Underwater target detection and segmentation in Side-Scan Sonar (SSS) imagery is challenged by low signal-to-noise ratios, geometric distortions, and Unmanned Underwater Vehicles (UUVs)’ computational constraints. This paper proposes CKAN-YOLOv8, a lightweight multi-task network integrating Kolmogorov–Arnold Networks Convolution (KANConv) into YOLOv8. The core innovation replaces conventional convolutions with KANConv blocks using learnable B-spline activations, dynamically adapting to noise and multi-scale targets while ensuring parameter efficiency. The KANConv-based Path Aggregation Network (KANConv-PANet) mitigates geometric distortions through spline-optimized multi-scale fusion. A dual-task head combines CIoU loss-driven detection and a boundary-sensitive segmentation module with Dice loss. Evaluated on a dataset (50 raw images augmented to 2000), CKAN-YOLOv8 achieves state-of-the-art performance as follows: 0.869 AP@0.5 and 0.72 IoU, alongside real-time inference at 66 FPS. Ablation studies confirm the contributions of KANConv modules to noise robustness and multi-scale adaptability. The framework demonstrates exceptional robustness to noise, scalability across target sizes.

Underwater target detection and segmentation in Side-Scan Sonar (SSS) imagery is challenged by low signal-to-noise ratios, geometric distortions, and Unmanned Underwater Vehicles (UUV) computational constraints. This paper proposes CKAN-YOLOv8, a lightweight multi-task network integrating Kolmogorov-Arnold Networks Convolution (KANConv) into YOLOv8. The core innovation replaces conventional convolutions with KANConv blocks using learnable B-spline activations, dynamically adapting to noise and multi-scale targets while ensuring parameter efficiency. KANConv-based feature pyramid (KANConv-PANet) mitigates geometric distortions through spline-optimized multi-scale fusion. A dual-task head combines CIoU loss-driven detection and a boundary-sensitive segmentation module with Dice loss. Evaluated on a dataset (50 raw images augmented to 2000), CKAN-YOLOv8 achieves state-of-the-art performance: 0.869 AP@0.5 and 0.72 IoU, alongside real-time inference at 66 FPS. Ablation studies confirm the contributions of KANConv modules to noise robustness and multi-scale adaptability. The framework demonstrates exceptional robustness to noise, scalability across target sizes.

This paper proposes a novel SDA-Mask R-CNN framework for precise seabed terrain edge feature extraction from Side-Scan Sonar (SSS) images to enhance Unmanned Underwater Vehicle (UUV) perception and navigation. The developed architecture addresses critical challenges in underwater image analysis, including low segmentation accuracy and ambiguous edge delineation, through three principal innovations. First, we introduce a Structural Synergistic Group-Attention Residual Network (SSGAR-Net) that integrates group convolution with an enhanced convolutional block attention mechanism, complemented by a layer-skipping architecture for optimized information flow and redundancy verification for computational efficiency. Second, a Depth-Weighted Hierarchical Fusion Network (DWHF-Net) incorporates depthwise separable convolution to minimize computational complexity while preserving model performance, which is particularly effective for high-resolution SSS image processing. This module further employs a weighted pyramid architecture to achieve multi-scale feature fusion, significantly improving adaptability to diverse object scales in dynamic underwater environments. Third, an Adaptive Synergistic Mask Optimization (ASMO) strategy systematically enhances mask generation through classification head refinement, adaptive post-processing, and progressive training protocols. Comprehensive experiments demonstrate that our method achieves 0.695 (IoU) segmentation accuracy and 1.0 (AP) edge localization accuracy. The proposed framework shows notable superiority in preserving topological consistency of seabed features, offering a reliable technical framework for underwater navigation and seabed mapping in marine engineering applications.

A stereo vision system provides important support for underwater robots to achieve autonomous navigation, obstacle avoidance, and precise operation in complex underwater environments. This article proposes an unsupervised underwater stereo matching method based on semantic attention. By combining deep learning and semantic information, it fills the challenge of insufficient training data, enhances the intelligence level of underwater robots, and promotes the progress of underwater scientific research and marine resource development. This article proposes an underwater unsupervised stereo matching method based on semantic attention, targeting the missing training supervised dataset for underwater stereo matching. An adaptive double quadtree semantic attention model for the initial estimation of semantic disparity is designed, and an unsupervised AWLED semantic loss function is proposed, which is more robust to noise and textureless regions. Through quantitative and qualitative evaluations in the underwater stereo matching dataset, it was found that D1 all decreased by 0.222, EPE decreased by 2.57, 3px error decreased by 1.53, and the runtime decreased by 7 ms. This article obtained advanced results.