homomorphic filtering. Chen et al 34 . proposed a depth-guided Retinex decomposition network that presents a reverse Retinex-based approach to single-image dehazing.…”

Section: Related Workmentioning

confidence: 99%

Retinex decomposition and fusion dehazing network

et al. 2023

View full text Add to dashboard Cite

show abstract

“…One of the key advantages of deep learning-based dehazing is its ability to generalize well to real-world scenarios, including those with non-uniform haze distributions, varying scene complexities, and diverse lighting conditions [6]. These methods can effectively restore details in challenging scenarios, such as underwater and nighttime dehazing, where traditional approaches often struggle to provide satisfactory results [7].…”

Section: A Backgroundmentioning

confidence: 99%

“…Moreover, researchers have explored innovative approaches that integrate domain-specific information, such as polarization information in underwater dehazing, to further enhance the dehazing process [7]. These specialized techniques leverage the physical properties of light polarization to better estimate and remove the scattered light from underwater scenes, resulting in improved visibility and image quality.…”

Section: A Backgroundmentioning

confidence: 99%

DSSCNet: Deep Custom Spatial and Spectral Consistency Layer-Based Dehazing Network

Kaur,

Singh,

Kumar

et al. 2024

IEEE Access

View full text Add to dashboard Cite

Poor weather conditions, such as haze, fog, and smog, present significant challenges in capturing clear and visually appealing images. To address this issue, we propose a Deep Custom Spatial and Spectral Consistency Layer-based Dehazing Network (DSSCNet) that effectively removes haze from images while preserving important spatial and spectral details. The network architecture includes a custom Haze Removal Layer (HRL), convolutional layers with ReLU activation, pooling layers, skip connections, and a custom Spatial and Spectral Consistency Layer (cSSCL). HRL estimates atmospheric light and transmission maps to generate an intermediate haze-free image. The proposed loss function combines Mean Squared Error (MSE) loss with a Consistency Loss (CL) to encourage content preservation during dehazing. The network is trained using the Adam optimizer to optimize the loss function, resulting in a powerful dehazing network capable of producing visually realistic and haze-free images. Extensive experimental results demonstrate the effectiveness and superiority of DSSCNet when compared to existing dehazing techniques. The outcomes highlight that DSSCNet outperforms competitive models in terms of various performance metrics, including Contrast gain (c g ), new visible edges (e), new edge gradients (r), Peak Signal-to-Noise Ratio (PSNR), and Structural Similarity Index (SSIM). These improvements amount to an average enhancement of around 1.27%, 1.12%, 1.18%, 1.21%, and 1.24%, respectively.

show abstract

“…The traditional hazy image enhancement method applies the image enhancement method to clear the image, directly improves the contrast of the image, and highlights the details, but it will cause the loss of part of the information, such as homogeneous filtering [1], histogram equaliza- tion [2] and Retinex [3], [4], etc. But this method does not achieve physical dehazing.…”

Section: Introductionmentioning

confidence: 99%

Single Image Dehazing Based on Sky Area Segmentation and Image Fusion

Chen

et al. 2023

IEICE Trans. Inf. & Syst.

View full text Add to dashboard Cite

Since the dark channel prior (DCP)-based dehazing method is ineffective in the sky area and will cause the problem of too dark and color distortion of the image, we propose a novel dehazing method based on sky area segmentation and image fusion. We first segment the image according to the characteristics of the sky area and non-sky area of the image, then estimate the atmospheric light and transmission map according to the DCP and correct them, and then fuse the original image after the contrast adaptive histogram equalization to improve the details information of the image. Experiments illustrate that our method performs well in dehazing and can reduce image distortion.

show abstract

Single-image dehazing via depth-guided deep retinex decomposition

Cited by 8 publications

References 35 publications

Retinex decomposition and fusion dehazing network

Retinex decomposition and fusion dehazing network

DSSCNet: Deep Custom Spatial and Spectral Consistency Layer-Based Dehazing Network

Single Image Dehazing Based on Sky Area Segmentation and Image Fusion

Contact Info

Product

Resources

About