Towards addressing the issues of poor image quality and limited sample quantity in underwater sonar images. A genetic encoding method for the fusion of image spot and local binary pattern texture features is employed in this study. By encoding the fusion features using genetic algorithms, the dimensionality of the fused features is reduced, thereby enhancing the efficiency of neural network operations while ensuring data transmission confidentiality. Additionally, a deep multi-classifier is introduced, which combines residual units with CNN-BiGRU-Attention. This classifier takes the genetically encoded fused features as input and outputs four predicted probabilities. Through an ensemble learning strategy that evenly assigns weights to the predicted probabilities, the final prediction result is obtained, achieving underwater target recognition. Results demonstrate that the proposed fusion features effectively enlarge inter-class differences among various types of sonar images and improve classifier accuracy. The deep multi-classifier model exhibits stable improvement in prediction accuracy compared to the original model, converging after only 15 training iterations with an average recognition accuracy of 99.1%. The research findings offer an effective approach for target sonar image recognition and classification, with potential applicability to a wider range of underwater target categorization.

The coupling mechanism between natural and anthropogenic noise in shallow marine areas is of great significance for maintaining the ecological safety of these regions. In this study, a section of Zhanjiang Bay’s entrance was selected as a typical research area, and environmental noise data at different depths were collected during the spring and autumn seasons. The spectral characteristics, sound pressure levels, and underwater noise frequency correlation matrices of environmental noise were analyzed to reveal the underwater noise characteristics of tidal channels in Zhanjiang Bay and their main influencing factors. The results show that underwater noise in this study area had a stable frequency band distribution. In the low-frequency range of 20–50 Hz, the main source of noise was the flow noise influenced by tides and topography, with a peak sound pressure level of approximately 97 dB. In the frequency range of 50 Hz to 500 Hz, the main noise sources were ships at sea, followed by wind-generated noise. At frequencies above 500 Hz, the noise intensity decreased. In addition, it was found that the sound pressure level in the low-frequency range had a significant correlation with the tidal level, increasing with the rise of the tide and decreasing during low tides. This study provides a research case on the impact that human noise activity has on environmental noise in shallow marine bays. These research findings can support the selection of sites and reduce construction noise from offshore wind farms, as well as ensure the acoustic ecological environment in the vicinity of marine ranches.