15 September 2026, Volume 52 Issue 9
-
Select all|
-
Computer Engineering. 2026, 52(9): 1-30. https://doi.org/10.19678/j.issn.1000-3428.0252895
As mega-constellations gradually become the core infrastructure of space—air—ground integrated networks, their resource scheduling faces multiple challenges, including high-dimensional constraints, dynamic task allocation, and multi-objective optimization. Intelligent scheduling methods can be broadly classified into three categories: model-driven methods, heuristic algorithms-based methods, and deep learning and Reinforcement Learning (RL) methods. Model-driven methods leverage tools, such as mixed-integer programming and graph-theoretical modeling, to construct optimization models and describe the constraints and objective functions of resource scheduling through precise mathematical formulations. These methods can provide theoretically optimal solutions in static scenarios. However, computational complexity increases exponentially as the problem size increases, posing difficulties for their application to large-scale dynamic scheduling. Heuristic algorithms-based methods, which are inspired by biological mechanisms, can rapidly generate approximate solutions and demonstrate high efficiency and flexibility in handling medium-scale problems. However, the quality of solutions is sensitive to parameter settings, and global optimality is not guaranteed. Deep learning and RL methods, which are driven by data and interactive learning mechanisms, can extract hidden patterns from massive scheduling datasets and continuously optimize decision strategies through agent-environment interactions. These approaches exhibit unique advantages in complex scenarios such as dynamic topologies and burst tasks. However, they are highly dependent on training data, and the interpretability of their decision-making processes remains limited. Research progress in areas such as cross-layer collaborative scheduling, robustness optimization, and heterogeneous resource integration remains limited. Future efforts need to further explore multimodal learning and adaptive decision-making mechanisms for driving mega-constellation resource scheduling toward greater intelligence, efficiency, and reliability and providing technological support for the large-scale deployment and application of space-air-ground integrated networks.
-
Computer Engineering. 2026, 52(9): 31-48. https://doi.org/10.19678/j.issn.1000-3428.0252663
The security of classical public-key cryptosystems, which operate based on the computational intractability of integer factorization and discrete logarithm problems, is profoundly threatened by the rapid development of quantum computing. Post-Quantum Cryptography (PQC) has been proposed to improve security against quantum attacks. Lattice-based cryptography is widely regarded as one of the most promising PQC method because of its proven security, efficient implementation, and functional versatility, and it has achieved significant progress in both algorithmic design and engineering implementation in recent years. This paper presents a systematic survey of the status of research on lattice-based cryptographic algorithms. First, the fundamental theory of lattices and their core hard problems are discussed. Next, the technical characteristics and performance of encryption and key encapsulation mechanisms, digital signature schemes, and key exchange protocols constructed based on lattice hard problems are reviewed and compared. Subsequently, the principles and optimization paths of two key technologies for lattice cryptography implementation, namely polynomial multiplication using the fast Number Theoretic Transform (NTT) and pre-image sampling algorithms, are discussed. Finally, considering the PQC standardization process of the National Institute of Standards and Technology (NIST), an outlook on future research directions for lattice-based cryptography is presented. The research directions include the development of more "succinct" fundamental hard problems, strengthening security proofs, improving algorithm efficiency, and countering novel attacks. This review serves as a reference for promoting the practical deployment and further research of lattice-based cryptography.
-
Computer Engineering. 2026, 52(9): 49-63. https://doi.org/10.19678/j.issn.1000-3428.0252962
The networked radar jamming resource allocation is a typical Non-deterministic Polynomial (NP) problem and a significant challenge, requiring the use of various optimization algorithms to solve it. To address the issues of low computational speed and poor adaptability in traditional jamming resource allocation optimization algorithms, advancements in intelligent optimization algorithms in this field are reviewed. First, a mathematical model and a solution framework for networked radar jamming resource allocation are constructed. The difficulties in solving this model are analyzed, and the advantages of intelligent optimization algorithms in terms of computational efficiency, global optimization capability, and robustness are emphasized. Subsequently, using the Genetic Algorithm (GA), Particle Swarm Optimization (PSO) algorithm, Ant Colony Algorithm (ACA), and various improved algorithms as typical examples, the implementation processes, solution effectiveness, and strengths and weaknesses of intelligent optimization algorithms in networked radar jamming resource allocation are analyzed in detail. Additionally, the application of fusion algorithms and other bionic/machine learning-based intelligent optimization algorithms in this field is summarized, and the advantages and disadvantages of various algorithms are compared and analyzed in terms of aspects such as adaptability, convergence, and global search capability, thus fully demonstrating the current development status of intelligent optimization algorithms. Finally, considering the multiple challenges currently encountered in networked radar jamming resource allocation, future development directions of intelligent optimization algorithms are proposed from four perspectives: algorithm comparison, optimization speed, fusion innovation, and dynamic adaptability. This study provides a valuable reference for the research and practical engineering of intelligent optimization algorithms in networked radar jamming resource allocation.
-
Computer Engineering. 2026, 52(9): 64-80. https://doi.org/10.19678/j.issn.1000-3428.0253388
In the digital era, complex interactions among modalities such as text, image, and audio have led to multimodal misinformation. The propagation speed and concealment levels of multimodal misinformation far exceed those of traditional unimodal misinformation, posing severe challenges to information security and social governance. However, research in this field is relatively scarce in China, and a comprehensive framework has yet to be established. Therefore, this paper provides a systematic overview of the research status and development trajectory of multimodal misinformation detection. Based on a clear understanding of the core concepts and task spectrum of multimodal misinformation detection, this review provides a detailed analysis of the characteristics of the datasets and evaluation metrics. It also analyzes the applicability and detection performance of different multimodal methods and models, such as SAFE, CAFE, CFFN, SSA-MFND, PSCC-Net, DGM4, CCN, SNIFFER, and KGAlign. Three core detection methods, namely cross-modal consistency, anomaly feature recognition, and external fact-driven approaches, are summarized. Furthermore, the interpretability and generalization robustness of multimodal misinformation detection are analyzed. With the rise of Large Vision-Language Models (LVLMs), their application to multimodal misinformation detection is continuously increasing. Various application scenarios, advantages, and limitations of LVLMs in this domain are discussed. Finally, future research directions in multimodal misinformation detection are outlined, aiming to provide insights and inspiration for further development in this field.
-
Computer Engineering. 2026, 52(9): 81-91. https://doi.org/10.19678/j.issn.1000-3428.0252092
The existing methods for tabular data prediction primarily focus on classical classification and regression tasks. However, a type of data in the tabular data domain contains labels with an ordinal relationship, and its prediction task is called tabular ordinal classification. Current methods for tabular ordinal classification rely primarily on retrieving similar features and augmenting sample feature representations by fusing features similar to the ordinal distance between classes. However, the existing methods neglect the full utilization of label ordinal knowledge. To address this, a method based on ordinal label entropy optimization is proposed, which effectively guides the model in learning ordinal information by mining the ordinal entropy embedded in the label order knowledge. First, an ordinal entropy calculation module is established that quantifies ordinal entropy based on the rank differences between the predicted and true labels. Through step-by-step analysis and derivation, the ordinal label entropy is designed as a novel rank loss function, which is introduced as a regularization term in the model. This encourages the model to learn the ordinal relationship between labels and reduces information loss caused by unordered predictions. The ordinal entropy optimized ranking loss function is combined with the original loss function of the model to improve its predictive ability. Finally, experiments on multiple ordinal tabular datasets reveal that this method outperforms various baseline models, demonstrating the effectiveness and advantages of the ordinal entropy optimization model in tabular ordinal classification tasks.
-
Computer Engineering. 2026, 52(9): 92-101. https://doi.org/10.19678/j.issn.1000-3428.0070496
In Named Entity Recognition (NER) methods, the nested semantic structure of named entities is effectively parsed by organizing all spans in a sentence into a planarized sentence representation. However, the spans in planarized sentence representation are abstract representations of phrases of different lengths in the sentence. Additionally, a span can represent either a single word or the entire sentence. This characteristic leads to two issues: insufficient contextual semantic dependency in short spans and loss of semantic information in the compression process for long spans. To address these issues, this study proposes a NER method based on semantic plane interpolation. This method expands the sentence plane representation by performing nearest neighbor interpolation on adjacent spans in the plane, which can significantly amplify the semantic features and contextual dependency between spans, effectively building up the semantic dependency between short spans and enhancing the semantic expression of long spans, thus better supporting named entities. Additionally, to avoid losing the original plane information, this study constructs a semantic perception network to extract and fuse global and local information in the original sentence plane representation, achieving fine-grained feature expression through multilayer refinement. The method achieves the highest F1 values of 88.87%, 88.52%, 81.55%, 96.49%, and 72.84% on the ACE2004, ACE2005, GENIA, Resume, and Weibo datasets, respectively, compared with the benchmark models, with improvements of 1.13, 1.61, 0.15, 0.37, and 0.46 percentage points, respectively. These results validate the generality and effectiveness of the proposed method for NER tasks.
-
Computer Engineering. 2026, 52(9): 102-112. https://doi.org/10.19678/j.issn.1000-3428.0252032
Existing time series forecasting methods fail to fully consider the spatiotemporal dependencies among variables, which hinders the improvement of forecasting accuracy. Spatial modeling methods based on Graph Neural Networks (GNN) also have limitations in graph structure construction. Specifically, static predefined graphs struggle to capture the dynamic interaction characteristics among variables, and adaptive graph structure learning is influenced by parameter initialization, rendering the method prone to being trapped in local optima. To address these issues, a multivariate time series forecasting method that combines spatiotemporal information and the Kolmogorov—Arnold Networks (KAN) is proposed. In the spatial dimension, a graph structure learning module is designed. It utilizes the Pearson correlation coefficient to establish the initial adjacency matrix of variables, introduces learnable parameters to dynamically adjust and optimize the graph structure, and effectively captures the spatial dependencies among variables by stacking multiple layers of graph convolutions. In the temporal dimension, the method combines the multi-head self-attention mechanism and Gated Recurrent Unit (GRU) to extract temporal dependencies in different subspaces, while capturing both local temporal patterns and global key information. To further enhance the representational capacity of the model, the KAN is used instead of the traditional Multi-Layer Perceptron (MLP), enabling nonlinear fusion of spatiotemporal features through learnable activation functions. Experimental results on seven benchmark datasets demonstrate that the proposed model achieves an average reduction of 36.9 percentage points in Mean Squared Error (MSE) and 24.8 percentage points in Mean Absolute Error (MAE). The generalization performance of the model is validated via testing using the Australian electricity load dataset. Compared with other mainstream models, this model can accurately capture the dependencies among variables and effectively fuse spatiotemporal features, thereby improving forecasting accuracy and robustness.
-
Computer Engineering. 2026, 52(9): 113-122. https://doi.org/10.19678/j.issn.1000-3428.0070677
Existing Knowledge Graph Completion (KGC) methods fail to effectively distinguish semantic differences among paths at different levels and cannot dynamically adjust relation representations by fully leveraging neighborhood context information. Consequently, contextual semantics remain insufficiently understood. To address these issues, this paper proposes a KGC model integrating relational context semantics, named RCSKGC, for enhancing the semantic representation of learned paths and neighborhood information. First, the local and global encoding of multi-hop paths at different levels is performed using a Bidirectional Gated Recurrent Unit (Bi-GRU) and an attention mechanism to extract effective path information. Meanwhile, the fine-grained semantic features of the path information at different levels are further enhanced through relational embedding contrastive learning. Second, a dual-attention mechanism and dynamic weighting strategy are employed to capture the neighborhood-level information of relations and learn their semantics through a "neighborhood—entity—relation" approach. Finally, the two types of relational representations are aggregated and used as inputs for the decoder to perform KGC. The results of experiments on the FB15k-237 dataset demonstrate that, compared to the best results of the baseline methods, RCSKGC achieves improvements of 1.4, 0.8, 1.3, and 2.1 percentage points in the Mean Reciprocal Rank (MRR), Hits@1, Hits@3, and Hits@10, respectively. Further, on the WN18RR dataset, RCSKGC shows improvements of 0.8 and 1.0 percentage points in MRR and Hits@3, respectively. These results validate the effectiveness of the proposed method.
-
Computer Engineering. 2026, 52(9): 123-132. https://doi.org/10.19678/j.issn.1000-3428.0070629
Entity Linking (EL), which links entity mentions in natural text to corresponding entities within a knowledge base, plays a crucial role in fields such as information retrieval and question-answering systems. The key challenge in EL lies in the effective utilization of the contextual information of mentions and feature information of entities in the knowledge base to generate candidate entities and select the correct entity among them. Although some methods generate relevant candidate entities by relying on specific strategies and select appropriate entities by combining feature information, they often fail to capture deeper semantic relationships, resulting in an inability to obtain high-quality candidate entities. In some cases, the correct entity may not be included among the candidate entities. Meanwhile, in specific domains, limited entity information may impair the capability of some methods for multi-level interaction. To address these issues, this paper proposes a two-stage EL method. This method first generates high-quality candidate entities and then aggregates entity feature information to achieve re-ranking at both coarse- and fine-grained levels. Specifically, it retrieves high-quality candidate entities using a contrastive learning method based on hybrid negative sample sampling. Subsequently, it predicts the fine-grained entity types in a weakly supervised manner and uses both coarse- and fine-grained type information to re-rank the candidate entities. Experimental results on three public datasets demonstrate the effectiveness of this method in enhancing EL performance.
-
Computer Engineering. 2026, 52(9): 133-142. https://doi.org/10.19678/j.issn.1000-3428.0070679
Personalized Federated Learning (pFL) algorithms have significant advantages in handling non-Independent and Identically Distributed (non-IID) datasets and enabling client-side model personalization. Hypernetwork-based pFL utilizes the client's own hypernetwork to achieve a personalized client model. However, the effect of sharing client-side hypernetwork parameters and client-side data on the accuracy of client-side personalized models remains unclear. A personalized Federated learning with Multi-layer Hypernetwork (pFedMHN) framework is proposed to optimize client models through the weighted aggregation of local and global hypernetworks. The server learns a global hypernetwork and each client's multi-layer local hypernetworks and then aggregates them. Clients use these aggregated hypernetwork parameters to iteratively update their models, resulting in more accurate personalized models. The experimental results show that on general public datasets, the pFedMHN outperforms the four benchmark algorithms in terms of accuracy, effectively solving the problems of data heterogeneity and model accuracy faced during personalized federated learning on non-IID datasets and achieving a more accurate personalized model for clients by utilizing hypernetwork parameters and client data sharing.
-
Computer Engineering. 2026, 52(9): 143-153. https://doi.org/10.19678/j.issn.1000-3428.0070449
In the modern industrial sector, the perception and analysis of text data are essential for promoting intelligent manufacturing and optimizing production processes. However, industrial text data are typically characterized by high specialization, diversity, complexity, and annotation costs, making traditional large-scale annotation methods unsuitable. Existing few-shot Named Entity Recognition (NER) methods often use prototypical networks to classify entities, where the prototype is the average of the features of all samples belonging to the same category. However, these methods are highly sensitive to the support set data and prone to sample-selection bias. To address this, a few-shot NER model based on Distribution Calibration called DC-NER is proposed. A two-stage approach for span detection and entity classification is adopted for the few-shot NER task. In the first stage, an entity extractor is used to select candidate entities. In the second stage, a type discriminator assigns these entities to predefined categories. To address the challenge of capturing the true class distribution with a small number of labeled samples in few-shot NER, which leads to inaccurate class prototype calculations, a method is proposed that leverages useful information from the source domain data to calibrate the class distribution in the target domain. Thus, more samples are generated based on the refined class distribution to construct more accurate prototypes, thereby improving the performance in few-shot NER tasks. Experiments on both the in-domain dataset Few-NERD and the cross-domain dataset Cross-NER reveal that DC-NER significantly outperforms the comparative models in terms of F1 value, validating its effectiveness for few-shot NER.
-
Computer Engineering. 2026, 52(9): 154-164. https://doi.org/10.19678/j.issn.1000-3428.0252119
Existing methods used for academic paper classification neglect relational information. To address this issue, this paper proposes a novel classification model that integrates a Graph Convolutional Network (GCN) with contrastive learning, called a Contrastive Graph Convolutional Network (CGCN). First, two distinct types of homogeneous-heterogeneous relational information are defined based on the content and citations of papers, and they are transformed into self-supervised information to construct contrastive loss. Second, contrastive loss is employed to enhance the feature extraction process of the GCN, such that homogeneous papers are extracted to be close to one another, while heterogeneous papers remain distant. Third, cross-entropy loss and the Softmax function are utilized to complete end-to-end academic paper classification. When tested on three benchmark academic datasets, the CGCN outperforms advanced baselines in the classification task. Its Micro-F1 and Macro-F1 scores are higher than those of the original GCN by 8.29 and 7.91 percentage points, respectively, on the Cora dataset. The CGCN enhances the capacity to represent potential information in papers by employing a contrastive loss based on a homogeneous—heterogeneous relationship, thereby improving the prediction accuracy and generalization. This method provides innovative ideas and methods for academic paper classification.
-
Computer Engineering. 2026, 52(9): 165-175. https://doi.org/10.19678/j.issn.1000-3428.0252022
Under complex sky backgrounds, the low local contrast between small targets and backgrounds reduces the accuracy of infrared intensity detection. Although infrared polarization imaging can effectively improve the local contrast between targets and backgrounds, it also raises the contrast of high-brightness background edges, bringing new challenges to target detection. To solve this problem, an aerial infrared small target detection method based on polarization feature reconstruction and local difference metric weighting is proposed in this paper. Firstly, considering the unique gradient characteristics of small targets, a global background gradient feature suppression matrix for Stokes vector components S1 and S2 is constructed by using image gradient vector information. Secondly, in view of the spatial correlation between S1 and S2, a polarization feature reconstruction method is established to suppress most background clutter while retaining target features. Meanwhile, according to the feature differences among targets, residual background and noise, an improved Variance Change Rate of Local Region (VSL) is proposed to quantify the complexity of local regions, so as to better suppress residual background. Comparative experiments with LCM, TLLCM, IPI, PSTNN and other algorithms are conducted in multiple scenarios. The experimental results show that the Signal-to-Clutter Ratio (SCR), Signal-to-Clutter Ratio Gain (SCRG) and Background Suppression Factor (BSF) of the proposed method are increased by 14.06%, 4.79% and 14.43% compared with the suboptimal algorithm, respectively. It can maintain excellent detection performance under various complex backgrounds and possesses strong robustness.
-
Computer Engineering. 2026, 52(9): 176-192. https://doi.org/10.19678/j.issn.1000-3428.0252073
Crowd escape behavior in public places is highly likely to cause serious public safety disasters. Traditional computer vision technology can detect a few characteristics of crowd escape behavior, but it is difficult to face complex dynamic visual scenes. To address this issue, based on the structural characteristics of the locust visual nerve, and leveraging the danger perception mechanism of the locust Lobula Giant Movement Detector (LGMD) as well as the mammalian retinal luminance adaptation mechanism, this paper proposes an Enhanced Crowd Escape Detection Neural Network (ECEDNN). First, the proposed neural network collects the luminance changes caused by crowd activities in the field of view. With the help of the mammalian retinal luminance adaptive mechanism, the visual response excitation is tuned to adapt to the lighting scene. Visual excitation and suppression are combined to filter background noise, and a center-surround mechanism is used to enhance motion edges. Finally, neural spike adaptive tuning is used to detect the burst escape behavior of the crowd and output strong membrane potential excitation. The experimental results show that ECEDNN can effectively detect and warn of crowd escape behavior in complex scenes, with an average accuracy of 98.90% on multiple video datasets. This work is involved the research of crowd activity detection inspired by biological visual perception mechanism, which can provide new ideas and methods for crowd behavior activity perception and anomaly detection in artificial intelligence.
-
Computer Engineering. 2026, 52(9): 193-203. https://doi.org/10.19678/j.issn.1000-3428.0070764
Biomedical imaging is crucial in the diagnosis and treatment of various diseases. The application of deep learning methods to medical image analysis can enhance the readability of medical images and provide more reliable support for clinical decision-making. However, traditional medical image processing methods are limited in effectively capturing spatial features and complex structural information in 3D images, especially when handling complex 3D medical images generated by different imaging modalities. This often challenges the accuracy and generalization ability of the model. To address this challenge, an MTM3D model is proposed for medical image classification tasks. This model combines the excellent performance of the Mamba model in complex sequential tasks with the external memory storage function of the improved Token Turning Machine (TTM) network. By introducing a cyclic chain storage structure, MTM3D enables effective interaction of features from different spatial structures within memory units, thus enhancing its ability to capture complex spatial relationships. Furthermore, the incorporation of Mamba further strengthens the interaction between the memory and processing units, strengthening the generalization capability of the model and enhancing its performance across different medical imaging datasets. Experimental results on the MedMNIST v2 dataset demonstrate that MTM3D exhibits outstanding capabilities in understanding medical images. Compared with the current best medical image analysis networks, MTM3D improves the average Accuracy (ACC) by 3.97% and the average Area Under the Curve (AUC) by 2.00%, fully showcasing its tremendous potential in interpreting medical images and assisting healthcare professionals in diagnosis and treatment planning.
-
Computer Engineering. 2026, 52(9): 204-217. https://doi.org/10.19678/j.issn.1000-3428.0070406
Previous studies on cross-view geolocalization have primarily focused on determining whether a query image accurately corresponds to a specific geographic location within a predefined gallery. However, this research paradigm often overlooks the extensive multiscale structural information inherent in the geographic space. To achieve more robust localization, a model must not only capture local architectural details but also understand the spatial relationships among targets reflected through building clusters and environmental features, thereby improving the localization accuracy across different spatial scales. To address these challenges, a multiscale cross-view geo-localization task is proposed and a Multi-Level Campus (ML-Campus) dataset is constructed specifically for this task. The ML-Campus dataset comprises multiview, multisource building images, each annotated with multiscale labels to reflect correlations and continuity across different spatial scales. Based on this dataset, an empirical evaluation of existing cross-view geo-localization methods is conducted, which is used as a benchmark to measure their performance in this context. To further enhance model performance, the proposed Cross-View HAPPIER (CV-HAPPIER) method is employed for training, which strengthens the model's feature representation capabilities across different spatial scales. Extensive experimental results on the ML-Campus dataset demonstrate that the CV-HAPPIER method significantly improves the spatial robustness of cross-view geo-localization retrieval ranking results.
-
Computer Engineering. 2026, 52(9): 218-227. https://doi.org/10.19678/j.issn.1000-3428.0070529
To address the problems of insufficient stability in complex network construction and poor classification performance of shape features extracted in complex situations in existing complex network-based image shape classification algorithms, this paper proposes an image shape classification algorithm based on persistent homology and complex networks. The algorithm combines complex networks with Vietoris-Rips filtration to construct a persistent complex network on the image contour point cloud. Persistent homology is used to compute global topological features of different dimensions within the persistent complex network. Degree features and correlation degree features extracted from the degree distribution are selected as local shape features, and are fused with global topological features respectively to obtain two feature sets, PHCND and PHCNJD, enabling a richer shape representation of the images. The fused feature vectors are classified using Linear Discriminant Analysis (LDA). Comparative experiments are conducted with other traditional algorithms as well as a deep learning algorithm ResNet-50 on nine public image datasets. Meanwhile, ablation studies are designed to validate the effectiveness of the global topological features and the complementarity between local shape features and global topological features within the persistent complex network. Experimental results show that the proposed algorithm achieves the highest accuracy and F1-score on five public datasets, with improvements of 2.2%—30.3% in classification accuracy and 2.2%—30.9% in F1-score compared to seven traditional image shape classification algorithms. These results validate the effectiveness of the proposed algorithm for image shape classification, and demonstrate that the extracted shape features exhibit certain robustness.
-
Computer Engineering. 2026, 52(9): 228-241. https://doi.org/10.19678/j.issn.1000-3428.0253485
Low-Light Image Enhancement (LLIE) is crucial in computer vision by restoring rich visual information from corrupted low-light images. However, existing LLIE methods often suffer from color bias due to color space and typically fail to balance denoising and color fidelity within a single-stage framework. To address these challenges, this study introduces a novel Dual-stage HVI-based Transformer Network (DHTNet) for LLIE. DHTNet significantly improves the quality of low-light images by decoupling I and HV maps, enabling their independent yet synergistic optimization within the HVI color space. In the first stage, a hierarchical Transformer network equipped with an Adaptive Guidance Interaction Module (AGIM) models long-range dependencies between I and HV features to achieve global noise suppression and accurate color calibration. In the second stage, the Multi-Scale Enhanced Synergistic Attention (MESA) module enhances localized color and detail representation through synergistic optimization across the I and HV branches. This dual-stage framework effectively addresses the limitations of existing LLIE methods by retaining complex image details while enhancing visual realism. Experimental results show that DHTNet achieves the highest Peak Signal-to-Noise Ratio (PSNR) on both the SICE and SID datasets, outperforming the second-best model by 0.717 dB and 1.897 dB, respectively. Furthermore, DHTNet attains PSNR values of 28.756 dB, 24.683 dB, and 25.950 dB on the LOLv1, LOLv2-Real, and LOLv2-Synthetic datasets, respectively, consistently outperforming existing models such as Retinexformer and CIDNet.
-
Computer Engineering. 2026, 52(9): 242-252. https://doi.org/10.19678/j.issn.1000-3428.0252122
To address the key issues in image stitching caused by wide-parallax, such as geometric distortion, ghosting artifacts, and visible seams, a seamless stitching algorithm with strong structure preservation for wide-parallax images is proposed. This algorithm achieves full-process structure preservation, from optimized registration to precise fusion, by constructing a multidimensional constraint model. First, during the registration phase, a scene-guided adaptive mesh deformation model is established, and dynamic weighted vectors are constructed based on the feature distribution and proportion of overlapping areas, significantly improving the spatial registration accuracy. Second, a dual-edge detection mechanism combining strong and weak detection is proposed to capture residual structural misalignment, and a novel significant texture measurement model is designed using feature-coordinate convolution, enabling precise perception of significant structures in the overlapping region. Finally, in the fusion phase, an optimal seam cost function is constructed considering color difference, structural difference, and significant texture degree as smoothness constraints, yielding a seam mask that balances structural integrity and naturalness of the image. Experimental results demonstrate that this algorithm effectively solve the problem of structural distortion in wide-parallax stitched images by generating high-definition and visually coherent panoramic images with a large viewing angle. Compared with six advanced image stitching algorithms, this algorithm improves the Peak Signal-to-Noise Ratio (PSNR) and Structural Similarity Index Measure (SSIM) by at least 8.14% and 10.44%, respectively, verifying its technical advantages.
-
Computer Engineering. 2026, 52(9): 253-264. https://doi.org/10.19678/j.issn.1000-3428.0070685
Murals, an important part of cultural heritage, have received widespread attention in recent years for their digital protection and restoration. However, Super-Resolution (SR) reconstruction of mural images often faces challenges such as texture blurring and loss of original information. To address these issues, this study proposes a Reference-based Two-stage Mural Image Super-Resolution Reconstruction (RTMISR) method. First, a Multi-Scale Residual Feature Extraction Module (MRFEM) is employed to accurately capture the feature relationships between High-Resolution (HR) and Low-Resolution (LR) mural images, ensuring complete retention of LR image information and achieving an initial reconstruction of mural contours and partial details. Subsequently, a Mural Texture Feature Enhancement Module (MTFEM) is introduced, which utilizes a coarse-to-fine feature matching method to extract high-quality texture information from reference images and effectively integrate it into the reconstructed images to enhance texture detail representation. Moreover, to ensure the relevance and quality of the reference images, a Reference Image Filtering Module (RIFM) is designed to select reference images that are highly correlated with the target LR images. Experimental results on mural datasets show that, compared to representative SR methods such as SRGAN, MADNet, and ESRT, RTMISR achieves superior performance in objective metrics: for 2× SR, Peak Signal-to-Noise Ratio (PSNR) improves by an average of 2.83 dB and Structural Similarity Index Measure (SSIM) by 0.04; for 4× SR, PSNR improves by an average of 2.00 dB and SSIM by 0.02. In terms of subjective visual quality, RTMISR effectively retains the original information while enhancing the texture details of mural images, achieving a better balance between model complexity and reconstruction performance.
-
Computer Engineering. 2026, 52(9): 265-281. https://doi.org/10.19678/j.issn.1000-3428.0260093
The demand for continuous time-series data sharing in Internet of Things (IoT) scenarios has been increasing. Traditional Certificateless Signcryption (CLSC) schemes based on bilinear pairings or elliptic curves offer limited security under quantum computing threats and incur high computational and communication overheads when repeatedly signcrypting multiple data segments. To address these issues, this paper proposes a post-quantum lightweight certificateless hybrid signcryption scheme tailored for time-series data sharing. First, a bidirectional hash chain-driven symmetric key generation and management mechanism is designed to enable rapid key generation for multiple consecutive data segments while ensuring forward and backward security. Subsequently, a lightweight hybrid signcryption scheme with conditional privacy protection is constructed using a module lattice framework. Combined with an offline/online separation that shifts part of the computation ahead of time, the scheme achieves post-quantum security while significantly reducing the burden on end devices and communication links. The scheme is proven to satisfy IND-CCA confidentiality and EUF-CMA unforgeability in a random oracle model. Prototype implementation and evaluation on resource-constrained terminal devices show that, the proposed scheme achieves significant advantages: on average, the end-side overall latency and transmission volume can be reduced to approximately 40% of those of comparable post-quantum schemes. The advantages of this scheme become even more pronounced in long-duration continuous time-series data scenarios. It effectively balances post-quantum security, lightweight design, and scalability, making it well-suited for time-series data sharing applications in next-generation intelligent IoT.
-
Computer Engineering. 2026, 52(9): 282-292. https://doi.org/10.19678/j.issn.1000-3428.0070637
Nowadays, open source software is widely used in various industries, particularly in important fields such as aerospace and automotive electronics. However, most open source software has security vulnerabilities. In China, the component analysis and verification of open source software are seriously missing in software development and evaluation, which makes it difficult to ensure the safety of software in critical areas. Therefore, Software Component Analysis (SCA) is indispensable to ensure software security, and accurate identification of Third-Party Dependencies (TPD) is the key to software vulnerability management and compliance assessment. To solve these problems, this paper proposes a lightweight dependency analysis method for SCA, which improves the accuracy of TPD identification and efficiency of large-scale project file processing. The main contents are as follows: first, the method includes an analysis algorithm for the Java Maven project, which constructs the project structure model and extracts TPD information by identifying the construction configuration file of the project; second, the method includes a redundant dependency detection algorithm based on the Winnowing algorithm. The algorithm detects the actual use of TPD and eliminates redundant dependencies by comparing the code file and hash fingerprint that identifies TPD information step by step; third, based on the proposed algorithm, a lightweight component analysis framework is designed and implemented. The framework wraps the analysis algorithm through a specific analyzer class and then registers and executes the analysis task using the ServiceLoader API in Java. To verify the effectiveness of the method, database containing 56 870 different versions of TPD is built, and four real open source projects from GitHub for experimental verification are collected. The results show that the proposed algorithm performs well in terms of detection accuracy. Compared with the clustering algorithm based on machine learning and technology based on code similarity comparison, the proposed algorithm has higher accuracy, F1 value, and lower detection time. In addition, the ServiceLoader API applied to the system makes it more extensible, convenient for adding different analysis algorithms, and has strong practicability, laying the foundation for the subsequent implementation of multilingual TPD detection.
-
Computer Engineering. 2026, 52(9): 293-305. https://doi.org/10.19678/j.issn.1000-3428.0070741
The Hypertext Transfer Protocol (HTTP), the core infrastructure of Internet communication, relies on the collaboration of multiple servers in its modern communication model. If the servers in the processing chain do not strictly follow the protocol specifications or have differences in semantic interpretation, systemic semantic inconsistency problems will occur, leading to security threats such as access control strategy bypass, multi-host issues, request smuggling, and cache pollution. Differential fuzz testing analyzes semantic inconsistency problems by observing the differences in the processing results of messages are processed across different servers. To solve the problems of an inaccurate field set range, low mutation efficiency, and single observation dimension in existing tools, this paper proposes an improved differential fuzz testing method. First, a message construction method based on key headers is used to select core fields and streamline the test space. Second, a mutation method based on field semantics is adopted, combining semantic classification and vulnerability characteristics to design mutation operations, thereby enriching the test cases. Third, an extended message analysis method is employed to expand the scope of message analysis to both request and response messages, fully observing the communication process and covering existing scenarios of semantic inconsistency problems. Finally, tests are conducted on 7 commonly used servers, identifying 18 types of server processing differences and verifying 9 pairs of combinations that exhibit semantic inconsistency problems. Compared with similar tools such as t-reqs, the proposed method reduces the size of the test set by an order of magnitude, increases the average proportion of valid test cases by 12.67 percentage points, discovers two additional types of difference problems under the same observation conditions, and expands the test scope to cover all four scenarios of current semantic inconsistency problems.
-
Computer Engineering. 2026, 52(9): 306-319. https://doi.org/10.19678/j.issn.1000-3428.0252239
Conventional clean label backdoor attacks often fail to establish a strong link between the trigger and target classes, resulting in a low Attack Success Rate (ASR). Extensive experimental evidence shows that this failure is even more severe in Federated Learning (FL). The main reason for attack failure is that the random selection of the trigger causes it to lack a direct connection with the target class. To this end, a learnable trigger backdoor attack is designed for FL, which fully utilized the task information and shared model issued by the central server to train a trigger that is strongly correlated with the target class. This training process is formalized into a dual-objective optimization problem and solved. The optimal perturbation under constraint conditions is found to blur the original features of the image as much as possible, thereby maximizing the model's learning ability for the trigger. Subsequently, these blurred images are trained by adding the triggers allowed within the specified range as inputs, minimizing their image classification loss and generating the optimal trigger quickly using the optimization method of small-batch Projected Gradient Descent (PGD). The backdoor attack activated by this trigger guarantees excellent attack performance in FL. The experimental results on three datasets show that the ASR of the proposed method in FL is much higher than that of all types of existing clean-label backdoor attacks, especially on CIFAR-10, which shows an improvement of more than 82 percentage points compared to the baseline method. The proposed attack method presents new challenges for the security of FL.
-
Computer Engineering. 2026, 52(9): 320-332. https://doi.org/10.19678/j.issn.1000-3428.0252251
With the development of multimedia technology, the difficulty of unauthorized forgery and dissemination of false information has significantly decreased. This can lead to several negative consequences. Effective content authentication algorithms are urgently required to ensure the authenticity and security of image content. Recently, perceptual image hashing has demonstrated excellent performance in the field of image authentication. However, existing algorithms are not ideal for processing images with a large proportion of text, and they cannot effectively cope with new content-preserving manipulations such as scribbling. Therefore, a Text—Picture Mixed Image (TPMI) content authentication algorithm based on perceptual hashing is proposed. The proposed algorithm adopts the image segmentation algorithm of ring partitioning and calculates the frequency and distribution characteristics of the Scale-Invariant Feature Transform (SIFT) keypoints within each ring. These features exhibit rotational invariance and effectively improve the anti-collision performance of the proposed algorithm. By obtaining keypoint information, the proposed algorithm performs well in terms of robustness against content-preservation manipulations, including irregular scribbles. A TPMI dataset is constructed to validate the performance of the proposed algorithm. Compared to some representative algorithms, this algorithm performs better in terms of perceptual robustness, anti-collision, and security. Partial tampering with images can effectively identify tampered images that are similar to the original image. Additionally, experiments on scribble attacks are conducted, and the results show that the proposed algorithm can effectively identify such attack images.
-
Computer Engineering. 2026, 52(9): 333-347. https://doi.org/10.19678/j.issn.1000-3428.0252041
This study proposes TGMM, a large-scale code clone detection method based on parse trees and Graphics Processing Unit (GPU) acceleration, to address the limitations of existing tools in multilingual adaptation and large-scale analysis. The method employs a three-stage architecture for clone detection. First, it generates standardized parse trees based on lexical and syntactic rules of each programming language and then extracts subtrees that meet specific granularity requirements. Second, it simplifies these subtrees by pruning and eliminates nonfunctional differences through semantic equivalence transformations. Finally, it leverages GPUs to construct a global suffix array in parallel, enabling rapid similarity calculation of large-scale code blocks. The proposed method is evaluated in terms of clone detection efficiency and language scalability. On the public benchmark dataset BigCloneBench, TGMM achieves a precision of 97%, significantly outperforming seven mainstream tools compared, while reducing execution time by over 50% compared to the second-best tool and maintaining a recall rate similar to the compared tools across various clone types. In the language scalability test, TGMM successfully parses 19 out of 22 mainstream programming languages. Furthermore, by applying TGMM to conduct a multi-granularity clone analysis of the top 45 GitHub projects (covering nine programming languages), the study reveals significant differences in clone density across different languages, provides an in-depth analysis of the underlying causes, and thus offers practical and effective references for software maintenance.
-
Computer Engineering. 2026, 52(9): 348-364. https://doi.org/10.19678/j.issn.1000-3428.0252080
Bi-triangle enumeration in bipartite graphs is a core operation for graph analysis tasks such as local Clustering Coefficient (CC) computation. As the scale of real-world bipartite graph data continues to increase, the data volume has exceeded the processing capacity of a single machine, necessitating efficient enumeration through distributed algorithms. However, existing distributed graph partitioning algorithm, called GP, suffer from issues such as large subgraph combination data volume, message overload, and redundant enumeration. To address these problems, two optimization algorithms based on the topological characteristics of bi-triangles are proposed and the graph partitioning strategies are customized accordingly. Method 1 treats a bi-triangle as composed of three wedge structures and generates subgraphs using wedge groups as the basic unit. It introduces a subgraph combination construction mechanism based on the concatenation of A-type and V-type wedge groups, significantly reducing the number and data size of subgraph combinations, and ultimately enumerates bi-triangles via wedge triplets. Additionally, to address message overload and redundant enumeration, Method 1 proposes a subgraph-reading mechanism based on a distributed storage system and a deduplication mechanism based on vertex ordering. Method 2 considers a bi-triangle as composed of two zedge structures. It first performs initial graph partitioning using wedge groups as the basic unit, then completes the second partitioning through a construction and restoration mechanism of "compressed zedge groups", ultimately enumerating bi-triangles via zedge pairs, achieving lower computational complexity than Method 1. Experimental results demonstrate that, compared with the GP algorithm, Method 1 processes only 0.5% of the subgraph combination data on average and the enumeration time is less than 2.2% of the original, whereas Method 2 processes approximately 3.3% of the data on average and the time is compressed to within 1.0% of the original.
-
Computer Engineering. 2026, 52(9): 365-375. https://doi.org/10.19678/j.issn.1000-3428.0260413
This paper studies operator-level reconstruction for the structural mismatch between YOLOv11 and Ascend Neural Processing Unit (NPU). The Spatial Pyramid Pooling-Fast (SPPF), C3K2, and C2PSA modules are optimized without changing network semantics or model scale. Three Ascend C operators are designed: the SPPF operator uses on-chip data loop and halo cache to reduce redundant global-memory traffic in multi-stage pooling; the C3K2 operator integrates multi-core task assignment and multi-queue asynchronous pipelining to reduce fine-grained kernel launch overhead; and the C2PSA operator reconstructs attention communication through a parallel reduction-broadcast primitive. On an Ascend 910B NPU, the complete reconstruction reduces the training time per epoch by 23.2% and improves the training throughput by 27.6% on the COCO dataset. The results show that matching Ascend on-chip memory, asynchronous queues, and multi-core synchronization mechanisms improves the training execution efficiency of key YOLOv11 modules and keeps inference performance stable. It can provide verifiable operator mapping schemes for the deployment of complex object detection networks on the Ascend platform.
-
Computer Engineering. 2026, 52(9): 376-387. https://doi.org/10.19678/j.issn.1000-3428.0070438
The efficient utilization of resources across nodes in a heterogeneous cluster is pivotal for enhancing the overall resource utilization of the cluster. One of the primary functions of Kubernetes, the preferred system for container orchestration, is to schedule pods onto appropriate nodes. A well-designed scheduling strategy can reduce resource fragmentation, enhance the utilization of node resources, and minimize job waiting times. In the context of resource scheduling within heterogeneous clusters, the default Central Processing Unit (CPU)- and memory scheduling algorithms of Kubernetes have become inadequate for addressing the demands of multi-dimensional resource job scheduling. To address this challenge, a multi-criteria scheduling algorithm integrating an Analytic Hierarchy Process (AHP) and binpacking scheduling concepts is proposed. This algorithm considers not only traditional CPU and memory resource indicators but also incorporates various other resource indicators such as storage, GPU-like accelerator cards, and images, enabling a more comprehensive resource assessment. It also combines subjective user decisions with quantified objective node resource constraints to select the most suitable node from multiple candidates for job scheduling. A custom scheduler is designed based on the Kubernetes scheduling framework. The use of a custom scheduler with a multi-criteria scheduling algorithm can more efficiently utilize node resources and reduce the job waiting time caused by resource fragmentation. Experiments on mixed deployments of different types of jobs in heterogeneous clusters demonstrate that the multi-criteria scheduling algorithm can effectively increase cluster resource utilization and reduce Pod scheduling wait times by 49.79%. This provides an important reference for optimizing scheduling strategies for future large-scale cloud platforms and data centers.
-
Computer Engineering. 2026, 52(9): 388-401. https://doi.org/10.19678/j.issn.1000-3428.0252206
In distributed computing frameworks, inefficient data transfer in the Shuffle phase has become a key bottleneck in data connectivity. Existing methods have certain limitations in dealing with table joins, such as broadcast joins and Hash joins in Spark are both susceptible to data skewing, which makes the load between nodes unbalanced. Aiming at this problem, the paper focuses on joining aggregated queries, and proposes a table joining method based on lattice structure: by precomputing the storage table partition data in the form of lattice structure, and utilizing the convex set property of equivalence class, i.e., the data cells containing the upper bound of equivalence class and contained by the lower bound of equivalence class, whose aggregation values are equal to the aggregation values of equivalence class, so as to realize the quick matching and calculation. Since the query data cells as a compressed form of basic table data, the data size and skew are more concise and uniform, the article uses the query data cells instead of table data to perform data transfer and connection, which greatly reduces the data Shuffle and computational complexity. The method proposed in the paper has been implemented in Spark, and experiments based on the TPC-H dataset show that: the method of the paper reduces the data Shuffle by about 45.06% in large dataset scenarios, meanwhile, the workload among the nodes is more balanced compared to the benchmark method, and the query response time is shortened by 14.23% on average.
-
Computer Engineering. 2026, 52(9): 402-412. https://doi.org/10.19678/j.issn.1000-3428.0070608
With the frequent occurrence of terrorist attacks, crowd evacuation path planning in indoor public places has received increasing attention. This paper proposes a path planning method based on the Proximal Policy Optimization (PPO) algorithm to improve the efficiency and safety of pedestrian evacuation. The indoor terrorist attack scenario is described, and static obstacles, idle locations, dynamic obstacles, exits, and pedestrians in indoor public places are modeled using a cellular automata model. Subsequently, a feature construction method based on distance information is proposed to construct pedestrian features that represent the escape difficulty of evacuation paths, including shortest path features and safe path features, by combining the distance from pedestrians to exits in non-threatening environments and threat-facing scenarios. Finally, by describing the evacuation path planning problem as a reinforcement learning problem, a reward function based on evacuation efficiency, death penalty and successful escape reward is designed. Based on the feedback from the real-time environment, an evacuation strategy is provided to pedestrians, which in turn realizes the overall optimization of the escape path by the PPO algorithm. Compared with existing field methods, the proposed method can improve the efficiency and safety of crowd evacuation in different simulation scenarios, particularly in complex and high-density environments. The effectiveness of the shortest and safe path features is verified through ablation experiments.
-
Computer Engineering. 2026, 52(9): 413-423. https://doi.org/10.19678/j.issn.1000-3428.0252052
Bayesian Optimization (BO) is a powerful tool for sequential optimal design in formulation development. However, conventional BO is sensitive to initial samples, easily falls into a local optimum, and cannot dynamically adapt to changes in the process easily. By combining the global search capability of Reinforcement Learning (RL) with the efficient exploration strategy of BO, this study proposes a sequential formulation optimization design method based on a Deep Deterministic Policy Gradient (DDPG) to achieve adaptive formulation optimization. By leveraging real-time feedback from the data, this approach dynamically adjusts the optimization strategy, enabling efficient global optimization starting from random initial points. It effectively addresses challenges such as limited data, inaccuracies in surrogate models, and measurement noise. During the optimization process, the method employs a Gaussian Process (GP) to construct a surrogate model of the objective function and adopts an inner-outer dual-loop framework to enhance optimization efficiency. In the inner loop, the actor network generates candidate actions, and the Upper Confidence Bound (UCB) strategy is used to select optimal experimental points. In the outer loop, the experimental strategy is dynamically adjusted based on historical experience, and the surrogate model is updated using new observations. The results from simulation cases and wine blending experiments demonstrate that the proposed method outperforms traditional optimization methods in terms of optimization efficiency, convergence speed, and quality of the final solution. Moreover, it exhibits a strong generalization ability in new environments, providing a novel solution for complex formulation optimization problems.
-
Computer Engineering. 2026, 52(9): 424-435. https://doi.org/10.19678/j.issn.1000-3428.0252768
Redundant dependencies in software projects can increase build size, performance overhead, and long-term maintenance burden. Although existing studies have investigated redundant dependencies within the Maven ecosystem, analyses of their distribution across different dependency scopes (e.g., compile and test), evolutionary patterns, and impact on project popularity are lacking. To address this gap, this study selects 2 214 Java Maven open-source projects from GitHub as research subjects. It employs a mvn command to identify dependencies that are declared but not actually used and quantitatively analyzes redundancy ratios based on their respective scopes. Furthermore, it applies the Mann-Kendall non-parametric trend test on 3 817 historical versions from 698 projects to identify evolutionary trends in redundant dependencies. To assess the relationship between redundant dependencies and project popularity or community activity, this study constructs five GitHub-based popularity and activity metrics, including star growth, fork growth, and Issue closing rates, and performs a Pearson correlation analysis. Experimental results show that redundant dependencies are primarily concentrated in the compile and test scopes, with median redundancy ratios of 33.33% and 30%, respectively. In terms of evolutionary trends, 48.1% of the projects maintain a stable redundancy ratio and 36.2% exhibit fluctuations, while a small proportion shows an increasing or decreasing trend. In the correlation analysis, only the issue closing rate exhibits a significantly weak negative correlation with the redundancy ratio. These findings provide developers with a detailed perspective on dependency management and can aid in optimizing project configurations and improving software maintainability.
-
Computer Engineering. 2026, 52(9): 436-448. https://doi.org/10.19678/j.issn.1000-3428.0070417
Electric power load forecasting is a crucial step in optimal dispatching of power grids. However, when confronted with complex dynamic environments and incompletely stationary load characteristics, existing purely data-driven methods and strategies incorporating domain knowledge insufficiently capture long-term trends and periodic patterns, which affects forecasting accuracy and robustness. To address this issue, an electric power load forecasting model based on the integration of domain prior knowledge, named DPK-ELF, is proposed. This model utilizes a prior knowledge extraction module to thoroughly analyze the dynamic behavioral characteristics of time-series data, constructs domain prior knowledge specific to the given data, and employs a dynamic piecewise stacking average smoothing method to extract prior trends in electric power loads. The prior-trend decomposition module decomposes the electric power load series into prior smoothed trends and residual local random fluctuations, which are then combined with the PatchTST data-driven model for forecasting. Soft constraint optimization techniques are adopted during the model training phase using domain prior knowledge as boundary constraints in the loss function to enhance the robustness of the model. Experiments on four publicly available electric power load datasets reveal that DPK-ELF outperforms comparison models such as PatchTST, DLinear, Autoformer, and Informer across three key performance metrics: Mean Squared Error (MSE), Mean Absolute Error (MAE), and Relative Squared Error (RSE). In the Australian electricity price and power load dataset, compared with the PatchTST model, DPK-ELF achieves improvements of 28.31%, 19.57%, and 14.94% in MSE, MAE, and RSE, respectively. In the PDB power demand dataset, compared with the PatchTST model, DPK-ELF achieves improvements of 12.25%, 7.77%, and 6.29% in MSE, MAE, and RSE, respectively. These results demonstrate the significant advantages of the DPK-ELF model in improving forecasting accuracy.
-
Computer Engineering. 2026, 52(9): 449-456. https://doi.org/10.19678/j.issn.1000-3428.0070580
Dissolved Gas Analysis (DGA) aims to identify potential fault types by monitoring the dissolved gases in insulating oil. However, existing DGA methods exhibit limited performance because of the constraints imposed by the scarcity of labeled data. To address this issue, a novel Graph Knowledge Distillation method (GKDG) is proposed to enhance the accuracy and efficiency of the DGA. It employs a dual-view graph construction strategy to obtain additional supervision from sample neighborhoods, aggregating information directly from other samples through propagation. Furthermore, knowledge from the teacher Graph Neural Network (GNN) is distilled into the student GNN model, ensuring that the student model can effectively capture and interpret the complex relationships among the dissolved gases. Additionally, to align the student and teacher graphs in the embedding space, multiple types of knowledge are introduced, thereby enhancing the learning capability of the student model and enabling it to learn better from the teacher model. The experimental results validate the effectiveness of the GKDG in improving the DGA performance, providing strong support for the maintenance and fault detection of power equipment.
-
Computer Engineering. 2026, 52(9): 457-468. https://doi.org/10.19678/j.issn.1000-3428.0252025
Annotation of diagenetic facies samples is a crucial step in ensuring the accuracy of intelligent diagenetic facies recognition. To address the challenges of high sample demand and limited accuracy in current automatic annotation technologies, this paper proposes an automatic annotation method AP-GCN, which integrates Affinity Propagation (AP) clustering and Graph Convolutional neural Network (GCN). This method effectively integrates the advantages of AP clustering in capturing complex correlations with the capability of GCN for mining spatial distribution features. The Fuyu oil layer within the Zhouliu Block of the Sanzhao Depression in the Songliao Basin is selected as the target area to implement the automatic annotation of diagenetic facies samples. First, diagenetic facies types are summarized and the logging curve data are preprocessed; a limited number of labels are annotated to construct an automatic annotation dataset, which forms the basis for the subsequent automatic annotation process. Second, a graph structure is constructed by using AP clustering to establish correlations between the depth nodes of the logging curves. Subsequently, the node features are aggregated through the graph convolutional layer to achieve rapid and accurate annotation of diagenetic facies. Finally, a comparative experiment is designed to verify the effectiveness of the proposed method. The experimental results demonstrate that the precision of the AP-GCN method for various diagenetic facies annotations exceeds 86%, with recall above 90% and F1 value above 88%. The proposed method achieves an annotation accuracy of 90.6%, confirming its effectiveness and practicality. It provides a novel solution for the automatic annotation of diagenetic facies samples.
Volume 52 No 8
Superintended by:
China Electronics Technology Group Corporation
Sponsored by：
East-China Institute of Computer Technology
Shanghai Computer Society
Edited & Published by：
Editorial Office of Computer Engineering
Chief Editor：Wu Jiangxing
Executive Chief Editor:
Hu Bing
Tel：021-67092514
E-mail: ecice06@ecict.com.cn
CN: 31-1289/TP
ISSN: 1000-3428