There are a large number and variety of microalgae in the ocean, which are important natural resources for extensive applications. However, how to identify and separate microalgae quickly and effectively remains a major challenge for investigation and utilization of microalgae resources. There are many methods for cell separation, while each of them has pros and cons. Therefore, it is difficult to meet the particular requirements of separating complex samples, such as microalgae, solely with a single method. Herein, a two-stage microfluidic separation chip was designed by combining inertial and dielectrophoretic (DEP) forces, which can achieve rapid and accurate separation of three common microalgae cells, including Platymonas, Closterium and Chlorella, with separation efficiency exceeding 90%. In the first-stage, high-throughput inertial separation technique was used to separate microalgae with their different equilibrium positions. Then, the DEP force was used to separate microalgae cells which were not separated in the first-stage through their dielectric properties. To the best of our knowledge, there is no prior reports of microalgae separation by combining DEP and inertial techniques. In addition, the DEP characteristics of microalgae cells under a non-uniform AC electric field were firstly reported in this study, and the DEP responses of each cell type were measured against AC electrical frequency changes. These findings may contribute to our understanding of the dielectric properties of microalgae, as well as the development of novel techniques for microalgae separation, detection, and marine environmental monitor and analysis.

The mechanical response of the marine reef sand–geogrid (RG) interface can be influenced by a high-temperature climate, grain size, and variable stress environments. These factors are critical to the effectiveness of geogrid reinforcement in reef sand engineering. However, there are few studies on the influences of grain size, temperature, and stress history on the mechanical characteristics of RG interfaces, with most studies centering on the influence of single factors on the mechanical characteristics of RG interfaces. In this paper, based on self-developed temperature-controlled large interface shear equipment, a series of before/post-cyclic shear tests were carried out on RG interfaces in the temperature range of 5–80 °C. The impact of different reef sand grain sizes on the RG interface was explored (S1: 1–2 mm; S2: 2–4 mm). It was shown that temperature and grain size had significant influences on the mechanical characteristics of the RS interface. Compared with the S1 RG interfaces, the S2 RG interfaces had higher sensitivity to temperature changes with respect to the before/post-cyclic maximum shear strength. Moreover, in comparison to the before-cyclic shear strength, the post-cyclic maximum shear strength is more responsive to temperature changes. The before/post-cyclic maximum shear strength of the S2 RG interfaces was greater than the maximum shear strength of the S1 RG interfaces as the temperature changed. Based on the results of physical tests, a machine learning model containing 450 datasets was constructed, which can accurately predict the shear strength of the RG interface.

Detection of live microalgae is critical to ballast water treatment. A novel handheld high-throughput ballast water rapid detection device based on microfluidic chip and fluorometry for counting single living microalgae cell is demonstrated in this article. The viability of individual microalgae cells can be evaluated and the number of the living microalgae cells in an aqueous solution sample can be obtained at a rate of one milliliter per minute in a fully automatic and real-time manner. It is experimentally confirmed that the detection results of this device are not influenced by free chlorophyll from the lysed microalgae, nor by the salinity and turbidity of the sample within the variation range of ballast water. Furthermore, this device has been used in a port to test the actual ballast water samples processed by ballast water treatment systems based on different principles or loaded at different locations, and the conclusion of this device on the qualification of the ship is consistent with the conclusion of the laboratory standard test. To the best of our knowledge, this is the first ballast water rapid detection device using a microfluidic chip as the detection platform, which can be used on board of ships for on-site counting the numbers of the living microalgae cells in ballast water.

The application potential of unmanned aerial vehicles (UAVs) in marine search and rescue is especially of concern for the ongoing advancement of visual recognition technology and image processing technology. Limited computing resources, insufficient pixel representation for small objects in high-altitude images, and challenging visibility conditions hinder UAVs’ target recognition performance in maritime search and rescue operations, highlighting the need for further optimization and enhancement. This study introduces an innovative detection framework, CFSD-UAVNet, designed to boost the accuracy of detecting minor objects within imagery captured from elevated altitudes. To improve the performance of the feature pyramid network (FPN) and path aggregation network (PAN), a newly designed PHead structure was proposed, focusing on better leveraging shallow features. Then, structural pruning was applied to refine the model and enhance its capability in detecting small objects. Moreover, to conserve computational resources, a lightweight CED module was introduced to reduce parameters and conserve the computing resources of the UAV. At the same time, in each detection layer, a lightweight CRE module was integrated, leveraging attention mechanisms and detection heads to enhance precision for small object detection. Finally, to enhance the model’s robustness, WIoUv2 loss function was employed, ensuring a balanced treatment of positive and negative samples. The CFSD-UAVNet model was evaluated on the publicly available SeaDronesSee maritime dataset and compared with other cutting-edge algorithms. The experimental results showed that the CFSD-UAVNet model achieved an mAP@50 of 80.1% with only 1.7 M parameters and a computational cost of 10.2 G, marking a 12.1% improvement over YOLOv8 and a 4.6% increase compared to DETR. The novel CFSD-UAVNet model effectively balances the limitations of scenarios and detection accuracy, demonstrating application potential and value in the field of UAV-assisted maritime search and rescue.

The identification and detection of microalgae are essential for the development and utilization of microalgae resources. Traditional methods for microalgae identification and detection have many limitations. Herein, a Feature-Enhanced YOLOv7 (FE-YOLO) model for microalgae cell identification and detection is proposed. Firstly, the feature extraction capability was enhanced by integrating the CAGS (Coordinate Attention Group Shuffle Convolution) attention module into the Neck section. Secondly, the SIoU (SCYLLA-IoU) algorithm was employed to replace the CIoU (Complete IoU) loss function in the original model, addressing the issues of unstable convergence. Finally, we captured and constructed a microalgae dataset containing 6300 images of seven species of microalgae, addressing the issue of a lack of microalgae cell datasets. Compared to the YOLOv7 model, the proposed method shows greatly improved average Precision, Recall, mAP@50, and mAP@95; our proposed algorithm achieved increases of 9.6%, 1.9%, 9.7%, and 6.9%, respectively. In addition, the average detection time of a single image was 0.0455 s, marking a 9.2% improvement.

To address the challenges of sparse point clouds in current MIMO millimeter-wave radar environmental mapping, this paper proposes a dense 3D millimeter-wave radar point cloud environmental mapping algorithm. In the preprocessing phase, a radar SLAM-based approach is introduced to construct local submaps, which replaces the direct use of radar point cloud frames. This not only reduces data dimensionality but also enables the proposed method to handle scenarios involving vehicle motion with varying speeds. Building on this, a 3D-RadarHR cross-modal learning network is proposed, which uses LiDAR as the target output to train the radar submaps, thereby generating a dense millimeter-wave radar point cloud map. Experimental results across multiple scenarios, including outdoor environments and underground tunnels, demonstrate that the proposed method can increase the point cloud density of millimeter-wave radar environmental maps by over 50 times, with a point cloud accuracy better than 0.1 m. Compared to existing algorithms, the proposed method achieves superior environmental map reconstruction performance while maintaining a real-time processing rate of 15 Hz.