Jianbing Chen

Solid mineral resources form the essential material foundation for the sustainable development of human society. The international seabed hosts vast and potentially valuable mineral deposits, among which deep-sea polymetallic nodules represent a key marine resource. Focusing on a specific area in the Western Pacific, this study aims to identify and classify small-scale polymetallic nodules in the deep-sea environment. We employ super-resolution reconstruction methods to enhance the resolution of deep-sea hydroacoustic images. Subsequently, a super-pixel segmentation approach is applied to construct a sample enhancement model for deep-sea objects, enabling in-depth extraction of multi-dimensional heterogeneous features from seabed targets and facilitating the effective development of training samples. Constrained by geological seabed samples, an accurate recognition model for seabed polymetallic nodules is established, achieving intelligent mineral classification based on multi-source data such as bathymetry and backscatter. Ultimately, by leveraging the generalization capability of the model, the recognition and classification of untrained samples can be accomplished, thereby promoting the application of the proposed algorithm in large-scale deep-sea mineral exploration.

The multibeam echo sounder (MBES) is a key tool for acquiring high-precision seabed topographic data. However, measurement gaps resulting from its swath-based measurement mode are prevalent, severely compromising the completeness of seabed terrain modeling. To address this issue, this study first categorizes multibeam data gaps into two data degradation patterns with clear hydrographic survey backgrounds: “random degradation” and “rule-based degradation.” Based on this categorization, a highly realistic training dataset that closely matches actual conditions is constructed. To improve the reconstruction accuracy and topographic fidelity, a novel multibeam echo sounder data reconstruction model, the MBES-DDPM, is proposed. Based on the denoising diffusion probabilistic model (DDPM) framework, this model innovatively incorporates gravity anomaly data as prior knowledge. Then, with a designed multisource data fusion guidance mechanism, macro-topographic structural constraints are injected during the diffusion process. Furthermore, a targeted quantitative and qualitative evaluation system is established. The experimental results show that compared with the baseline methods, the MBES-DDPM achieves the best performance across various complex scenarios. Its restored results exhibit an average reduction in root mean square error of at least 34.21% and an average increase in peak signal-to-noise ratio of more than 3.71 dB. Furthermore, it achieves the highest reconstruction fidelity in teams of the terrain slope accuracy metrics. Thus, this research provides a new and reliable solution for accurately restoring large-scale MBES data.