Yifan Huang

Optical attenuation caused by absorption and scattering in turbid water significantly degrades underwater image quality, making reliable underwater imaging a challenging problem. Underwater polarization imaging has attracted increasing attention because of its ability to suppress scattered light and provide additional polarization cues. However, existing polarization-based enhancement approaches often adapt conventional underwater image enhancement strategies, and the multi-dimensional characteristics of polarization information are not always fully utilized, which may limit detail restoration in complex underwater environments. To address this issue, this paper proposes a bio-inspired underwater polarization image enhancement framework motivated by the polarization vision mechanism of marine organisms. Specifically, a two-stage architecture consisting of a Polarization Adversarial Network (PAN) and a Polarization Enhancement Network (PEN) is designed. The PAN incorporates a Bionic Antagonistic Module (BAM) to exploit complementary information among polarization channels, while Salient Feature Extraction (SFE) is introduced to reduce redundant feature interference. The subsequent PEN integrates a frequency-aware Mamba-based structure to enhance feature representation and improve detail reconstruction. Experiments on simulated underwater polarization datasets indicate that the proposed framework can effectively suppress backscattering and improve structural detail visibility in challenging underwater scenes, demonstrating competitive performance compared with representative traditional and learning-based methods.

Underwater optical imaging, known for its high resolution and rich color information, is widely used in underwater exploration and monitoring. However, it is affected by light absorption and scattering, leading to color attenuation, blurring, and low contrast, which impact visual tasks. Meanwhile, the development of underwater applications towards mobile platforms poses challenges for the efficient deployment of models due to limited computational resources. To address these challenges, this paper proposes a lightweight underwater image enhancement network, WDGA L (Water DOOC GAN Air Light), which integrates human visual dual-opponent characteristics with knowledge distillation. Inspired by human color constancy, the network is designed with a dual-opponent mechanism to enhance color restoration and detail clarity, thereby improving adaptability. Additionally, a knowledge distillation strategy is incorporated, introducing fog density feature imitation loss and adaptive local self-feature distillation loss. These components enable the student network to significantly reduce parameter size while maintaining enhancement performance and improving computational efficiency. The proposed lightweight model is successfully deployed on the RK3588 platform, achieving 32.8 FPS for real-time underwater image observation. Experimental results demonstrate that, compared to state-of-the-art (SOTA) methods, WDGA L improves UCIQE and UIQM metrics by 23.7%, verifying its practicality and feasibility in resource constrained underwater environments. WDGA L shows promising applications in marine intelligent devices, offering an efficient and lightweight solution for underwater image enhancement.

Underwater optical images are a crucial information source for autonomous underwater vehicles during underwater development and exploration. When these vehicles are in operation, they need to capture high-quality images for information extraction and analysis and require depth map information from the underwater scene to maintain the vehicle's posture balance, obstacle avoidance, and navigation. However, the absorption and scattering of light in water result in low-quality underwater images, significantly affecting the execution of these tasks. In response to these challenges, this article proposes a physically guided, semi-supervised dual-loop network for underwater image enhancement. This network is designed to accomplish high-quality underwater image enhancement and depth map estimation simultaneously. First, the revised underwater image formation model is employed to guide a two-stage network in decomposing and reconstructing underwater images. The depth map consistency of the scene and piecewise cycle consistency loss are utilized to ensure the reliability of the image transformation process. In another loop, a self-augmentation module based on inherent optical properties is introduced to enhance the robustness of the decomposition network. A multimodal discriminator is incorporated to form piecewise adversarial loss to improve the visual quality of the images. Through extensive experimental evaluation and analysis, the proposed method not only demonstrates outstanding performance in underwater image enhancement and depth map estimation but also reveals the relationships between various physical quantities during the degradation process of underwater images, enhancing the physical interpretability of the neural network.

Optical remote sensing serves as a critical technology for exploring underwater environments. However, light absorption and scattering underwater significantly degrade underwater optical images, affecting the extraction and analysis of information. Underwater image enhancement methods aim to eliminate this degradation and improve the visual quality of images. Nonetheless, the complex and dynamic underwater imaging environment, limited computing resources, and scarce training data/data pairs restrict the practical application of existing methods. To solve these problems, we propose an underwater image enhancement network based on neuromorphic computing, which simulates the pathway of the visual system to perceive and process light information, and can use a lightweight network structure to achieve good performance through unsupervised learning. Specifically, we propose a visual perception module comprising a two-dimensional Duffing oscillator with pixel-wise potential barrier parameters. This module can generate the stochastic resonance phenomenon to enhance the degraded image. Inspired by physics-informed learning, a dual-path neural network is employed to estimate the potential barrier parameters and solve the partial differential equation that describes the visual perception module. Subsequently, we introduce three non-reference losses to guide the network training and improve the enhanced image’s visual quality. Extensive experiments demonstrate that the proposed method can achieve outstanding performance with less computing resource cost compared to state-of-the-art methods. Furthermore, we examine the generalization and versatility of the proposed method to establish its reliability across various degradation types and tasks in practical applications of optical remote sensing.

Underwater detection faces uncomfortable illumination conditions, and traditional optical images sensitive to intensity often cannot work well in these conditions. Polarization imaging is a good solution for underwater detection under adverse lighting conditions. However, the process of obtaining polarization information causes it to be more sensitive to noise; serious noise reduces the quality of polarized images and subsequent performance in advanced visual tasks. Unfortunately, the flourishing low-light image enhancement methods applied to intensity images have not demonstrated satisfactory performance when transferred to polarized images. In this paper, we propose a low-light image enhancement paradigm based on the antagonistic properties of polarization parameters. Furthermore, we develop a dual-branch network that relies on a gradient residual dense feature extraction module (GRD) designed for polarized image characteristics and polarization loss, effectively avoiding noise introduced during the direct amplification of brightness, and capable of restoring target contour details. To facilitate a data-driven learning method, we propose a simulation method for underwater low-light polarized images. Extensive experimental results on real-world datasets demonstrate the effectiveness of our proposed approach and its superiority against other state-of-the-art methods.

Due to underwater light absorption and scattering, underwater images usually suffer from severe color attenuation and contrast reduction. Most mainstream underwater image processing methods based on deep learning require a large amount of underwater paired training data, leading to a complex network structure, longer training time, and higher computational cost. To address this problem, a novel Zero-Reference Deep Network for Underwater Image Enhancement (Zero-UIE) is proposed in this paper, which transforms the enhancement of an underwater image into a specific parameter map estimation by using a deep network. The underwater curve model based on the classical haze image formation principle is specially designed to remove underwater color dispersion and cast. A lightweight deep network is designed to estimate the dynamic adjustment parameters of the underwater curve model, and then adjust the dynamic range of the given image pixels according to the model. A set of non-reference loss functions are designed according to the characteristics of underwater images, which can implicitly drive the network learning. In addition, adaptive color compensation can be optionally used as the pre-processing step to further improve the robustness and visual performance. The significant contribution of the proposed method is zero reference, i.e., it does not require any paired or unpaired reference data for training. Extensive experiments on various benchmarks demonstrate that the proposed method is superior to state-of-the-art methods subjectively and objectively, which is competitive and applicable to diverse underwater conditions. Most importantly, it is an innovative exploration of zero reference for underwater image enhancement.

In turbid water, the attenuation and scattering of light caused by scatterers make underwater optical images degraded, blurred, and contrast reduced, limiting the extraction and analysis of information from images. To address such problems, a turbid underwater image enhancement method based on parameter-tuned stochastic resonance (SR) is proposed in this article. First, an SR algorithm framework for underwater image enhancement is constructed, including the dimensionality reduction and normalization of input images, the solution and parameter optimization of the SR system, the dimensionality upgrading of output images, etc. This framework can apply the SR's ability to enhance weak signals to the enhancement of turbid underwater images. Second, to measure the performance of the system, a synthetic turbid underwater image data set (UWCHIC) is constructed using the underwater imaging model and an image set with simulated scatterers. Based on this data set, the relationship between various image quality evaluation metrics and system parameters is analyzed, and then the suitable no-reference (NR) metrics for system performance evaluation are selected and an adaptive parameter tuning strategy of the SR system is proposed to guide the image enhancement. Lastly, the proposed method is evaluated on the UWCHIC, a dataset to evaluate underwater image restoration methods (TURBID), marine underwater environment database (MUED), and underwater image enhancement benchmark (UIEB) data sets and the turbid underwater images captured from natural waters. Different experimental evaluations demonstrated that the proposed method not only effectively enhances the visual quality of turbid underwater images but also improves the performance of downstream vision tasks.

As acoustic waves are affected by the channel characteristics, such as scattering and reverberation when propagating in water, sonar images often exhibit speckle noise which will cause visual quality of the image to decrease. Therefore, denoising is a crucial preprocessing technique in sonar image applications. However, speckle noise is mainly caused by the sediment echo signals which are related to the background of seafloor sediment and can be obtained by prior modeling. Although deep learning-based denoising algorithms represent a research hotspot now, they are not suitable for such applications due to the high calculation amount and the large requirement of original images considering that sonar is carried by Autonomous Underwater Vehicles (AUVs) for collecting sonar images and performing calculation. In contrast, dictionary learning-based denoising method is more suitable and easier to be modeled. Compared with deep learning, it can greatly reduce the calculation amount and is more easily integrated into AUV systems. In addition, dictionary learning method based on image sparse representation can effectively achieve image denoising similarly. In order to solve the above problems, we propose a new adaptive dictionary learning method based on multi-resolution characteristics, which combines K-SVD dictionary learning with wavelet transform. Our method has the characteristics of dictionary learning and inherits the features of wavelet analysis as well. Compared with several classical methods, the proposed method is better at speckle noise reduction and edge detail preservation. At the same time, the calculation time is greatly reduced and the efficiency is significantly improved.