Most papers are machine-translated to English; some are originally in English.
Showing 50 of 1733 papers
To address the limitations of conventional salient object detection methods in detecting multiple salient objects at different scales, we propose a salient object detection algorithm based on deep reuse of multi-scale features. The network model consists of vertically stacked bidirectional dense fea…
Emerging intelligent transportation systems hold significant promise for improving traffic flow, optimizing fuel efficiency, reducing delays, and enhancing overall driving experience. Currently, traffic congestion constitutes an extremely serious challenge for humanity, particularly severe at urban …
To address the issues of insufficient feature extraction for facial expressions and inadequate generalization capability of lightweight networks in complex environments, as well as ambiguous expressions caused by single-label datasets' inability to effectively describe complex emotional tendencies, …
To improve the real-time estimation accuracy of human vigilance, a real-time vigilance estimation method based on differential entropy (DE), improved moving average, and bidirectional two-dimensional principal component analysis (TD-2DPCA) is proposed. First, the total frequency band is decomposed i…
Deep Reinforcement Learning (DRL) can be widely applied in the field of urban traffic signal control; however, in existing research, the vast majority of DRL agents utilize only the current traffic state for decision-making, resulting in limited control effectiveness under conditions of significant …
To address the issue of attackers exploiting background knowledge and other information to launch attacks during user location privacy protection, this paper proposes a location privacy protection method for mobile terminals. The scheme leverages k-anonymity and local differential privacy techniques…
For the Dual-Resource Constrained Flexible Job Shop Scheduling Problem (DRCFJSP), a Cuckoo Algorithm with an improved decoding scheme is proposed to optimize the makespan. Since DRCFJSP requires consideration of both machine allocation and worker processing status, the traditional decoding method is…
Recommendation systems have found extensive applications across diverse domains, profoundly influencing daily life. Training an effective recommendation system typically requires vast amounts of "user-item" interaction data; however, data obtained in practice is often extremely sparse, frequently le…
Hand-drawn sketch image translation is a challenging research topic in computer vision, with significant application value in artistic design and e-commerce. Currently, GAN-based hand-drawn sketch image translation remains in its infancy. This article analyzes the challenging issues confronting sket…
To address the issue that inpainting networks in generative adversarial networks fail to simultaneously preserve global and local consistency of images while incurring substantial computational load, we incorporate the concept of progressive inpainting into an asymmetric U-Net architecture. First, w…
Max-SAT is the optimization version of the SAT problem, with the objective of finding a variable assignment that maximizes the number of satisfied clauses in a given clause set. This problem is typically NP-hard. With the in-depth development of big data and artificial intelligence, existing algorit…
To address issues such as base station coverage holes and local traffic overload in urban vehicular networks, this paper proposes a dynamic pre-deployment scheme based on vehicle trajectory prediction information. First, to train a unified Seq2Seq-GRU trajectory prediction model, multiple UAVs equip…
Traffic flow prediction constitutes a crucial component of intelligent transportation systems. Due to the complexity of traffic data, long-term and accurate traffic flow prediction has consistently remained one of the most challenging tasks in time series forecasting. In recent years, researchers ha…
Few-shot relation extraction is a prominent research problem in natural language processing, which aims to train relation extraction models with low-cost annotated data. The currently widely-used prototype networks suffer from issues of inaccurate and incomplete class prototype representation. To ad…
In mobile crowdsensing systems, tasks exhibit spatiotemporal coverage overlap, which may lead to redundant data collection and consequently cause data redundancy issues. To address this, we propose a task allocation method that can simultaneously control both intra-task and inter-task data redundanc…
When swarm intelligence heuristic algorithms solve the Discounted {0-1} Knapsack Problem (D{0-1}KP), some repair and optimization strategy is required to transform abnormal encoding individuals into encoding individuals that satisfy the solution constraints, in order to improve solution efficiency a…
In open source communities, the promptness and quality of responses to developers' issues critically determine community vitality. Therefore, identifying and recommending appropriate problem solvers for newly submitted issues facilitates community development. This paper proposes a problem solver re…
Graph Convolutional Networks (GCNs) can extract effective information from graph data through graph convolutions, but they are vulnerable to adversarial attacks that degrade model performance. While adversarial training can enhance neural network robustness, the discrete nature of graph structures a…
To address the challenges of next-generation networks in coverage, deployment cost, and capacity, Mobile Edge Computing (MEC) often necessitates the assistance of relay nodes to execute computation-intensive and delay-sensitive tasks. This paper first introduces the fundamental architecture of relay…
Existing short-term demand forecasting models for public bicycles have neglected the nature differences in how various environmental factors influence user demand, as well as the temporal dependencies of variable environmental factors. This paper distinguishes environmental factors into invariant fa…
Slater voting rule is a tournament-based voting rule that primarily constructs an acyclic tournament, finds one with minimal difference from the original tournament, and selects the winner from it. For the NP-hard Slater voting algorithm, a Picat method for optimized solving of the Slater problem ba…
A novel network framework (SR-Net) for RGB-D salient object detection is proposed. To effectively integrate the complementarity of multimodal features, depth feature extraction is employed as an independent branch, the Convolutional Block Attention Module (CBAM) is adopted for depth feature enhancem…
Given that edge clouds lack more powerful computational processing capabilities compared to central clouds, they are susceptible to unnecessary scaling jitter or insufficient resource processing capability when handling dynamic workloads. This paper therefore conducts experimental evaluations on mic…
To address the limitation that existing clothing compatibility models predominantly focus on exploring compatibility between paired items, we propose a clothing compatibility prediction model based on hypergraph representation. The model first constructs a clothing hypergraph based on the different …
To address issues such as key abuse and digital certificate management in existing lattice-based proxy re-encryption schemes, we introduce an accountability mechanism and propose a novel identity-based accountable proxy re-encryption scheme. This scheme employs user identity ID to compute and genera…
Most existing algorithms utilizing Graph Neural Networks for text classification have overlooked the over-smoothing problem inherent in GNNs and the discrepancies arising from text graph topological differences, leading to suboptimal performance in text classification. To address this issue, we prop…
To address the issue that existing vertical handover algorithms in heterogeneous wireless network environments struggle to achieve seamless handover for mobile users accessing networks and fail to provide stable communication services, a vertical handover algorithm based on mobile user location pred…
To address issues such as poor car-following stability, ineffective tracking, or unsafe conditions caused by significant speed fluctuations of vehicles in congested environments, a multi-objective optimization car-following scheme based on vehicle models and deep reinforcement learning is proposed. …
Rain streaks severely degrade the quality of captured images, affecting subsequent computer vision tasks. To improve the quality of rainy images, we propose a Transformer-based single image deraining algorithm. First, the algorithm obtains a large receptive field through a transformer with a windowi…
To address the inability of traditional software-based protocol stacks to meet the demands of high-speed data transmission and processing, a hardware-accelerated UDP protocol stack design scheme is proposed. Based on the highly efficient parallel characteristics of hardware, this scheme implements a…
To address the problem of inconsistency between control-plane and data-plane flow rules caused by hardware/software failures and misconfigurations in the SDN data plane, we propose a P4-based consistency verification mechanism for SDN control-data plane flow rules (P4CV, P4-based Consistency Verific…
To satisfy the personalized requirements of end users and reduce the transmission delay in D2D networks, a cooperative retransmission scheme based on instantly decodable network coding (IDNC) with terminal differentiation is proposed. First, this scheme introduces a novel IDNC algorithmic framework …
With the rapid development of the social economy, the sustainability of employee altruistic behavior has become increasingly important for organizations and their members to better adapt and develop in the current rapidly changing work environment. Recent studies have found that altruistic behavior …
Residential mobility refers to the frequency with which individuals change their place of residence (i.e., moving). As a key socio-ecological factor, it exerts significant influence on individuals' cognition, emotion, and behavior. Previous scholars have reviewed the consequences of residential mobi…
This study proposes a personal fit statistic R for polytomous scoring items to investigate its performance in detecting six common aberrant response patterns (cheating, guessing, random responding, carelessness, innovative responding, and mixed aberrance), and compares it with the standardized log-l…
This year marks the centennial anniversary of the School of Information Management at Wuhan University (its predecessor being the Library Science Department of Boone University, founded in 1920), signifying that library science education in China has passed through a century of development. In retro…
[Purpose/Significance] Library and information science graduate students constitute the emerging force in the development of the LIS profession. Development in the new era has presented new demands for LIS education, requiring further enhancement of core competencies among LIS graduate students. [Re…
［Purpose/Significance］Knowledge service represents both the core capability and key trend for libraries, as well as a focal content and business direction for publishing institutions. In the digital intelligence era, investigating the similarities and differences between knowledge services in librar…
Objective: Pregnancy is a process during which the body's physiological state undergoes profound changes. The normal course of pregnancy is crucial for ensuring proper fetal development and maintaining maternal health. We attempted to explore whether changes in the urinary proteome can reflect the p…
In this paper, we propose and study a novel continuous-time model, based on the well-known constant elasticity of variance (CEV) model, to describe the asset price process. The basic idea is that the volatility elasticity of the CEV model cannot be treated as a constant from the perspective of stoch…
Hibiscus mutabilis has a long cultivation history and is an ancient garden tree species and medicinal plant native to China. To investigate the evolutionary characteristics of H. mutabilis cultivars and their relatives, clarify the phylogenetic relationships among H. mutabilis cultivars and between …
SPL transcription factors are widely involved in plant growth, development, and stress response processes. Currently, there are no studies on SPL membrane-bound (SPL with Transmembrane Motif) transcription factors, i.e., STM transcription factors, in Dendrobium officinale. To investigate the roles o…
Phenological changes driven by global warming have already exerted significant impacts on biodiversity and ecosystems. Research on subtropical phenology remains relatively limited compared to that in temperate and frigid zones. Studies of autumn phenology are particularly scarce, and whether phenolo…
Late embryogenesis abundant (LEA) proteins are closely associated with plant stress resistance, protecting plant cells under drought stress, reducing plant damage, and are widely distributed across organisms. Selaginella pulvinata is a fern species with exceptional survival capacity under drought st…
The Baiku Yao people have a strong tradition of raising and consuming local black pigs, and possess unique and extensive experience and knowledge regarding the collection and management of feed plants. This study aims to conduct an ethnobotanical inventory of feed plants and their related indigenous…
Isoëtes is the most anciently originated genus of aquatic vascular plants, with all species listed as National First-Class Protected Plants. Through nationwide investigation and research on Isoëtes plants, significant morphological differences were found among tetraploid individuals from different l…
This study investigated the response characteristics of soil enzyme activity to biochar input to provide a theoretical reference for the rational application of eucalyptus branch biochar. Based on a field experiment in eucalyptus plantations in northern Guangxi, biochar prepared from eucalyptus plan…
Keteleeria davidiana var. calcarea is a precious tree species in mid-subtropical karst regions, characterized by a narrow distribution range and scarcity of wild seedlings, which severely constrains its natural regeneration. To investigate the causes of its endangerment and provide a reference for i…
To investigate the effects of different land use types and ecological restoration patterns on soil enzyme activities and their C:N:P ratios in karst regions, this study selected three land use types [degraded disturbed land, grassland, and fruit tree (loquat) forest land] and four restoration patter…
This study conducted a taxonomic investigation on a morphologically distinct specimen of the genus Carcinomyces collected from Liancheng National Nature Reserve in Gansu Province, using morphological characterization and molecular phylogenetic analysis based on ITS sequences. The results indicate th…
