Sea fog is a particular kind of atmospheric aerosol that often poses hidden risks to ship navigation, ocean exploration, human productivity, and life. In light of the aforementioned issues, this research conducted a thorough investigation of the polarization transmission properties in a sea fog environment. We studied the physical characteristics of sea fog and established a polarized radiative transfer model based on RT3/PolRadtran (polarized radiative transfer) in a sea fog environment based on the theory of Mie scattering. The effects of wavelength, polarization state, sea fog concentration, and salt content on the polarization degree were simulated by using the polarization transport model. An indoor sea fog simulation device was designed and built. The simulation test results were compared with the experimental test results from many aspects, and the existing errors were analyzed so that they could be mutually verified for the overall trend. According to the modeling results, distinct polarization states of light exhibit evident depolarization as the sea fog concentration rises. Overall, circularly polarized light has superior polarization-maintaining properties compared to linearly polarized light under the same contrast settings. The penetrating impact of incoming light in the visible range improves with increasing wavelength, and the amount of salt in sea fog has some bearing on the degree of polarization.

The color division of focal plane (DoFP) polarization sensor structure mostly uses Bayer filter and polarization filter superimposed on each other, which makes the polarization imaging unsatisfactory in terms of photon transmission rate and information fidelity. In order to obtain high-resolution polarization images and high-quality RGB images simultaneously, we simulate a sparse division of focal plane polarization sensor structure, and seek a sweet spot of the simultaneous distribution of the Bayer filter and the polarization filters to obtain both high-resolution polarization images and high-quality RGB images. In addition, From the perspective of sparse polarization sensor imaging, leaving aside the traditional idea of polarization intensity interpolation, we propose a new sparse Stokes vector completion method, in which the network structure avoids the introduction and amplification of noise during polarization information acquisition by mapping the S₁ and S₂ components directly. The sparsely polarimetric image demosaicing (Sparse-PDM) model is a progressive combined structure of RGB image artifact removal enhancement network and sparsely polarimetric image completion network, which aims to compensate sparsely polarimetric Stokes parameter images with the de-artifacts RGB image as a guide, thus achieving high-quality polarization information and RGB image acquisition. Qualitative and quantitative experimental results on both self-constructed and publicly available datasets prove the superiority of our method over state-of-the-art methods.

In the research of heterogeneous wireless sensor networks, clustering is one of the most commonly used energy-saving methods. However, existing clustering methods face challenges when applied to heterogeneous wireless sensor networks, such as energy balance, node heterogeneity, algorithm efficiency, and more. Among these challenges, a well-designed clustering approach can lead to extended node lifetimes. Efficient selection of cluster heads is crucial for achieving optimal clustering. In this paper, we propose an Enhanced Pelican Optimization Algorithm for Cluster Head Selection (EPOA-CHS) to address these issues and enhance cluster head selection for optimal clustering. This method combines the Levy flight process with the traditional POA algorithm, which not only improves the optimization level of the algorithm, but also ensures the selection of the optimal cluster head. The logistic-sine chaotic mapping method is used in the population initialization, and the appropriate cluster head is selected through the new fitness function. Finally, we utilized MATLAB to simulate 100 sensor nodes within a configured area of 100 × 100 m${}^2$. These nodes were categorized into four heterogeneous scenarios: $m=0,\alpha =0$, $m=0.1,\alpha =2$, $m=0.2,\alpha =3$, and $m=0.3,\alpha =1.5$. We conducted verification for four aspects: total residual energy, network survival time, number of surviving nodes, and network throughput, across all protocols. Extensive experimental research ultimately indicates that the EPOA-CHS method outperforms the SEP, DEEC, Z-SEP, and PSO-ECSM protocols in these aspects.

Underwater object detection is critical for marine ecological monitoring and biodiversity research, yet existing algorithms struggle in detecting densely packed objects of varying sizes, particularly in occluded and complex underwater environments. This study introduces Underwater-Yolo, a novel detection network that enhances performance in these challenging scenarios by integrating a dual-branch occlusion-handling attention mechanism (GLOAM) and a Cross-Stage Partial Dilated Deformable Convolution (CSP-DDC) backbone. The dilated deformable convolutions (DDCs) in the backbone and neck expand the receptive field, thereby improving the detection of small objects, while the deformable convolutions enhance the model’s adaptive feature extraction capabilities for unstructured objects. Additionally, the CARAFE up-sampling operator in the neck aggregates contextual information across a broader spatial domain. The GLOAM, consisting of a global branch (using a Vision Transformer to capture global features and object–background relationships) and a local branch (enhancing the detection of occluded objects through depthwise–pointwise convolutions), further optimizes performance. By incorporating these innovations, the model effectively addresses the challenges of detecting small and occluded objects in dense underwater environments. The evaluation on the CLfish-V1 dataset shows significant improvements over state-of-the-art algorithms, with an AP50 of 93.8%, an AP75 of 88.9%, and an AP-small of 76.4%, marking gains of 4.7%, 16.7%, and 6%, respectively. These results demonstrate the model’s effectiveness in complex underwater scenarios.

A collection of smaller, less expensive sensor nodes called wireless sensor networks (WSNs) use their sensing range to gather environmental data. Data are sent in a multi-hop manner from the sensing node to the base station (BS). The bulk of these sensor nodes run on batteries, which makes replacement and maintenance somewhat difficult. Preserving the network’s energy efficiency is essential to its longevity. In this study, we propose an energy-efficient multi-hop routing protocol called ESO-GJO, which combines the enhanced Snake Optimizer (SO) and Golden Jackal Optimization (GJO). The ESO-GJO method first applies the traditional SO algorithm and then integrates the Brownian motion function in the exploitation stage. The process then integrates multiple parameters, including the energy consumption of the cluster head (CH), node degree of CH, and distance between node and BS to create a fitness function that is used to choose a group of appropriate CHs. Lastly, a multi-hop routing path between CH and BS is created using the GJO optimization technique. According to simulation results, the suggested scheme outperforms LSA, LEACH-IACA, and LEACH-ANT in terms of lowering network energy consumption and extending network lifetime.

The importance of panoramic traffic perception tasks in autonomous driving is increasing, so shared networks with high accuracy are becoming increasingly important. In this paper, we propose a multi-task shared sensing network, called CenterPNets, that can perform the three major detection tasks of target detection, driving area segmentation, and lane detection in traffic sensing in one go and propose several key optimizations to improve the overall detection performance. First, this paper proposes an efficient detection head and segmentation head based on a shared path aggregation network to improve the overall reuse rate of CenterPNets and an efficient multi-task joint training loss function to optimize the model. Secondly, the detection head branch uses an anchor-free frame mechanism to automatically regress target location information to improve the inference speed of the model. Finally, the split-head branch fuses deep multi-scale features with shallow fine-grained features, ensuring that the extracted features are rich in detail. CenterPNets achieves an average detection accuracy of 75.8% on the publicly available large-scale Berkeley DeepDrive dataset, with an intersection ratio of 92.8% and 32.1% for driveableareas and lane areas, respectively. Therefore, CenterPNets is a precise and effective solution to the multi-tasking detection issue.