In the research on navigation efficiency optimization in complex waters, accurate vessel traffic flow prediction has emerged as a critical challenge in the field of intelligent maritime navigation. Based on Resolution A.1158(32) of the International Maritime Organization (IMO) and expert knowledge of Vessel Traffic Service (VTS), this study proposes a novel framework for vessel port reporting and VTS decision-making, and constructs a feature-fused database integrated with meteorological data. A hybrid deep learning model based on the DeepAR framework is developed, which extracts the spatial dependencies of vessel traffic via a convolutional neural network (CNN), captures temporal features using a bidirectional long short-term memory network (BiLSTM), and acquires long-range dependencies with a self-attention mechanism (SAM) to achieve multi-dimensional feature fusion for vessel traffic flow prediction. A case study is conducted on the narrow section of the Lüsi inbound waterway in Tongzhou Bay, China, to verify the effectiveness of the proposed model. The results demonstrate that the proposed method can realize high-precision probabilistic prediction of vessel traffic flow. Compared with the benchmark methods, the proposed model reduces both the Mean Absolute Scaled Error (MASE) and the Mean Absolute Error (MAE) simultaneously for medium- and long-term forecasting ( > 24 hours) under specific working conditions (VTS six-shift mode). This study can provide probabilistic decision support for intelligent traffic management in narrow navigable waters, and serve as an important reference for VTS scheduling as well as port and shipping operation dispatching.

Diffusion models (DMs) are advanced generative models, yet recent research reveals their vulnerability to backdoor attacks, which establish hidden associations between input patterns and targeted model behavior, potentially causing malicious outputs during inference. These attacks pose significant risks, including model owner reputation damage and harmful content generation. However, existing defense methods often fail against backdoor attacks on diffusion models. To address this gap, we propose Diff-Cleanse, a two-stage defense framework. The first stage introduces a novel trigger inversion method for backdoor detection, and the second stage applies a structural pruning-based method for backdoor removal. Experiments on 373 models poisoned by three state-of-the-art attacks show that Diff-Cleanse achieves > 97% detection accuracy, completely remove backdoors and maintains the models’ benign performance. Code is available at https://github.com/shymuel/diff-cleanse.

We propose an instance segmentation model, named CoMask, that effectively alleviates the scale variation issue and addresses the precise localization. Specifically, we develop a multi-scale feature extraction module (MSFEM) to exploit multi-scale spatial cues. Besides, the channel attention mechanism is also adopted to further enhance the discriminating ability. Equipped with MSFEMs, multi-scale and multi-level features can be extracted to better characterize objects of various sizes and provide affluent high-level semantic information. For precise localization, we propose a collaborative learning framework to compute coarse masks and regresses position-sensitive dense offsets. The foreground confidence of each position is then assigned as the weight of the corresponding bounding box to calculate a weighted average. Thus, we can mitigate interference of background regions. After obtaining the final regressed bounding boxes, finer foreground masks can be calculated. We conduct experiments on MS COCO dataset. Experimental results validate that CoMask is competitive compared with state-of-the-art models.

Research on designing local feature descriptors has gradually shifted to deep learning. Different from other computer vision tasks, the biggest challenge for local descriptor learning lies with the formulation of loss functions. Existing methods solve the problem by leveraging Siamese loss or triplet loss and improve the performance of the learned descriptors by a significant margin. However, the widely used Siamese loss and triplet loss cannot fully utilize the context information. In this paper, we propose a novel loss function to introduce more context information to facilitate training. After incorporating the proposed loss function into training, our learned descriptor demonstrates state-of-the-art performance in patch verification, image matching and patch retrieval benchmarks. The pretrained model will be publicly available at https://github.com/clelouch/CANet.