Underwater Polarization Imaging Research Articles

Research of the underwater polarization imaging process based on the Oren–Nayar polarization bidirectional reflection distribution function

Research of the underwater polarization imaging process based on the Oren–Nayar polarization bidirectional reflection distribution function

A robust underwater polarization image recovery based on Angle of Polarization with low-rank and sparse decomposition

Underwater imaging method based on the Angle of Polarization (AoP) is extremely popular due to its ability to remove the backscattered light effectively. However, existing methods overlook a potentially important phenomenon: the AoP contains significant noise in the real world, which leads to inaccurate estimations of key parameters. In this study, we propose a novel robust underwater polarization image recovery method based on AoP with low-rank and sparse decomposition. Based on the analysis of polarization, the AoP of backscattering exhibits low-rank characteristics, whereas the AoP of noise and targets demonstrates sparsity. Considering different sparsity of AoP, we combine Robust Principal Component Analysis and Alternating Direction Multiplier Method (RPCA-ADMM) with an adaptive threshold shrinkage method for low-rank and sparse decomposition, the AoP of the backscattering is obtained. Moreover, according to the low-rank component of the AoP, the Degree of Linear Polarization (DoLP) of backscattering can be effectively and automatically estimated without considering whether the background region exists, relying on a frequency prior strategy. Finally, the degraded image is recovered utilizing underwater polarization imaging model. Extensive experimental results demonstrate that the recovered images from the proposed method have abundant detail, high contrast, and minimal color distortion, which is competitive with most state-of-the-art methods.

Transformer-based improved U-net for high-performance underwater polarization imaging

Polarization imaging plays an important role in underwater imaging technology, as it allows for the suppression of backscattering light and thus enhances the quality of underwater images. Furthermore, with advancements in computer hardware technology, deep learning (DL) has experienced rapid development in recent years. However, existing methods based on convolutional neural networks (CNN) have the limitations of local feature extraction and fail to full use of the salient target’s polarization features. In this paper, we propose a transformer-based improved U-Net (TIU-Net) to further improve the performance of underwater polarization imaging. The proposed TIU-Net leverages CNN and transformer module to extract short-range and long-range features respectively for underwater target reconstruction. Meanwhile, by entering multi-dimensional information, each of which has a different emphasis on the expression of target features, to enhance the performance and stability of the reconstruction model. Experimental results on our established polarization underwater dataset show the superiority of our proposed method for underwater imaging, achieving the efficient imaging and high-quality generalization imaging in highly turbid underwater environments.

Research on improving the clarity of underwater polarization imaging image restoration

Research on improving the clarity of underwater polarization imaging image restoration

Underwater motion scene image restoration based on an improved U-Net network.

Active underwater polarization imaging is a common underwater imaging method, which uses the polarization difference between the reflected light and the scattered light in the underwater scene to suppress the scattered light, so as to improve the imaging quality of the underwater scene. However, the implementation often requires the acquisition of multiple polarization images, which is not suitable for the restoration of images of underwater motion scenes. To address the problem, a U-AD-Net deep learning network model based on a single polarized image is proposed, taking the polarization information of the single polarized image as the feature input, based on the classic U-Net network model, and introducing Dense-Net and spatial attention module. The learning ability and generalization ability of the proposed model for deep features are enhanced, and the polarization information that is most helpful to the image restoration is extracted, so as to restore the scene image more comprehensively. IE, AG, UCIQE, and SSIM are selected as evaluation metrics to assess the quality of the restored images. Experimental results show that the images restored through this proposed method contain richer detail information, having an obvious advantage to the existing network models. Since only a single polarized image is needed for restoration, this method has dynamic adaptability to underwater moving scene restoration.

Underwater polarization de-scattering imaging based on orthogonal polarization decomposition with low-pass filtering

Previous polarization imaging methods tend to assume that the degree of polarization of the target reflected light is a constant, while the targets with nonuniform polarization characteristics are less considered. The required background area or prior knowledge hinders the practical application of underwater polarization imaging technology. In this paper, we analyzed the basic physical model of underwater imaging, and proposed a new underwater polarization de-scattering method by considering orthogonal polarization decomposition of both the target reflected and backscattered light. The polarized-difference intensity of the backscattered light can be directly calculated from the polarization orientation angle of the two horizontal decomposition components. Meanwhile, a low-pass filter was used to suppress noise and globally evaluate the degree of polarization of the backscattered light. The proposed method can automatically distinguish the underwater targets polarization characteristics, and no background area or any prior information is required. The laboratory simulation experimental results demonstrate that this method can effectively improve the quality of underwater images under different conditions.

Real-time underwater polarization imaging without relying on background

Polarization imaging has excellent potential for use in underwater application scenarios due to efficient elimination of backscattered light. In most studies, two images that contain the maximum and minimum visible backscatter are obtained by finding the brightest and darkest moments during the rotation of the polarizer. In addition, the background region must be included in the image. To prevent these limitations, this study directly generates the optimal image pair via the Stokes vector. These images are then transformed into the frequency domain, and the global distribution of the degree of polarization of backscattered light is obtained using low-pass filtering. Recovery of the clear scene requires neither relying on background nor a search for the brightest and darkest images, significantly enhancing the real-time characteristics of the imaging process. Because of this, imaging of moving targets comes true. Additionally, experimental results at continuously varying distances verify the effectiveness of the proposed method for targets with different polarization properties. Especially, the imaging distance can be increased by 30% compared to the conventional active polarization imaging model at the same level of contrast.

Underwater polarization image restoration based on a partition method

The quality of underwater polarization imaging is mainly affected by the polarization properties of the target and impurity particles. Traditional methods often assume uniform polarization characteristics of the target, which make it difficult to address the restoration issues of complex targets. In response, a partition-based method for recovering underwater polarization images is proposed. The method involves pre-processing the image using the Gaussian curvature filtering algorithm, partitioning the image based on polarization information. In addition, a joint image evaluation method is used to achieve restoration of complex polarized characteristic targets. The method estimates the value of the reflected light polarization of one partition to estimate the value of the next partition and links the polarization values of each partition. Our approach achieves clear restoration results for multiple targets or complex structural objects underwater. Achieving significant improvement in image quality in multi-target underwater scenes, our method is highly effective for complex underwater environments. Experimental results show that our method, when compared with three other newer methods on multiple images of different targets and under varying scattering conditions, achieves an average increase of 617% in the standard deviation image evaluation index and a 61% optimization in the natural image quality evaluator index. Furthermore, our method is robust for different degrees of water turbidity.

An underwater image restoration based on global polarization effects of underwater scene

Currently, most underwater polarization imaging methods assume that the backscattering is polarized and the reflected light of the target is unpolarized. In reality, this assumption is unreliable since both backscattering and reflected light contribute to polarization. Therefore, we propose an underwater image restoration method that combines the polarization effects of reflected light and backscattering. Firstly, Stokes vectors are used to automatically obtain polarization difference images. Based on the enhancement measure evaluation (EME) constraint, the initial estimate of backscattering is obtained. Then, an adaptive local optimization (ALO) method is designed to estimate global backscattering light. A variable factor is introduced for local optimization to obtain global backscattering. Finally, in order to optimize the system, particle swarm optimization (PSO) is introduced, which improves the processing speed of the system. The experimental results show that it is necessary to consider the polarization of the target during underwater polarization descattering, which not only has a good ability to eliminate scattering but also has an excellent effect on restoring target details.

Research on polarization characteristics of background light by modified polarization difference imaging method

<sec>The random scattering event of light by water medium is the primary reason for the degradation in underwater imaging. Underwater polarization imaging technology can enhance the signal-to-noise ratio of imaging effectively by utilizing the polarization information difference between background scattered light and target light. However, as scattering events increase in the water body, it is difficult to maintain the polarization characteristics of light, which reduces the effect of removing scattering based on polarization characteristics. In addition, the polarization rule of background scattered light in water is unclear, and there is a lack of quantitative description of the polarization characteristics of scattered light. Therefore, the study of polarization transmission characteristics of underwater scattered light is of great significance in reducing the scattering light of underwater polarization imaging.</sec><sec>In order to clarify the polarization characteristics of underwater background scattered light, especially the polarization angle information, this paper proposes a method for ascertaining polarization angle of background light based on modified polarization difference imaging method. In this method, the coupling relationship between optimal weight coefficient and enhancement measure evaluation (EME) value of the Stokes vector difference result is analyzed, and the background light polarization angle is calculated based on the optimal weight coefficient. Combined with the experimental results, the EME distribution trend of the optimal weight coefficient and the modified polarization difference imaging method results in different turbidity water bodies are determined, the scattering suppression limit is explored, and the trend of background scattered light polarization direction with turbidity of water is analyzed. The results show that the proposed method can obtain the exact polarization angle of background scattered light in different water environments, revealing a trend that the polarization direction of background scattered light becomes orthogonal to the incident light direction as the turbidity of the water increases. This research provides a methodological basis for determining the polarization direction of the background scattered light in underwater imaging.</sec>

Underwater polarization imaging based on two-layer multi-index optimization

<sec>Underwater imaging is of great significance in exploring seabed resource , monitoring marine environment, implementing underwater rescue and military reconnaissance, etc. by providing clear vison. Among various underwater imaging techniques, the polarization imaging is considered to be an effective way to improve the quality of underwater imaging. It can realize underwater image restoration by using the difference in polarization characteristic between the target light and backscattered light. A classical underwater active polarization imaging method was presented by Treibitz [Treibitz T, Schechner Y Y <ext-link ext-link-type="uri" xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="http://doi.org/10.1109/TPAMI.2008.85">2009 <i>IEEE Trans. Pattern Anal. Mach. Intell.</i> <b>31</b> 385</ext-link>], in which the degrees of linear polarization (DoLPs) of target light and backscattered light are used to recover clear image. A variety of improved methods have been derived from this, but most of them require background areas and human-computer interaction. Then, a new underwater active polarization imaging method without prior knowledge was presented by Zhao [Zhao Y, He W, Ren H, Li Y, Fu Y <ext-link ext-link-type="uri" xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="http://doi.org/10.1016/j.optlaseng.2021.106777">2022 <i>Opt. Lasers Eng.</i> <b>148</b> 106777</ext-link>], in which the DoLPs of target light and backscattered light can be automatically obtained without background region. However, sometimes the above two parameters are very close and thus introduce a lot of noise into the restored images, for this method takes only the contrast into account.</sec><sec>In this work, an underwater active polarization imaging method based on two-layer multi-index optimization is proposed. First, the mutual information and contrast are taken as the upper objective functions, and the Pareto optimal solution set is obtained by the multi-objective genetic optimization algorithm. Second, the information entropy is taken as the lower objective function to obtain the optimal parameters from this optimal solution set. Based on the optimal parameters, the restored images are obtained. According to the difference between the DoLPs of target light and backscattered light, these restored images are further improved by the digital image processing method.</sec><sec>The experimental results indicate that our method can not only enhance image details effectively but also balance various evaluation indexes of the imaging quality to obtain high-quality restored images. The proposed algorithm is suitable for underwater targets with low and high DoLPs, with or without background regions.</sec>

Underwater polarization imaging for visibility enhancement of moving targets in turbid environments.

Polarization imaging techniques have more prominent advantages for imaging in strongly scattered media. Previous de-scattering methods of polarization imaging usually require the priori information of the background region, and rarely consider the effect of non-uniformity of the optical field on image recovery, which not only reduces the processing speed of imaging but also introduces errors in image recovery, especially for moving targets in complex scattering environments. In this paper, we propose a turbid underwater moving image recovery method based on the global estimation of the intensity and the degree of polarization (DOP) of the backscattered light, combined with polarization-relation histogram processing techniques. The full spatial distribution of the intensity and the DOP of the backscattered light are obtained by using frequency domain analysis and filtering. Besides, a threshold factor is set in the frequency domain low-pass filter, which is used to adjust the execution region of the filter, which effectively reduces the error in image recovery caused by estimating the DOP of the backscattered light as a constant in traditional methods with non-uniform illumination. Meanwhile, our method requires no human-computer interaction, which effectively solves the drawbacks that the moving target is difficult to be recovered by traditional methods. Experimental studies were conducted on static and moving targets under turbid water, and satisfactory image recovery quality is achieved.

Active non-uniform illumination-based underwater polarization imaging method for objects with complex polarization properties.

Active polarization imaging is one of the most effective underwater optical imaging methods that can eliminate the degradation of image contrast and clarity caused by macro-molecule scattering. However, the non-uniformity of active illumination and the diversity of object polarization properties may decrease the quality of underwater imaging. This paper proposes a non-uniform illumination-based active polarization imaging method for underwater objects with complex optical properties. Firstly, illumination homogenization in the frequency domain is proposed to extract and homogenize the natural incident light from the total receiving light. Then, the weight values of the polarized and non-polarized images are computed according to each pixel's degree of linear polarization (DoLP) in the original underwater image. By this means, the two images can be fused to overcome the problem of reflected light loss generated by the complex polarization properties of underwater objects. Finally, the fusion image is normalized as the final result of the proposed underwater polarization imaging method. Both qualitative and quantitative experimental results show that the presented method can effectively eliminate the uneven brightness of the whole image and obtain the underwater fusion image with significantly improved contrast and clarity. In addition, the ablation experiment of different operation combinations shows that each component of the proposed method has noticeable enhancement effects on underwater polarization imaging. Our codes are available in Code 1.

Underwater Polarization Imaging Recovery Based on Polarimetric Residual Dense Network

Application of deep-learning to polarization imaging technology for image restoration has led to many technological breakthroughs, especially in underwater image recovery and recognition. In this work, a four-input deep learning model with the Polarimetric Residual Dense Network is proposed for underwater image recovery. The diverse polarization component images are trained and tested in different processes in the network for the recognition and dehazing by considering the physical model of polarization dehazing imaging. Our study reveals that the proposed method can efficiently recover the hazed images, and provide good performance for improving the quality of image restoration even in a high-turbidity complex underwater environment.

Local optimum underwater polarization imaging enhancement based on connected domain prior

Estimating the degree of polarization (DOP) of target signal light () is a critical step in the underwater polarization differential imaging process. It is challenging to accurately estimate the of objects containing multiple attributes in the imaging range. Inspired by the optical polarization characteristics of objects with different DOP, we find a prior of connected domains (CDs) of objects in polarization orthogonal images. Based on this prior, the CDs of low-DOP (LDOP) and high-DOP (HDOP) objects are extracted from the darkest image () and brightest image () without additional image acquisition. The of LDOP and HDOP objects in the scene can be estimated through the extracted CDs, which realizes the transition from global optimum estimation in the traditional method to local optimum estimation. According to the estimated , the enhanced results of the entire image marked by can be obtained after fusing the enhanced results of the objects in each CD. Experimental results of simple and complex combination scenes of objects show that our method can recognize CDs of LDOP and HDOP objects. We obtained the enhanced results of the entire image, verified the feasibility and robustness of our method, and found that the CDs of LDOP and HDOP objects could be effectively extracted in up to NTU of turbid water. The local optimum improves the application ability of underwater polarization imaging in unpredictable scenarios, while the enhanced result of target labeling has potential application value in underwater target recognition and classification.

Image-restoration algorithm based on an underwater polarization imaging visualization model.

The polarization bidirectional reflection distribution function theory of a target is combined with microfacet theory, and the Monte Carlo method is used to establish an underwater laser active-polarization imaging model based on Mie scattering theory. The model presented herein can simulate imaging of an underwater target with a high degree of polarization, and the effects of optical thickness and target surface roughness on active underwater laser imaging results are demonstrated by the simulation image. Combined with histogram equalization and the traditional polarization differential imaging algorithm, an algorithm is presented herein that globally estimates the mutual information value between the target polarization degree and the correction factor of backscattered light polarization degree. The images received from the simulation test can be restored, and results show that the algorithm can restore the target image with a high degree of polarization to some extent. Finally, the correctness of the active underwater laser polarization imaging model and the feasibility of global estimation based on the polarization differential restoration algorithm are verified experimentally.

Experimental Study on Bottom-Up Detection of Underwater Targets Based on Polarization Imaging.

Previous studies on the polarization imaging of underwater targets mainly focused on top-down detection; however, the capacities of bottom-up detection were poorly known. Based on in situ experiments, the capability of bottom-up detection of underwater targets using polarization imaging was investigated. First, to realize the objective of bottom-up polarization imaging, a SALSA polarization camera was integrated into our Underwater Polarization Imaging System (UPIS), which was integrated with an attitude sensor. At Qiandao Lake, where the water is relatively clear, experiments were conducted to examine the capacity of the UPIS to detect objects from the bottom up. Simultaneously, entropy, clarity, and contrast were adopted to compare the imaging performance with different radiation parameters. The results show that among all the used imaging parameters, the angle of polarization is the optimal parameter for bottom-up detection of underwater targets based on polarization imaging, which may result from the different diffused reflectance of the target surface to the linear polarization components of the Stokes vector.

Fast processing of underwater polarization imaging based on optical correlation.

Underwater polarization differential imaging requires the estimation of different parameters, and the parameters can be accurately obtained by using optical correlation. However, optical correlation as a criterion function to estimate parameters takes a lot of time. To expedite the parameters' estimation process, we propose two operations to process underwater polarization images. One operation is to update the analyzer angle range to reduce the number of processed images. The other is image downsampling, which reduces the amount of calculation for the corresponding images. In experiments, we confirmed the feasibility of our method. We have obtained an average of 42 times the calculation speed increase under the conditions of updating the analyzer angle range 3 times and reducing the image scale by 16 times. The results of our method are consistent with those of traditional methods. This established method is conducive to the practical application of underwater polarization differential imaging.

Clear underwater vision in non-uniform scattering field by low-rank-and-sparse-decomposition-based olarization imaging

<sec>Underwater imaging plays a critical role in marine rescue, seabed resource exploration, underwater archaeology, etc. by providing human-vision-system-friendly information. A variety of approaches have been exploited to realize clear underwater imaging. Noticeably, underwater polarization imaging has attracted attention due to its simple imaging system and clear vision. It can remove the backscattered light from degraded image and recover abundant high-fidelity information of target. Descattering is conducted by using the difference in polarization characteristics between the target and background. A classical underwater polarization imaging method is presented by Schechner [Tali T, Schechner Y Y <ext-link ext-link-type="uri" xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="http://doi.org/10.1109/JPHOT.2017.2698000">2009 <i>IEEE Trans. Pattern Anal. Mach. Intell.</i> <b>31</b> 385</ext-link>], in which the differential polarization characteristics of backscattered light and target light are used to recover clear image. More researches were conducted including Huang et al.’s research [Huang B J, Liu T G, Hu H F, et al. <ext-link ext-link-type="uri" xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="http://doi.org/10.1364/OE.24.009826">2016 <i>Optics Express</i> <b>24</b> 9826</ext-link>], Liu et al.’s study [Liu F, Han P L, Wei Y, et al. <ext-link ext-link-type="uri" xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="http://doi.org/10.1364/OL.43.004903"> 2018 <i>Opt. Lett.</i> <b>43</b> 4903</ext-link>], etc.</sec><sec>However, in the polarization imaging methods, the uniform underwater backscattered light and polarization parameters over the whole image are usually assumed. In most practical applications, these assumptions cannot hold true. Therefore, the inaccurate estimation of backscattered light makes it difficult to completely descatter an image, leading many methods to fail to detect the target in non-uniform turbid water.</sec><sec>In this study, we propose a low-rank-and-sparse-decomposition-based polarization imaging combined with common mode rejection feature of polarization information in scattered light field to eliminate non-uniformity and scattering caused by severe scattering during active polarization imaging of turbid water. The backscattered light is highly reduced and the information contained in background is single and highly correlated. It conforms to the low-rank characteristics of the image. What is more, the target in underwater scene occupies a relatively small proportion, which conforms to the sparsity characteristics of the image. Therefore, combining the low-rank characteristics of backscattered light with the sparse characteristics of target information light, we separate them through low-rank and sparse matrix decomposition to recover clear underwater image. Both experimental and objective image quality evaluation results demonstrate the validity of the proposed method.</sec><sec>The proposed method works well in improving polarization vision in non-uniform turbid water, which is due to its ability to make the underwater scene uniform and the target and background information separated through their distribution difference of polarization characteristics. It possesses potential applications in turbid water imaging.</sec>

Polarization descattering imaging: a solution for nonuniform polarization characteristics of a target surface

This paper presents a polarization descattering imaging method for underwater detection in which the targets have nonuniform polarization characteristics. The core of this method takes the nonuniform distribution of the polarization information of the target-reflected light into account and expands the application field of underwater polarization imaging. Independent component analysis was used to separate the target light and backscattered light. Theoretical analysis and proof-of-concept experiments were employed to demonstrate the effectiveness of the proposed method in estimating target information. The proposed method showed superiority in accurately estimating the target information compared with other polarization imaging methods.