An innovative marine fine-scale drifting buoy and its networked observation system were designed. The buoy hardware featured a self-developed detachable wave-resistant structure and low-power core component integration technology. With the help of its dual-antenna communication architecture, the buoy was subjected to laboratory and field lake tests to quantitatively evaluate its measurement accuracy, environmental adaptability, and data stability. Based on LoRa-enabled distributed IoT communication technology, an innovative topological networking architecture was used to construct a four-node, 10 km-scale 3D observation network with a packet loss rate of less than 0.1%. FFT spectral analysis was conducted to analyze data features, and a noise reduction method was proposed. Test results showed that the network could synchronously capture quasi-steady-state ocean environmental data, providing a scalable solution for high-density, multi-dimensional, fine-scale ocean monitoring. The research results were applied in teaching practice. In addition, expansion to a 10-node network has been planned to advance the engineering application of fine-scale ocean observation technology.

Synthetic aperture radar (SAR) is widely employed in geospatial mapping, national defense, and microwave remote sensing due to its capability for high-resolution imaging under all-weather, all-day conditions. Consequently, developing effective jamming techniques to degrade SAR-based target reconnaissance and identification has become a critical challenge in radar countermeasure research. This study proposes a method to generate importance distribution maps through spatial coordinate transformation, capturing the spatial relationships and relative importance of targets within a scene. Based on this map, a SAR passive jamming control method is developed. A U-Net-based architecture is constructed to design a decision-making algorithm for radar jamming strategies, aiming to prioritize the protection of critical targets with minimal resource expenditure. Simulation results demonstrate that this method visualizes the interference problem by constructing an importance distribution map and uses advanced artificial intelligence algorithms for the control and decision-making of passive motion interference devices, achieving collaborative decision-making among multiple interference devices and significantly enhancing the intelligence level of SAR passive interference. Across different scenarios, the algorithm-generated jamming strategies effectively disrupt the target areas, with jamming strips fully covering the regions of interest in each case.

The volume scattering function (VSF) of aquatic particles is one of the most important inherent optical properties of water. However, a VSF measurement instrument in China that can cover multiple wavelengths and a wide angular range remains to be investigated. Herein, we developed a dual-wavelength (488 nm and 532 nm) aquatic particle VSF measurement system that integrates dual-periscope optical configuration with a rotating detector. This system enables VSF measurements across a scattering angle range of 1.5°~178.5° for both wavelengths. According to the optical configuration and radiative transfer principles of the system, baseline correction, angular calibration, and amplitude calibration experiments were conducted on the system. The measurement results for 3 μm polystyrene standard particles agreed well with the Mie scattering theoretical values, demonstrating the VSF measurement accuracy of the system. The system was used to perform dual-wavelength VSF measurements of natural seawater particles from the East China Sea and South China Sea and to analyze the differences in VSF characteristics between the two wavelengths. The measured VSFs can provide a basis for VSF parameterization in the radiative transfer models. In addition, asymmetry factors were calculated from the measured data to analyze the differences in scattering characteristics between the two wavelengths from the volume scattering perspective.

The extraction of cage aquacultural areas was investigated using high-resolution GF-1 and GF-2 remote sensing images from northern Fujian Province. Image enhancement was performed by correction, fusion, and cropping. The sample database of inland cage culture areas of two kinds of images was constructed; The sample bank is used to train the in-depth learning fully convolutional networks (FCN) model extracted from inland cage culture area and verify the accuracy. The results of the test experiment show that the F-measure of GF-1 and GF-2 reaches 83.37% and 92.56%,respectively. It shows that the inland cage culture area extraction based on FCN has high accuracy, and can be used for large-scale inland cage acquaculture area extraction, which provides an important basis for the monitoring of inland aquaculture area.