Semantic segmentation of remote sensing images has been widely used in environmental protection, geological disaster discovery, and natural resource assessment. With the rapid development of deep learning, convolutional neural networks (CNNs) have dominated semantic segmentation, relying on their powerful local information extraction capabilities. Due to the locality of convolution operation, it can be challenging to obtain global context information directly. However, Transformer has excellent potential in global information modeling. This paper proposes a new hybrid convolutional and Transformer semantic segmentation model called CTFuse, which uses a multi-scale convolutional attention module in the convolutional part. CTFuse is a serial structure composed of a CNN and a Transformer. It first uses convolution to extract small-size target information and then uses Transformer to embed large-size ground target information. Subsequently, we propose a spatial and channel attention module in convolution to enhance the representation ability for global information and local features. In addition, we also propose a spatial and channel attention module in Transformer to improve the ability to capture detailed information. Finally, compared to other models used in the experiments, our CTFuse achieves state-of-the-art results on the International Society of Photogrammetry and Remote Sensing (ISPRS) Vaihingen and ISPRS Potsdam datasets.

Landslide detection is crucial for natural disaster risk management. Deep-learning-based object-detection algorithms have been shown to be effective in landslide studies. However, advanced algorithms currently used for landslide detection require high computational complexity and memory requirements, limiting their practical applicability. In this study, we developed a high-resolution dataset for landslide-prone regions in China by extracting historical landslide remote sensing images from the Google Earth platform. We propose a lightweight LP-YOLO algorithm based on YOLOv5, with a more-lightweight backbone that incorporates our designed PartitionNet and neck equipped with CSPCrossStage. We constructed and added the vertical and horizontal (VH) block to the backbone, which explores and aggregates long-range information with two directions, while consuming a small amount of computational cost. A new feature fusion structure is proposed to boost information flow and enhance the location accuracy. To speed up the model learning process and improve the accuracy, the SCYLLA-IoU (SIoU) bounding box regression loss function was used to replace the complete IoU (CIoU) loss function. The experimental results demonstrated that our proposed model achieved the highest detection performance (53.7% of Precision, 49% of AP50 and 25.5% of AP50:95) with a speed of 74 fps. Compared to the YOLOv5 model, the proposed model achieved 4% improvement for Precision, 2.6% improvement for AP50, and 2.5% for AP50:95, while reducing the model parameters and FLOPs by 38.4% and 53.1%, respectively. The results indicated that the proposed lightweight method provides a technical guidance for achieving reliable and real-time automatic landslide detection and can be used for disaster prevention and mitigation.

Accurately forecasting demand for intermittent items is essential for effective inventory control, improved service levels, and cost reduction. This study focuses on highly sparse, irregular, and volatile demand patterns and proposes a generalizable multi-source data-driven framework for intermittent demand forecasting, using automotive spare parts as a representative application scenario. The proposed framework integrates Transformer networks, multi-graph convolutional networks (GCNs), and a Mamba-based feature fusion module. The Transformer captures long-term temporal dependencies in historical demand sequences, while the multi-graph GCN incorporates prior knowledge—including traffic geography, socioeconomic indicators, and environmental attributes—to model spatial correlations across multiple supply nodes. The Mamba-based fusion module then integrates temporal and spatial features into a unified representation, enhancing predictive accuracy and robustness. Extensive experiments on real-world datasets of automotive spare parts in China show that the proposed framework exhibits competitive and often superior performance compared with TiDE, FSNet, Informer, and DLinear across multiple forecasting horizons (3-, 6-, and 9-step), as measured by RMSE, MAE, and $R^2$. The proposed approach provides a practical and adaptable solution for forecasting intermittent demand, offering valuable support for dynamic inventory management.

In the past decade, the high computational overhead of asymmetric cryptography has remained a central challenge in end-to-end secure communication systems. To mitigate the performance bottlenecks inherent in the full SM2 encryption and decryption workflow, this paper introduces ParaSM2, a parallel restructuring optimization framework tailored for SM2-based cryptographic operations. ParaSM2 exploits the observed 2:1 processing ratio between KDF and HASH to perform cross-component parallel restructuring and applies fixed-prefix reuse together with dynamic task parallelism to eliminate 39.7% of redundant KDF computations. Furthermore, a vectorized reconstruction of the HASH message extension is incorporated to leverage SIMD parallel acceleration, thereby substantially enhancing throughput. Experimental evaluations against SM4-GCM and SM4-CBC on data blocks larger than 64 KB demonstrate that ParaSM2 achieves up to a 5.1× performance improvement on both x86 and ARM architectures, effectively reducing end-to-end latency and providing a scalable pathway for algorithmic optimization in cryptography across heterogeneous platforms.