Search Results (10)

Infrared small target detection (IRSTD) aims to segment small targets from an infrared clutter background. However, the long imaging distance, complex background, and extremely limited number of target pixels pose great challenges for IRSTD. In this paper, we propose a new IRSTD method based on the deep graph neural network to fully extract and fuse the texture and structural information of images. Firstly, a clustering algorithm is designed to divide the image into several subgraphs as a prior knowledge to guide the initialization of the graph structure of the infrared image, and the image texture features are integrated to graph construction. Then, a graph feature extraction module is designed, which guides nodes to interact with features within their subgraph via the adjacency matrix. Finally, a hierarchical graph texture feature fusion module is designed to concatenate and stack the structure and texture information at different levels to realize IRSTD. Extensive experiments have been conducted, and the experimental results demonstrate that the proposed method has high interaction over union (IoU) and probability of detection (P_d) on public datasets and the self-constructed dataset, indicating that it has fine shape segmentation and accurate positioning for infrared small targets. Full article

Knowledge base question answering (KBQA) aims to answer a question using a knowledge base (KB). However, a knowledge base is naturally incomplete, and it cannot cover all the knowledge needed to answer the question. Therefore, obtaining accurate and comprehensive answers to complex questions is difficult when KBs are missing relations and entities. To mitigate this challenge, we propose an incomplete KBQA approach based on Relation-Aware Interactive Network and Text Fusion (RAIN-TF). Specifically, we provide essential textual knowledge by finely filtering the question-related text to compensate for the missing relations and entities in the KB. We propose a question-related subgraph construction method that fuses the knowledge from the text and KB and enhances the interactions among questions, entities, and relations. On this basis, we propose a relation-aware interactive network, which is a relation-aware multi-head attention graph neural network (GNN) model, to promote the deep semantic integration of unstructured texts and structured KBs, thus effectively compensating for the lack of knowledge. Comprehensive experiments on three mainstream incomplete KBQA datasets verify the effectiveness of the proposed approach. Full article

Urban land use classification plays a significant role in urban studies and provides key guidance for urban development. However, existing methods predominantly rely on either raster structure deep features through convolutional neural networks (CNNs) or topological structure deep features through graph neural networks (GNNs), making it challenging to comprehensively capture the rich semantic information in remote sensing images. To address this limitation, we propose a novel urban land use classification model by integrating both raster and topological structure deep features to enhance the accuracy and robustness of the classification model. First, we divide the urban area into block units based on road network data and further subdivide these units using the fractal network evolution algorithm (FNEA). Next, the K-nearest neighbors (KNN) graph construction method with adaptive fusion coefficients is employed to generate both global and local graphs of the blocks and sub-units. The spectral features and subgraph features are then constructed, and a graph convolutional network (GCN) is utilized to extract the node relational features from both the global and local graphs, forming the topological structure deep features while aggregating local features into global ones. Subsequently, VGG-16 (Visual Geometry Group 16) is used to extract the image convolutional features of the block units, obtaining the raster structure deep features. Finally, the transformer is used to fuse both topological and raster structure deep features, and land use classification is completed using the softmax function. Experiments were conducted using high-resolution Google images and Open Street Map (OSM) data, with study areas on the third ring road of Shenyang and the fourth ring road of Chengdu. The results demonstrate that the proposed method improves the overall accuracy and Kappa coefficient by 9.32% and 0.17, respectively, compared to single deep learning models. Incorporating subgraph structure features further enhances the overall accuracy and Kappa by 1.13% and 0.1. The adaptive KNN graph construction method achieves accuracy comparable to that of the empirical threshold method. This study enables accurate large-scale urban land use classification with reduced manual intervention, improving urban planning efficiency. The experimental results verify the effectiveness of the proposed method, particularly in terms of classification accuracy and feature representation completeness. Full article

Driven by the rise of intelligent manufacturing and Industry 4.0, the manufacturing industry faces significant challenges in adapting to flexible and efficient production methods. This study presents an innovative approach to solving the Flexible Job-Shop Scheduling Problem (FJSP) by integrating Heterogeneous Graph Neural Networks based on Relation (HGNNR) with Deep Reinforcement Learning (DRL). The proposed framework models the complex relationships in FJSP using heterogeneous graphs, where operations and machines are represented as nodes, with directed and undirected arcs indicating dependencies and compatibilities. The HGNNR framework comprises four key components: relation-specific subgraph decomposition, data preprocessing, feature extraction through graph convolution, and cross-relation feature fusion using a multi-head attention mechanism. For decision-making, we employ the Proximal Policy Optimization (PPO) algorithm, which iteratively updates policies to maximize cumulative rewards through continuous interaction with the environment. Experimental results on four public benchmark datasets demonstrate that our proposed method outperforms four state-of-the-art DRL-based techniques and three common rule-based heuristic algorithms, achieving superior scheduling efficiency and generalization capabilities. This framework offers a robust and scalable solution for complex industrial scheduling problems, enhancing production efficiency and adaptability. Full article

Port state control (PSC) inspections, considered a crucial means of maritime safety supervision, are viewed by the industry as a critical line of defense ensuring the stability of the international supply chain. Due to the high level of globalization and strong regional characteristics of PSC inspections, improving the accuracy of these inspections and efficiently utilizing inspection resources have become urgent issues. The construction of a PSC inspection ontology model from top to bottom, coupled with the integration of multisource data from bottom to top, is proposed in this paper. The RoBERTa-wwm-ext model is adopted as the entity recognition model, while the XGBoost₄ model serves as the knowledge fusion model to establish the PSC inspection knowledge graph. Building upon an evolutionary game model of the PSC inspection knowledge graph, this study introduces an evolutionary game method to analyze the internal evolutionary dynamics of ship populations from a microscopic perspective. Through numerical simulations and standardization diffusion evolution simulations for ship support, the evolutionary impact of each parameter on the subgraph is examined. Subsequently, based on the results of the evolutionary game analysis, recommendations for PSC inspection auxiliary decision-making and related strategic suggestions are presented. The experimental results show that the RoBERTa-wwm-ext model and the XGBoost₄ model used in the PSC inspection knowledge graph achieve superior performance in both entity recognition and knowledge fusion tasks, with the model accuracies surpassing those of other compared models. In the knowledge graph-based PSC inspection evolutionary game, the reward and punishment conditions ($n,f$) can reduce the burden of the standardization cost for safeguarding the ship. A ship is more sensitive to changes in the detention rate $\beta$ than to changes in the inspection rate $\alpha$. To a certain extent, the detention cost $C_{DC}$ plays a role similar to that of the detention rate $\beta$. In small-scale networks, relevant parameters in the ship’s standardization game have a more pronounced effect, with detention cost $C_{DC}$ having a greater impact than standardization cost $C_S$ on ship strategy choice and scale-free network evolution. Based on the experimental results, PSC inspection strategies are suggested. These strategies provide port state authorities with auxiliary decision-making tools for PSC inspections, promote the informatization of maritime regulation, and offer new insights for the study of maritime traffic safety management and PSC inspections. Full article

With the development of the digital and intelligent transformation of the power grid, the structure and operation and maintenance technology of the power grid are constantly updated, which leads to problems such as difficulties in information acquisition and screening. Therefore, we propose a recommendation method for power knowledge retrieval based on a graph neural network (RPKR-GNN). The method first uses a graph neural network to learn the network structure information of the power fault knowledge graph and realize the deep semantic embedding of power entities and relations. After this, it fuses the power knowledge graph paths to mine the potential power entity relationships and completes the power fault knowledge graph through knowledge inference. At the same time, we combine the user retrieval behavior features for knowledge aggregation to form a personal subgraph, and we analyze the user retrieval subgraph by matching the similarity of retrieval keyword features. Finally, we form a fusion subgraph based on the subgraph topology and reorder the entities of the subgraph to generate a recommendation list for the target users for the prediction of user retrieval intention. Through experimental comparison with various classical models, the results show that the models have a certain generalization ability in knowledge inference. The method performs well in terms of the MR and Hit@10 indexes on each dataset, and the F1 value can reach 87.3 in the retrieval recommendation effect, which effectively enhances the automated operation and maintenance capability of the power system. Full article

Over the last decade, various deep neural network models have achieved great success in image recognition and classification tasks. The vast majority of high-performing deep neural network models have a huge number of parameters and often require sacrificing performance and accuracy when they are deployed on mobile devices with limited area and power consumption. To address this problem, we present an SSD-MobileNet-v1 acceleration method based on network compression and subgraph fusion for Field-Programmable Gate Arrays (FPGAs). Firstly, a regularized pruning algorithm based on sensitivity analysis and Filter Pruning via Geometric Median (FPGM) was proposed. Secondly, the Quantize Aware Training (QAT)-based network full quantization algorithm was designed. Finally, a strategy for computing subgraph fusion is proposed for FPGAs to achieve continuous scheduling of Programmable Logic (PL) operators. The experimental results show that using the proposed acceleration strategy can reduce the number of model parameters by a factor of 11 and increase the inference speed on the FPGA platform by a factor of 9–10. The acceleration algorithm is applicable to various mobile edge devices and can be applied to the real-time monitoring of forest fires to improve the intelligence of forest fire detection. Full article

Spatiotemporal graph convolution has made significant progress in skeleton-based action recognition in recent years. Most of the existing graph convolution methods take all the joints of the human skeleton as the overall modeling graph, ignoring the differences in the movement patterns of various parts of the human, and cannot well connect the relationship between the different parts of the human skeleton. To capture the unique features of different parts of human skeleton data and the correlation of different parts, we propose two new graph convolution methods: the whole graph convolution network (WGCN) and the part graph convolution network (PGCN). WGCN learns the whole scale skeleton spatiotemporal features according to the movement patterns and physical structure of the human skeleton. PGCN divides the human skeleton graph into several subgraphs to learn the part scale spatiotemporal features. Moreover, we propose an adaptive fusion module that combines the two features for multiple complementary adaptive fusion to obtain more effective skeleton features. By coupling these proposals, we build a whole and part adaptive fusion graph convolution neural network (WPGCN) that outperforms previous state-of-the-art methods on three large-scale datasets: NTU RGB+D 60, NTU RGB+D 120, and Kinetics Skeleton 400. Full article

To visually detect sea-surface targets, the objects of interest must be effectively and rapidly isolated from the background of sea-surface images. In contrast to traditional image detection methods, which employ a single visual feature, this paper proposes a significance detection algorithm based on the fusion of multi-visual features after detecting the sea-sky-lines. The gradient edges of the sea-surface images are enhanced using a Gaussian low-pass filter to eliminate the effect of the image gradients pertaining to the clouds, wave points, and illumination. The potential region and points of the sea-sky-line are identified. The sea-sky-line is fitted through polynomial iterations to obtain a sea-surface image containing the target object. The saliency subgraphs of the high and low frequency, gradient texture, luminance, and color antagonism features are fused to obtain an integrated saliency map of the sea-surface image. The saliency target area of the sea surface is segmented. The effectiveness of the proposed method was verified. The average detection rate and time for the sea-sky-line detection were 96.3% and 1.05 fps, respectively. The proposed method outperformed the existing saliency models on the marine obstacle detection dataset and Singapore maritime dataset, with mean absolute errors of 0.075 and 0.051, respectively. Full article

Surveillance videos contain a considerable amount of data, wherein interesting information to the user is sparsely distributed. Researchers construct video synopsis that contain key information extracted from a surveillance video for efficient browsing and analysis. Geospatial–temporal information of a surveillance video plays an important role in the efficient description of video content. Meanwhile, current approaches of video synopsis lack the introduction and analysis of geospatial-temporal information. Owing to the preceding problems mentioned, this paper proposes an approach called “surveillance video synopsis in GIS”. Based on an integration model of video moving objects and GIS, the virtual visual field and the expression model of the moving object are constructed by spatially locating and clustering the trajectory of the moving object. The subgraphs of the moving object are reconstructed frame by frame in a virtual scene. Results show that the approach described in this paper comprehensively analyzed and created fusion expression patterns between video dynamic information and geospatial–temporal information in GIS and reduced the playback time of video content. Full article