Image dehazing based on polarization information and deep prior learning

Bi, Pengshuai; Wang, Dongliang; Chen, Wei; Yang, Lan; Liang, Jian; Li, Guangchao; Zhang, Fengjiao; Wang, Zhe; Zhang, Xuepeng

doi:10.1016/j.ijleo.2022.169746

Cited by 12 publications

(3 citation statements)

References 40 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In 2022, China University of Mining and Technology proposed a new image dehazing algorithm based on polarization information and deep prior learning [ 13 ]. This method achieved a 31% improvement in contrast for heavily hazy images, providing a new direction for image dehazing processing.…”

Section: Introductionmentioning

confidence: 99%

Analysis of multispectral polarization imaging image information based on micro-polarizer array

Fu,

Ma,

Liu

et al. 2024

PLoS ONE

View full text Add to dashboard Cite

As a new detection technology, polarization imaging is of great significance in the field of target detection. At present, polarization imaging technology usually adopts visible light polarization imaging. The technique is difficult to image the target in complex background due to its narrow working spectrum and short detection distance. Therefore, based on the principle of full Stokes micro-polarizer array, this paper proposes a multi-spectral polarization imaging scheme and designs a multi-spectral polarization imaging detection system penetrating haze. Conducting indoor and outdoor polarized imaging experiments. Finally, image quality was assessed using metrics such as information entropy (EN), average gradient (AG), and standard deviation (STD). The results show that compared with traditional strength detection, the imaging system has significantly improved detection distance and imaging quality in smoky environments. The imaging system can effectively enhance the contours and details of the target object and improve detection and recognition capabilities.

show abstract

Section: Introductionmentioning

confidence: 99%

Analysis of multispectral polarization imaging image information based on micro-polarizer array

Fu,

Ma,

Liu

et al. 2024

PLoS ONE

View full text Add to dashboard Cite

show abstract

“…These algorithms seek to maximize color and detail restoration while enhancing overall clarity. Currently, such algorithms fall into two broad categories: prior-knowledgebased algorithms [7,8] and deep-learning-based algorithms [9][10][11][12].…”

Section: Introductionmentioning

confidence: 99%

Discerning Reality through Haze: An Image Dehazing Network Based on Multi-Feature Fusion

Wang,

Jiang

et al. 2024

Applied Sciences

View full text Add to dashboard Cite

Numerous single-image dehazing algorithms have been developed, employing a spectrum of techniques ranging from intricate physical computations to state-of-the-art deep-learning methodologies. However, conventional deep-learning approaches, particularly those based on standard convolutional neural networks (CNNs), often result in the persistence of residual fog patches when applied to images featuring high fog concentration or heterogeneous fog distribution. In response to this challenge, we propose an innovative solution known as the multi-feature fusion image dehazing network (MFID-Net). This approach employs an end-to-end methodology to directly capture the mapping relationship between hazy and fog-free images. Central to our approach is the introduction of a novel multi-feature fusion (MF) module, strategically designed to address channel and pixel characteristics in regions with uneven or high fog concentrations. Notably, this module achieves effective haze reduction while minimizing computational resources, thereby mitigating the issue of residual fog patches. Experimental results underscore the superior performance of our algorithm compared to similar dehazing methods, as evidenced by higher scores in structural similarity (SSIM), peak signal-to-noise ratio (PSNR), and computational velocity. Moreover, MFID-Net exhibits significant advancements in restoring details within expansive monochromatic areas, such as skies and white walls.

show abstract

“…The light intensity and atmospheric polarization at infinity are calculated to jointly recover the defogging image. Due to the use of light intensity information for defogging, this method is very light dependent and cannot be applied when the image brightness is dark.Liang et al [2] proposed a new method for estimating atmospheric light intensity using polarization angle, which can suppress the effect of transmitted light on the image to the maximum extent.Gao et al [3] used a polarization filtering method to estimate the polarization distribution of atmospheric scattered light on foggy days, and used an adaptive adaptive bright channel method to calculate the intensity distribution of atmospheric light at infinity to reconstruct the defogged image, and finally the texture information of polarization degree is used to enhance the reconstructed defogged image.Bi et al [4] proposed a polarization-based unsupervised defogging network PUDN, which uses three joint sub-networks to divide the given haze image into three parts, and then reconstructs the input blurred image by an atmospheric scattering model. A priori knowledge is also added to the model to guide and constrain the learning process of the model.Zhou et al [5] proposed a generalized physical formation model of the fuzzy image without assuming that the transmitted light is not significantly polarized, while considering spatially varying real-world scattering, based on the physical model to estimate the atmospheric light at infinity and the polarization of transmitted and atmospheric light through the network.Most of the above methods rely on a priori knowledge or ideal atmospheric transport models, and use the difference in polarization characteristics of atmospheric light and the target to process the image effectively.…”

Section: Introductionmentioning

confidence: 99%

Polarization image defogging based on detail recovery generative adversarial network

Yang,

2023

Third International Conference on Signal Image Processing and Communication (ICSIPC 2023)

View full text Add to dashboard Cite

Most of the existing polarization image based defogging methods use a priori and assumptions to recover images, and although these methods have made great progress, the a priori or assumptions are not always reliable in practical scenarios, which limits the performance of defogging methods. In this paper, we design a two branch network to learn image features; the defogging network uses the fusion module to adaptively assign weights to different path features to ensure the defogging effect while focusing on the haze region; the detail recovery network uses convolutional layers and smoothly expanding convolution to expand the perceptual field and fully obtain local and global feature information; finally, the features obtained from the detail recovery network and the defogging network are fused to improve the defogging Finally, the features obtained from detail recovery network and defogging network are fused to improve the defogging effect. The experimental results show that this method can reconstruct clear images in foggy environment with 2db improvement in PSNR and 0.016 improvement in SSIM, and the quality of reconstructed images is better than existing algorithms.

show abstract

Image dehazing based on polarization information and deep prior learning

Cited by 12 publications

References 40 publications

Analysis of multispectral polarization imaging image information based on micro-polarizer array

Analysis of multispectral polarization imaging image information based on micro-polarizer array

Discerning Reality through Haze: An Image Dehazing Network Based on Multi-Feature Fusion

Polarization image defogging based on detail recovery generative adversarial network

Contact Info

Product

Resources

About