期刊:
FRONTIERS IN PHYSICS,2023年11:1180413 ISSN:2296-424X
通讯作者:
Lin, H.;Yu, F.;Pham, V.-T.
作者机构:
[Yu, Fei] School of Computer and Communication Engineering, Changsha University of Science and Technology, Changsha, China;[Lin, Hairong] College of Computer Science and Electronic Engineering, Hunan University, Changsha, China;[Pham, Viet-Thanh] Faculty of Electrical and Electronics Engineering, Ton Duc Thang University, Ho Chi Minh City, Viet Nam
通讯机构:
[Yu, F.] S;[Lin, H.] C;[Pham, V.-T.] F;School of Computer and Communication Engineering, China;Faculty of Electrical and Electronics Engineering, Viet Nam
关键词:
application;editorial;non-linear device;non-linear networks;non-linear systems
摘要:
If there are nonlinear elements in the system or network, and the input and output are not superimposed and uniform, such system or network is called nonlinear system or nonlinear network. Nonlinearity makes the whole not equal to the sum of parts, and the superposition principle fails. At some joint points of nonlinear system, small changes in parameters often lead to qualitative changes in the form of motion, and behaviors that are essentially different from external excitation appear. It is the nonlinear effect that forms the infinite diversity, richness, tortuosity, singularity, complexity, variability and evolution of the material world.Nonlinear systems and networks have broad application prospects in the engineering fields of the Internet of Things, medical care, intelligent systems [1-5], etc. With the development of science and technology, according to the current research frontier, it is not difficult to find that the research fields of nonlinear systems and networks are also expanding, including chaotic systems and circuits [6][7][8][9], nonlinear device models [10][11][12], memristors [13][14][15][16], neural networks [17][18][19][20][21], neural circuits [22][23][24], synchronous control [25][26][27] and application research in related fields [28][29][30][31].Therefore, in this research topic, 12 articles about nonlinear systems and networks and their applications are reported. For nonlinear systems, a reverse single side band (RSSB) system with orthogonal fre...
摘要:
Trackers based on the Siamese network have received much attention in recent years, owing to its remarkable performance, and the task of object tracking is to predict the location of the target in current frame. However, during the tracking process, distractors with similar appearances affect the judgment of the tracker and lead to tracking failure. In order to solve this problem, we propose a Siamese visual tracker with spatial-channel attention and a ranking head network. Firstly, we propose a Spatial Channel Attention Module, which fuses the features of the template and the search region by capturing both the spatial and the channel information simultaneously, allowing the tracker to recognize the target to be tracked from the background. Secondly, we design a ranking head network. By introducing joint ranking loss terms including classification ranking loss and confidence&IoU ranking loss, classification and regression branches are linked to refine the tracking results. Through the mutual guidance between the classification confidence score and IoU, a better positioning regression box is selected to improve the performance of the tracker. To better demonstrate that our proposed method is effective, we test the proposed tracker on the OTB100, VOT2016, VOT2018, UAV123, and GOT-10k testing datasets. On OTB100, the precision and success rate of our tracker are 0.925 and 0.700, respectively. Considering accuracy and speed, our method, overall, achieves state-of-the-art performance.
期刊:
IEEE Transactions on Geoscience and Remote Sensing,2023年61:1-19 ISSN:0196-2892
作者机构:
[He, Ziping] School of Computer and Communication Engineering, Changsha University of Science and Technology, Changsha, China;[Wu, Panpan] College of Computer and Information Engineering, Tianjin Normal University, Tianjin, China;[Zu, Baokai; Li, Yafang; Li, Jianqiang; Wang, Hongyuan] Faculty of Information Technology, Beijing University of Technology, Beijing, China
摘要:
Hyperspectral image (HSI) classification attempts to classify each pixel, which is an important means of obtaining land-cover knowledge. HSIs are cubic data with spectral–spatial knowledge and can generally be considered as sequential data alongside spectral dimension. Unlike convolutional neural networks (CNNs), which mainly focus on local relationship models in images, transformers have been shown to be a powerful structure for qualifying sequence data. However, it lacks the excellent ability of CNNs in establishing local relationships in images and cannot perform good generalization in the case of insufficient data. In addition, the gradient disappearance problem hinders the convergence stability of deep learning networks as the layers get deeper. To address these problems, we propose a Cascaded Convolution-based Transformer with Densely Connected Mechanism (CDCformer) for HSI classification. First, we propose a cascaded convolution feature tokenization to extract spectral–spatial information, which will introduce some inductive bias properties of CNN into the transformer. In addition, we design a simple and effective densely connected transformer to enhance feature propagation and transfer memorable information from shallow to deep layers. It efficiently improves the performance of the transformer and extracts more discriminative spectral–spatial features from the HSI. Extensive experimental evaluation of three public hyperspectral datasets shows that CDCformer achieves competitive classification results.
摘要:
Wireless sensor node coverage optimization is a critical issue in wireless sensor networks (WSN), which is a commonly typical NP-hard problem. To enhance the coverage of wireless sensor networks, coverage optimization refers to the prudent placement of resource-constrained wireless sensor nodes. Current coverage optimization techniques frequently result in local optimums and have poor optimization performance. Based on the excellent optimization performance of artificial bee colony (ABC) algorithm, this paper presents a novel self-adaptive multi-strategy artificial bee colony (SaMABC) algorithm, which designs an appropriate strategy pool and a fine-grained adaptive selection mechanism according to the coverage optimization problem. Furthermore, the algorithm is improved through using simulated annealing approach and the dynamic search step to enhance its ability to jump out of the local optimum. Compared with the state-of-the-art optimization algorithms, the evaluation results carried out in several scenarios show that SaMABC obtains the best performance in terms of coverage optimization. Specifically, the coverage of wireless sensor networks in SaMABC achieves around 99.1% and outperforms the initial coverage by up to 14.1%.
关键词:
Deployment;unmanned aerial vehicles (UAVs);Internet of vehicle (IoV);highway interchange
摘要:
The three-dimensional deployment of Unmanned Aerial Vehicles (UAVs) has attracted extensive attention, especially for the Internet of Vehicles (IoV) in an emergency or to help the overloaded edge servers in traffic peaks. However, most existing works assume a two-dimensional road to simplify the design and modeling, while ignoring the interchange bridges scenario. In this scenario, UAVs deployment will face new challenges: the line-of-sight (LoS) transmission between the vehicles and UAVs is weakened due to the occlusion of the bridge body and vehicle movement. Meanwhile, energy consumption and the quantity of UAVs also need to be considered. In this paper, we propose an energy-aware 3D-deployment of UAVs, named 3D-UAV, to guarantee a high uplink rate with a minimized number of UAVs in IoV with Highway Interchange. First, considering the channel gain over bridges, 3D-UAV divides vehicles into several clusters. In each time slot, the number of clusters is iteratively optimized. Based on the clustering result, the flight altitude of the UAV is optimized in a stochastic gradient ascent (SGA) way aiming at maximizing the average uplink rate of transmission. Numerical results show that the proposed 3D-UAV can cover all vehicles on the highway interchange with the number of UAVs close to the theoretical lower bound. Meanwhile, it outperforms SOA, DRL, and HOLD methods in terms of the uplink rate and energy.
关键词:
Global contextual information;multiscale transformer;remote-sensing image;semantic segmentation
摘要:
Convolutional neural networks (CNNs) are powerful in extracting local information but lack the ability to model long-range dependencies. In contrast, the transformer relies on multihead self-attention mechanisms to effectively extract the global contextual information and thus model long-range dependencies. In this article, we propose a novel encoder–decoder structured semantic segmentation network, named CNN and multiscale transformer fusion network (CMTFNet), to extract and fuse local information and multiscale global contextual information of high-resolution remote-sensing images. Specifically, to further process the output features from the CNN encoder, we build a transformer decoder based on the multiscale multihead self-attention (M2SA) module for extracting rich multiscale global contextual information and channel information. Additionally, the transformer block introduces an efficient feed-forward network (E-FFN) to enhance the information interaction between different channels of the feature. Finally, the multiscale attention fusion (MAF) module fully fuses the feature information from different levels. We have conducted extensive comparison experiments and ablation experiments on the International Society for Photogrammetry and Remote Sensing (ISPRS) Vaihingen and Potsdam datasets. The extensive experimental results demonstrate that our proposed CMTFNet can obtain superior performance compared to the currently popular methods. The codes will be available at https://github.com/DrWuHonglin/CMTFNet .
通讯机构:
[Zhou, SR ] C;Changsha Univ Sci & Technol, Sch Comp & Commun Engn, Changsha 410114, Peoples R China.
关键词:
Person re-identification;Occluded person re-identification;Convolutional neural network;Transformer;Attention mechanism;Pose estimation
摘要:
In spite of Convolutional Neural Network (CNN) has dominated in the area of Person Re-Identification, Transformer-based methods have emerged with their advantages in computer vision for processing long sequences in recent two years. In this work, for the purpose of reinforcing complementary advantages of Transformer and CNN in computer vision, a concise method combining Convolution and Transformer is proposed to boost the performance. Firstly, a convolutional network with attention mechanism is employed to generate features with channel and inter-channel relationship information. Moreover, a feature enhancement module is designed to combine pose information and ViT information, and the heatmap generated by the pose estimator is applied to guide ViT features to become good discriminative features. Finally, a relationship reinforced transformer layer is proposed to effectively increase the relationship between features. Experimental results show that the proposed method achieves superior results than interrelated advanced methods on two large-scale person re-Identification benchmark datasets and one occlusion dataset. For Market-1501, our method called Fusion Pose Guidance and Transformer Feature Enhancement for Person Re-Identification gain 94.3% and 87.0% for Rank-1 and mAP respectively. For DukeMTMC-reID our method reaches 88.7% and 77.2% for Rank-1 and mAP respectively. Especially, for the dataset Occluded-Duke, compared with the state of art model HONet, our method, with up to 2.7% and 4.5% performance gains in Rank-1 and mAP respectively.
通讯机构:
[Zheng, B ] C;Changsha Univ Sci & Technol, Sch Comp & Commun Engn, Changsha 410114, Peoples R China.
关键词:
small target detection;Mul-BiFPN;M-SimAM;Focal EIoU
摘要:
At present, UAV aerial photography has a good prospect in agricultural production, disaster response, and other aspects. The application of UAVs can greatly improve work efficiency and decision-making accuracy. However, owing to inherent features such as a wide field of view and large differences in the target scale in UAV aerial photography images, this can lead to existing target detection algorithms missing small targets or causing incorrect detections. To solve these problems, this paper proposes a small target detection algorithm for UAV aerial photography based on improved YOLOv5s. Firstly, a small target detection layer is applied in the algorithm to improve the detection performance of small targets in aerial images. Secondly, the enhanced weighted bidirectional characteristic pyramid Mul-BiFPN is adopted to replace the PANet network to improve the speed and accuracy of target detection. Then, CIoU was replaced by Focal EIoU to accelerate network convergence and improve regression accuracy. Finally, a non-parametric attention mechanism called the M-SimAM module is added to enhance the feature extraction capability. The proposed algorithm was evaluated on the VisDrone-2019 dataset. Compared with the YOLOV5s, the algorithm improved by 7.30%, 4.60%, 5.60%, and 6.10%, respectively, in mAP@50, mAP@0.5:0.95, the accuracy rate (P), and the recall rate (R). The experiments show that the proposed algorithm has greatly improved performance on small targets compared to YOLOv5s.
摘要:
We propose and experimentally demonstrate a digital method to mitigate inter-symbol interference (ISI) in demultiplexing Nyquist optical time-division multiplexing (N-OTDM) signal using a simple but powerfully serviceable k-nearest neighbors (KNN) classifier. The proposed method is evaluated under four types of sampling pulse-width corresponding to various extents of ISI distortion in 160 Gbaud down to 40 Gbaud demultiplexing experiments, and the mitigation performance is found to be better along with the increasement of sampling pulse-width. Compared with the traditional decision-directed least mean square (DD-LMS) equalization method, a 3.9 dB required-OSNR (at a BER of 10-4) improvement is observed under the pulse -width of 5.3 ps. Demultiplexing performance is investigated under the pulse-widths of 3.2 ps and 3.9 ps as well, in which the improvement of required-OSNR are found to be 1.33 dB and 2.95 dB, respectively. The performance improvement clearly indicates the strong ISI mitigation ability of the KNN classifier for N-OTDM signal demultiplexing. In addition, experimental results also confirm that the proposed digital ISI mitigation method could release the pulse-width requirement of the N-OTDM demultiplexing, which paves the way for future practical application of N-OTDM technique.
关键词:
Deep learning;Person re-identification;Feature pyramid;Bidirectional fusion branch network;Penalty term-based trihard loss
摘要:
Person re-identification (Re-ID) is the recognition of the same person in different camera views. Because of the existence of highly similar persons and great differences of the same person in different scenes, and the fact that the features extracted by current mainstream models lose some fine-grained information, it is likely for the models to misidentify the query person. To tackle these challenges, we introduce a bidirectional fusion branch network with penalty term-based trihard loss (BFB-PTT). The BFB-PTT constructs a bidirectional fusion branch (BFB) network based on feature pyramid, where low-level features are transferred to a high-level feature space through fewer convolutional layers than most of the traditional CNN-based models have, thus retaining more local features to discriminate different pedestrians more accurately and effectively. Meanwhile, we propose using the penalty term-based trihard loss (PTT) to optimize the spatial structure of pedestrian’s samples, so that the similar samples are drawn closer together in order to reduce the variabilities of the same person in different scenes. We have conducted comprehensive experiments and analyses on the proposed method’s effectiveness on three challenging benchmarks, and the results show that our approach achieves competitive performance with the state-of-art models.
摘要:
Multiview clustering has attracted increasing attention to automatically divide instances into various groups without manual annotations. Traditional shadow methods discover the internal structure of data, while deep multiview clustering (DMVC) utilizes neural networks with clustering-friendly data embeddings. Although both of them achieve impressive performance in practical applications, we find that the former heavily relies on the quality of raw features, while the latter ignores the structure information of data. To address the above issue, we propose a novel method termed iterative deep structural graph contrast clustering (IDSGCC) for multiview raw data consisting of topology learning (TL), representation learning (RL), and graph structure contrastive learning to achieve better performance. The TL module aims to obtain a structured global graph with constraint structural information and then guides the RL to preserve the structural information. In the RL module, graph convolutional network (GCN) takes the global structural graph and raw features as inputs to aggregate the samples of the same cluster and keep the samples of different clusters away. Unlike previous methods performing contrastive learning at the representation level of the samples, in the graph contrastive learning module, we conduct contrastive learning at the graph structure level by imposing a regularization term on the similarity matrix. The credible neighbors of the samples are constructed as positive pairs through the credible graph, and other samples are constructed as negative pairs. The three modules promote each other and finally obtain clustering-friendly embedding. Also, we set up an iterative update mechanism to update the topology to obtain a more credible topology. Impressive clustering results are obtained through the iterative mechanism. Comparative experiments on eight multiview datasets show that our model outperforms the state-of-the-art traditional and deep clustering competitors.
通讯机构:
[Deke Guo] S;Science and Technology Laboratory on Information Systems Engineering, National University of Defense Technology, Changsha, China<&wdkj&>State Key Laboratory of High Performance Computing, National University of Defense Technology, Changsha, China
关键词:
Service function chain;SFC migration;Long-term budget;Dynamic networks
摘要:
Zeroing neural network (ZNN) has been applied to various time-varying problems solving, and numerous ZNN models have been developed in recent years, such as power-type varying-parameter ZNN (PT-VR-ZNN) for solving time-varying quadratic minimization problems, adaptive fuzzy-type ZNN (AFT-ZNN) for solving time-variant matrix inversion and fuzzy power ZNN (FPZNN) for solving time-varying quadratic programming problems. As a time-varying problem and imperative research hot spot in science and engineering, the synchronization of chaotic systems has developed for decades. However, the research on chaos synchronization using ZNN method is rarely reported. Therefore, this paper proposes a time-varying fuzzy parameter ZNN (TVFP-ZNN) model to realize chaotic systems synchronization against the external noises. The most prominent feature of the TVFP- ZNN model is that the time-varying fuzzy parameter generated by the fuzzy logic system is applied in this model. Moreover, the above mentioned three models are also applied to realize the same chaotic systems synchronization for comparison. Compared with above three models, the proposed TVFP-ZNN model not only possesses the fastest convergence speed, but also maintains strongest robustness to noises. Besides, the excellent performances of the TVFP-ZNN model are verified by rigorous mathematical validation. Furthermore, the effectiveness and robustness of the proposed TVFP-ZNN model for chaotic systems synchronization are verified by comparative numerical simulation results. Finally, the process of the proposed TVFP-ZNN model for chaotic system synchronization is displayed on the oscilloscope based on the field programmable gate array (FPGA) to further illustrate its practical application ability.
摘要:
Synchronization of memristive neural networks (MNNs) by using network control scheme has been widely and deeply studied. However, these researches are usually restricted to traditional continuous-time control methods for synchronization of the first-order MNNs. In this paper, we study the robust exponential synchronization of inertial memristive neural networks (IMNNs) with time-varying delays and parameter disturbance via event-triggered control (ETC) scheme. First, the delayed IMNNs with parameter disturbance are changed into first-order MNNs with parameter disturbance by constructing proper variable substitutions. Next, a kind of state feedback controller is designed to the response IMNN with parameter disturbance. Based on feedback controller, some ETC methods are provided to largely decrease the update times of controller. Then, some sufficient conditions are provided to realize robust exponential synchronization of delayed IMNNs with parameter disturbance via ETC scheme. Moreover, the Zeno behavior will not happen in all ETC conditions shown in this paper. Finally, numerical simulations are given to verify the advantages of the obtained results such as anti-interference performance and good reliability.
通讯机构:
[Zhuofan Liao] S;School of Computer and Communication Engineering, Changsha University of Science and Technology, Changsha, Hunan, 410114, China
关键词:
Edge computing (EC);Internet of Things (IoT);Blockchain;Consensus;Voting;Reputation
摘要:
By deploying edge servers around devices, edge computing brings computing resources far away from the cloud center close to the Internet of Things (IoT), which reduces latency and promotes the rapid development of IoT. Since edge servers and devices (hereafter referred to as NODEs) are highly scattered, blockchain is becoming one of the most promising solutions to enhance security issues for IoT. In the blockchain, a consensus mechanism determines how to achieve an agreement among nodes, hence it is an essential element for the operation and efficiency of the blockchain. However, due to the exponentially increasing of nodes in IoT, the consensus efficiency of the traditional consensus mechanism will be greatly reduced, and due to the lack of detection process for malicious nodes, its security will also reduce. To solve the above problems, in this paper, a Reputation and Voting based Consensus mechanism (RVC) is proposed. To reduce the time consumption of the consensus process, RVC adopts a reputation evaluation algorithm without complex hash calculations to select block proposers, which both consider the behaviors in edge computing and blockchain consensus. To prevent malicious nodes from participating in consensus, a filtering algorithm is designed for RVC, which can detect and filter nodes with malicious behaviors. Simulation results show that, RVC outperforms some traditional work. On time consumption, compared with AirBC and hybrid blockchain, RVC improved by 73.6% and 93.7% respectively. And on consensus security, RVC improved by 14% compared with LVBS. When the network scale and the proportion of malicious nodes change exponentially, RVC shows good scalability in terms of time consumption, successful consensus rate and transaction throughput.
摘要:
Scratch作为图形化编程中的热门课程吸引了广大中小学生,而对于学生所做的作品与标准作品之间差异性的评定通常是靠教师通过人工对比检查,对于教师不仅工作量大且耗费巨大精力,因此对于Scratch作品相似性的识别就可以辅助教师快速检测学生作品,从而提高教学效率.针对该问题,提出Siamese-BERT模型对两个Scratch作品之间的相似度进行检测.首先,对Scratch源文件进行解析提取原始积木块序列,根据积木块逻辑特征提出一种积木块重构算法,将原始积木块序列排序成Token序列,将Token序列作为CBOW(Continuous Bag of Words)模型的输入文本进行预训练,从而得到Scratch的词向量模型;再使用Siamese神经网络框架结合BERT(Bidirectional Encoder Representation from Transformers)模型组合训练,最终输入到余弦相似度函数进行相似度计算.数据集来自于长沙市Scratch培训机构的培训作品和学生的练习作品,在该数据集上,Siamese-BERT模型准确度能达到 0.82,对比其它的文本相似度模型,Siamese-BERT模型在Scratch作品相似度检测上更加准确.
作者机构:
[Cao, Dun; Wang, Jin; Ru, Jia; Qin, Jian] Changsha Univ Sci & Technol, Coll Comp & Commun Engn, Changsha 410114, Peoples R China.;[Tolba, Amr] King Saud Univ, Community Coll, Dept Comp Sci, Riyadh 11437, Saudi Arabia.;[Zhu, Min] Zhejiang Shuren Univ, Coll Informat Sci & Technol, Hangzhou 310015, Peoples R China.
通讯机构:
[Zhu, M ] Z;Zhejiang Shuren Univ, Coll Informat Sci & Technol, Hangzhou 310015, Peoples R China.
关键词:
Internet of vehicles;road networks;3D road model;structure recognition;GIS
摘要:
Internet of Vehicles (IoV) is a new system that enables individual vehicles to connect with nearby vehicles, people, transportation infrastructure, and networks, thereby realizing a more intelligent and efficient transportation system. The movement of vehicles and the three-dimensional (3D) nature of the road network cause the topological structure of IoV to have the high space and time complexity. Network modeling and structure recognition for 3D roads can benefit the description of topological changes for IoV. This paper proposes a 3D general road model based on discrete points of roads obtained from GIS. First, the constraints imposed by 3D roads on moving vehicles are analyzed. Then the effects of road curvature radius (Ra), longitudinal slope (Slo), and length (Len) on speed and acceleration are studied. Finally, a general 3D road network model based on road section features is established. This paper also presents intersection and road section recognition methods based on the structural features of the 3D road network model and the road features. Real GIS data from a specific region of Beijing is adopted to create the simulation scenario, and the simulation results validate the general 3D road network model and the recognition method. Therefore, this work makes contributions to the field of intelligent transportation by providing a comprehensive approach to modeling the 3D road network and its topological changes in achieving efficient traffic flow and improved road safety.
摘要:
In recent years, there has been a great development in the research of automated detection of diabetic retinopathy, and deep learning algorithms have been more and more widely used in this field. In this paper, we propose a channel cross enhancement network based on a two-stream model for diabetic retinopathy severity grading for the detailed performance of diabetic retinopathy images on different channels (RGB). The model takes the features of the full-channel input image as global features and the features extracted from the green channel of the original image as local features, and the local features complement the global features to enhance the model's ability to extract the global channel information of the image. In addition, a channel cross-attention module (CCAM) is designed to achieve the effective extraction of global channel features and the interaction of local channel features with global channel features. The proposed method is validated on the Messidor-2 dataset, and the experimental results show that the proposed method outperforms the existing methods in terms of accuracy and AUC values. After experimental validation, the method proposed in this paper can be effectively used for the auxiliary diagnosis of diabetic retinopathy, helping doctors to provide an effective basis for early clinical treatment.