作者机构:
[Dhelim, Sahraoui; Dhelim, S; Aung, Nyothiri; Kechadi, Tahar] Univ Coll Dublin, Dublin, Ireland.;[Zhu, Tao] Univ South China, Henyang, Peoples R China.;[Zerdoumi, Saber] United Nation Int Solar Energy Technol Transferri, Henyang, Peoples R China.;[Guerbouz, Tahar] Univ Ouargla, Ouargla, Algeria.
会议名称:
IEEE 7th International Conference on Intelligent Transportation Engineering (ICITE)
会议时间:
NOV 11-13, 2022
会议地点:
Beijing, PEOPLES R CHINA
会议主办单位:
[Aung, Nyothiri;Kechadi, Tahar;Dhelim, Sahraoui] Univ Coll Dublin, Dublin, Ireland.^[Zhu, Tao] Univ South China, Henyang, Peoples R China.^[Zerdoumi, Saber] United Nation Int Solar Energy Technol Transferri, Henyang, Peoples R China.^[Guerbouz, Tahar] Univ Ouargla, Ouargla, Algeria.
关键词:
blockchain;IoV;VANET;trust;security;privacy
摘要:
With the rapid development of the Internet of Things (IoT) and its potential integration with the traditional Vehicular Ad-Hoc Networks (VANETs), we have witnessed the emergence of the Internet of Vehicles (IoV), which promises to seamlessly integrate into smart transportation systems. However, the key characteristics of IoV, such as high-speed mobility and frequent disconnections make it difficult to manage its security and privacy. The Blockchain, as a distributed tamper-resistant ledge, has been proposed as an innovative solution that guarantees privacy-preserving yet secure schemes. In this paper, we review recent literature on the application of blockchain to IoV, in particular, and intelligent transportation systems in general.
会议名称:
9th International Forum on Electrical Engineering and Automation (IFEEA)
会议时间:
NOV 04-06, 2022
会议地点:
ELECTR NETWORK
会议主办单位:
[Wang, Shoutao;Chen, Wenguang;Liu, Zhijian;Wei, Caiyi;Peng, Yuanyuan] Univ South China, Sch Elect Engn, Hengyang, Peoples R China.
关键词:
Printed circuit board;Bidirectional pulse power supply;Forward and reverse pulse duty cycle ratio;Current density;Frequency
摘要:
Printed circuit boards are the most important basic electronic components in the electronics industry, and copper plating technology is widely used in the field of printed circuit board manufacturing. Pulse plating is more efficient and environmentally friendly than DC plating, and the plating layer of pulse plating is more uniform and detailed. The effect of bidirectional pulse power supply parameters on the copper plating effect of printed circuit board surface was investigated by using a self-developed bidirectional pulse power supply. The experimental results show that the frequency, forward and reverse pulse duty cycle and current density of the bidirectional pulse power supply have certain effects on the copper plating effect.
作者机构:
[Zhang, XZ; Zhang, Xiaozhi; Zhou, Liu] Univ South China, Sch Elect Engn, Hengyang 421001, Peoples R China.;[Zhu, Minjie; Ouyang, Lijun; Ouyang, Yan; Xiong, Dongping; Ouyang, LJ] Univ South China, Sch Comp Software, Hengyang 421001, Peoples R China.
会议名称:
14th International Conference on Graphics and Image Processing (ICGIP)
会议时间:
OCT 21-23, 2022
会议地点:
ELECTR NETWORK
会议主办单位:
[Zhou, Liu;Zhang, Xiaozhi] Univ South China, Sch Elect Engn, Hengyang 421001, Peoples R China.^[Zhu, Minjie;Xiong, Dongping;Ouyang, Lijun;Ouyang, Yan] Univ South China, Sch Comp Software, Hengyang 421001, Peoples R China.
会议论文集名称:
Proceedings of SPIE
关键词:
Magnetic resonance imaging (MRI);fast MRI;deep learning;Non-Local;remote dependencies
摘要:
As an advanced medical imaging technology, magnetic resonance imaging (MRI) has great advantages and application potentials in medical clinical diagnosis. However, since the long scanning time and the artifacts caused by patient movements, the imaging results are always not satisfactory. Therefore, accelerating MRI and improving the imaging quality are the key problems. In this work, we propose a novel deep network that combines the U-net architecture with non-local attention blocks for MRI reconstruction. We employ the U-net to construct the basic network. The non-local attention is exploited to capture the remote dependencies in MRI images which calculates the weighted average of the remaining multiple location features as the value of the response location. The U-net has limitations in capturing long-term dependencies, however, the non-local attention can solve this problem well. Furthermore, we develop the residual module to better retain the detail information. The proposed model is compared with some recent leading MRI reconstruction methods, including the state-of-the-art deep learning-based methods. Compared with these methods, the proposed residual non-local attention network provides superior MRI reconstruction results and retains better perceptual image details.
作者机构:
[Xie, Fuqiang; Guo, Lei; Yang, Bin] Univ South China, Sch Elect Engn, Hengyang, Peoples R China.
会议名称:
8th International Conference on Hydraulic and Civil Engineering - Deep Space Intelligent Development and Utilization Forum (ICHCE)
会议时间:
NOV 25-27, 2022
会议地点:
Xian, PEOPLES R CHINA
会议主办单位:
[Guo, Lei;Xie, Fuqiang;Yang, Bin] Univ South China, Sch Elect Engn, Hengyang, Peoples R China.
关键词:
component;Deep learning;Lane detection;non-local attention;frame-similarity loss
摘要:
Lane detection plays an important role in autonomous driving. For video instance lane detection, both global spatial and temporal information is significantly important. However, the global spatial features and the temporal features are not been well exploited in recent studies. In this work, we address the video instance lane detection task by capturing global context based on non-local attention network. Specifically, we designed a twin non-local attention network to extract long-range dependencies along the spatial and temporal dimensions, respectively. Meanwhile, the global spatial and temporal features can be adaptively fused by gating mechanisms for better results. In addition, the frame-similarity loss is proposed to further exploit the information of adjacent frames. The experimental results on the video instance lane detection (VIL-100) dataset verify that our method achieves better results compared with other comparison methods. Ablation experiments further demonstrate the effectiveness of each sub-module.
摘要:
Early warning of small civilian UAVs is a critical issue in the field of public security and is a challenging task in object detection. Video object detection methods based on deep convolution neural networks have attracted increasing attention over recent years. The object detection algorithm designed for static images can hardly be used directly in video object detection, due to motion blur, out-of-focus, etc. Traditional methods mainly apply temporal information to address this problem. In this paper, we proposed a method based on optical flow for UAV video object detection, employing optical flow information and the object appearance visual features in different stages. We establish a motion enhancement module to utilize motion features of moving objects while suppressing complex backgrounds. In addition, direction information is used to associate detection boxes of adjacent frames, improving the stability of the detection. The results of the experiments on the Anti-UAV2021 Challenge dataset show that the average precision of our model is 89.8%, significantly higher than that of FGFA, SELSA, and YOLOv5 models. The experimental results also demonstrate that the motion enhancement module and direction post-processing module can improve the recall rate of small UAVs under complex backgrounds.
作者机构:
[Wang, Chongjie; Wang, Qianwen; Wei, Caiyi; Chen, Wenguang] Univ South China, Hengyang, Peoples R China.
会议名称:
9th International Forum on Electrical Engineering and Automation (IFEEA)
会议时间:
NOV 04-06, 2022
会议地点:
ELECTR NETWORK
会议主办单位:
[Wei, Caiyi;Chen, Wenguang;Wang, Chongjie;Wang, Qianwen] Univ South China, Hengyang, Peoples R China.
关键词:
Interleaved parallel;Buck converters;Hall current sensor;sharing load
摘要:
There is no unified standard for the measurement and detection of Hall current sensors at presently. Manufacturer makes equipment for measurement and detection according to their own verification scheme. The High current detection standard device is the key device to verify the Hall current sensor, and its output current accuracy directly determines the measurement accuracy of the high-current Hall sensor. Therefore, a Hall high current sensor detection device scheme is proposed in this paper. The main circuit adopts a multi-phase Buck converter interleaved and parallel topology, which can not only reduce the total output current ripple, improve the output precision, but also reduce the current stress of each phase Buck converter. Aiming at the problem of current imbalance caused by parameter differences between components, the master-slave current sharing method is used to perform closed-loop feedback control to achieve current sharing between modules. Finally, the simulation results in PSIM show that the total output current of the four-phase Buck converter is 100A, and the output current ripple ratio is 0.1%.
作者机构:
[Li, Longjin; Jia, Lei] Univ South China, Dept Elect Engn, Hengyang, Peoples R China.
会议名称:
25th International Conference on Electrical Machines and Systems (ICEMS)
会议时间:
NOV 29-DEC 02, 2022
会议地点:
Rajamangala Univ Technol Lanna, Chiang Mai, THAILAND
会议主办单位:
Rajamangala Univ Technol Lanna
会议论文集名称:
International Conference on Electrical Machines and Systems ICEMS
关键词:
Brushless doubly-fed generator;Pole-changing;Finite element
摘要:
In this paper, a nested-loop rotor brushless doubly-fed generator (BDFG) with improved stator single winding pole-changing design scheme is presented. The BDFG is simulated at 400 rpm and 600 rpm through finite element (FE) method. It is found that the improved stator single winding pole-changing design scheme is effective in reducing the machine harmonic contents and improving the PW generated voltage, CW current waveforms quality.
作者机构:
[Ouyang, Chunping; Tian, Wenlong; Liu, Yongbin; Liu, Qifei; Li, Jing; Geng, Yuqing] Univ South China, Sch Comp Sci & Technol, Hengyang, Peoples R China.;[Tian, Wenlong] Nanyang Technol Univ, Sch Phys & Math Sci, Singapore, Singapore.;[Ouyang, Chunping; Tian, Wenlong; Liu, Yongbin] Hunan Prov Base Sci & Technol Innovat Cooperat, Hengyang, Peoples R China.;[Li, Ruixuan] Huazhong Univ Sci & Technol, Sch Comp Sci & Technol, Wuhan, Peoples R China.;[Xiao, Weijun] Virginia Commonwealth Univ, Elect & Comp Engn, Richmond, VA 23284 USA.
会议名称:
IEEE/ACM 15th International Conference on Utility and Cloud Computing (UCC) / 9th International Conference on Big Data Computing, Applications and Technologies (BDCAT)
会议时间:
DEC 06-09, 2022
会议地点:
Vancouver, WA
会议主办单位:
[Geng, Yuqing;Tian, Wenlong;Ouyang, Chunping;Liu, Yongbin;Liu, Qifei;Li, Jing] Univ South China, Sch Comp Sci & Technol, Hengyang, Peoples R China.^[Tian, Wenlong] Nanyang Technol Univ, Sch Phys & Math Sci, Singapore, Singapore.^[Tian, Wenlong;Ouyang, Chunping;Liu, Yongbin] Hunan Prov Base Sci & Technol Innovat Cooperat, Hengyang, Peoples R China.^[Li, Ruixuan] Huazhong Univ Sci & Technol, Sch Comp Sci & Technol, Wuhan, Peoples R China.^[Xiao, Weijun] Virginia Commonwealth Univ, Elect & Comp Engn, Richmond, VA 23284 USA.^[Xu, Zhiyong] Suffolk Univ, Math & Comp Sci Dept, Boston, MA 02114 USA.
关键词:
Cloud Storage;Resemblance Detection;Context-Aware;Deduplication Ratio Prediction
摘要:
With the prevalence of cloud storage, people prefer to outsource their data to the cloud for flexibility and reliability. Undoubtedly, there are lots of redundancy among these data. However, high-end storage with deduplication costs heavy computation and increases the data management complexity. Potential customers need the redundancy proportion information of their outsourced data to decide whether high-end storage with deduplication is worthwhile. Thus, many researchers have previously attempted to predict the redundant ratio. However, existing mechanisms ignore the redundancy proportion among similar chunks containing many duplicate data. Although resemblance detection, detecting the duplicate parts among similar data, has become a hot issue, it is hardly applied to the conventional deduplication ratio estimation because of unacceptable calculation cost. Therefore, we analyze the limitations and challenges of deduplication ratio prediction in prediction scope and response time and further propose a novel prediction scheme. By leveraging the context-aware resemblance detection, and confidence interval theory, our method can achieve faster estimation speed with higher accuracy in deduplication ratio compared with the state-of-the-art work. Finally, the results show that our method can efficiently and effectively estimate the proportion of duplicate chunks and redundant data among similar chunks by conducting experiments on real workloads.
作者机构:
[Shan, Wenyu; Chen, Cheng; Luo, Lingyun; Ding, Pingjian; Luo, Hanyu] Univ South China, Sch Comp Sci, Hengyang 421001, Hunan, Peoples R China.;[Luo, Lingyun] Hunan Med Big Data Int Sci & Technol Innovat Coop, Hengyang 421001, Peoples R China.
会议名称:
18th International Conference on Intelligent Computing (ICIC)
会议时间:
AUG 07-11, 2022
会议地点:
Xian, PEOPLES R CHINA
会议主办单位:
[Luo, Hanyu;Chen, Cheng;Shan, Wenyu;Ding, Pingjian;Luo, Lingyun] Univ South China, Sch Comp Sci, Hengyang 421001, Hunan, Peoples R China.^[Luo, Lingyun] Hunan Med Big Data Int Sci & Technol Innovat Coop, Hengyang 421001, Peoples R China.
摘要:
Enhancers are small segments of DNA that bind to proteins (transcription factors) and the transcription of a gene is strengthened after binding to the protein, thus playing an essential role in gene expression. Recently, machine learning-based methods have become a trend in identifying enhancers and their strength. In this study, we propose iEnhancer-BERT, a novel transfer learning method based on pre-trained DNA language model using the whole human genome. More specifically, iEnhancer-BERT consists of a BERT layer for feature extraction and a CNN layer for classification. We initialize our parameters of the BERT layer using a pre-trained DNA language model, and fine-tune it with transfer learning on the enhancer identification tasks. Unlike common fine-tuning strategies, we extract the output of all Transformer Encoder layers to form the feature vector. Experiments show that our method achieves state-of-the-art results in both enhancer identification tasks and strong enhancer identification tasks. The code and data are publicly available at https://github.com/lhy0322/iEnhancer-BERT.
作者机构:
[Zhang, Liangfa] Univ South China, Hengyang, Peoples R China.
会议名称:
IEEE International Conference on Electrical Engineering, Big Data and Algorithms (EEBDA)
会议时间:
FEB 25-27, 2022
会议地点:
Changchun, PEOPLES R CHINA
会议主办单位:
[Zhang, Liangfa] Univ South China, Hengyang, Peoples R China.
关键词:
Face recognition;Pose variation;Deep learning
摘要:
Face recognition beneath pose changes, recently, is a very tough problem and key point in face recognition, a lot of research institutes have tried to figure out the best way to solve the tough problem by producing different method. Recently, as the success development of deep learning in many fields of computer vision, deep learning is also applied to this problem and obtains the state-of-the-art performance. This review is trying to analyze the basic idea and introduce the state-of-the-art algorithms in this field.
摘要:
Nowadays, falling is a growing threat to the elderly. This paper combines millimeter wave radar technology, machine learning algorithm, wireless communication technique and cloud platform to realize a fall detection system. In this project, the millimeter wave radar is used to sample the human posture point cloud data, and we create a data set that consist of point of clouds of two different human poses. Random forest and BP neural network are used to train the fall detection model. The system will send the human posture point data to the trained model and realize the fall detection. Besides, the system will use 4G communication technology to transform the data to the web cloud platform. This web page serves as an warning function, which can report the acceleration, speed and other information. According to our experiments, the millimeter wave radar system that we built in this paper can effectively detect human point cloud, and can send human point cloud data packets to the recognition model to detect human falls. In machine learning part, both Random forest and BP neural network models show very strong robustness after repeated adjusting parameters. Random forest has the advantages of light weight and interpretability, which can reach 93% recognition accuracy. The recognition accuracy of BP neural network is even up to 95%, which is higher than other detection models in previous works. Besides, BP neural network model also has higher recall rate in the categories of human falls, which fully meets the requirements of this project.
作者机构:
[He, Jianchong; Lu, Yao] Sun Yat Sen Univ, Sch Data & Comp Sci, Guangzhou, Peoples R China.;[Liang, Xiaowen] Guangzhou Med Univ, Affiliated Hosp 3, Guangzhou, Peoples R China.;[Wei, Jun] Percept Vis Med Technol Co Ltd, Guangzhou, Peoples R China.;[Wei, Jun; Chen, Zhiyi] Univ South China, Affiliated Hosp 1, Med Imaging Ctr, Hengyang, Peoples R China.
会议名称:
13th International Conference on Graphics and Image Processing (ICGIP)
会议时间:
AUG 18-20, 2021
会议地点:
Yunnan Univ, Kunming, PEOPLES R CHINA
会议主办单位:
Yunnan Univ
会议论文集名称:
Proceedings of SPIE
关键词:
Endometrium thickness;ultrasound image;multi-task learning;image process
摘要:
Endometrial receptivity assessment based on the ultrasound image is a common and non-invasive way in clinician practice. Clinicians consider that the thickness of the endometrium is one of the most important assessment markers, which can be calculated with the endometrial region in ultrasound images. Suffering from low contrast of the boundaries in ultrasound images, it's a challenge that makes accurate segmentation of endometrial for thickness calculation. An automated assessment framework with a multi-task learning segmentation network is proposed in this paper. The VGG-based U-net is trained with an auxiliary pattern classification task, the losses of different tasks are combined by weighted sum based on uncertainty in the training phase. Experiment shows that the network has a more accurate prediction than single-task learning and the framework does a better thickness calculation.
作者机构:
[Liu, Chong; Yang, Mei] Univ South China, Sch Elect Engn, Hengyang 421001, Peoples R China.;[Shao, CP; Shao, Cuiping; Yang, Mei; Li, Huiyun] Chinese Acad Sci, Shenzhen Inst Adv Technol, Shenzhen 518055, Peoples R China.
会议名称:
21st IEEE International Conference on Trust, Security and Privacy in Computing and Communications (IEEE TrustCom)
会议时间:
DEC 09-11, 2022
会议地点:
Wuhan, PEOPLES R CHINA
会议主办单位:
[Yang, Mei;Liu, Chong] Univ South China, Sch Elect Engn, Hengyang 421001, Peoples R China.^[Yang, Mei;Li, Huiyun;Shao, Cuiping] Chinese Acad Sci, Shenzhen Inst Adv Technol, Shenzhen 518055, Peoples R China.
会议论文集名称:
IEEE International Conference on Trust Security and Privacy in Computing and Communications
关键词:
Internet of Vehicles;Security certification;SM2 Digital signature;HF(2(233));FPGA
摘要:
The rapid development of the Internet of Vehicles (IoV) provides a strong technical guarantee for intelligent transportation, greatly facilitating people's daily travel. At the same time, its security problems are becoming increasingly prominent. Fortunately, the cryptographic integrated circuits (ICs) provide a security guarantee for the IoV access authentication, traffic management, data communication, etc., which is the core and cornerstone of the IoV cryptographic technology. However, the IoV has high requirements for timeliness, and its communication resources are precious, so its necessary to ensure that the overhead of the cryptographic module is small and the delay is low. In this paper, we adopt SM2 Elliptic Curve Public Key Digital Signature Algorithm-with fast operation speed and short signature data to implement cryptographic ICs. We design and optimize its hardware design to balance overhead and efficiency. Based on the Montgomery point multiplication algorithm in Lopez-Dahab (LD) projection coordinates, we have researched the core point multiplication operation in SM2 and optimized the time-consuming operation in finite fields, which improved the computational efficiency of SM2. Finally, we completed the hardware design of SM2 on GF(2(233)) domain and verified it on a Xilinx Kintex-7 FPGA development board. The experiment results show that the design occupies a total of 77,665 Slice LUTs and merely takes 2.57 mu s to complete a signature verification. The signature verification rate is 389,105 times/s. Compared with traditional solutions, our proposed method achieves less overhead and little latency.
摘要:
Sensing intrusion is a new threat to information security of automatic driving system, which employs digital noise like adversarial sample to show sensors fake information, aiming to mislead decision making and eventually achieve the hacker's illegal intention. Unfortunately, it is difficult for most traditional information security techniques to deal with this novel risk. Even if the methods like outlier detection can pick out the abnormal sensing data, they still hardly tell the samples containing adversarial noise. In this paper, a method based on semantic similarity check is proposed to address this issue. Scene semantic centroid is introduced to represent the core semantics of a category of driving scene. Correspondingly, each component of the centroid is used to depict the standard sensing semantics for each sensor in this kind of scene. On this basis, a straightforward algorithm is proposed to simultaneously detect both abnormal driving scene and the sensors that may be hacked. Considering the raw sensing semantic space is too high dimensional and too sparse to handle efficiently, a scene semantic autoencoder is developed to extract scene semantic centroid by semantic embedding. The experiments on Nuscenes dataset show that the proposed method is not only feasible to identify the suspicious intruded sensors, but also more effective and accurate than the traditional abnormal sensing data detection.
摘要:
The current mainstream knowledge tracking model is based on the neural network of deep learning, which has a certain improvement in performance. However, due to the difficulty of interpretability of the deep learning methods, and the previous literature did not involve the high-dimensional information between problems and knowledge points when their model used the answer record, there is a situation that the relevant information is not sufficiently extracted. In order to solve the above problems, a knowledge tracing model based on the graph attention network mechanism is proposed, which uses the graph attention network to reveal the potential graph structure between knowledge points in answer records, and aggregates the correlation degree through the attention mechanism, so that the input information of the model includes the relationship information between problems and knowledge points, which enhances the interpretability of the model and improves the prediction accuracy of the model. On the three commonly used public datasets, the proposed model can better reflect learners' mastery of knowledge points.
摘要:
This work investigates a joint power allocation and transmission scheduling algorithm for underwater acoustic (UWA) networks to improve the throughput by leveraging the low sound speed in water. We consider a round-based UWA system where transmission links are active concurrently to send multiple signal blocks during each round. To mitigate the interference, a quasi-interference alignment (IA) method is proposed where the interference signals are aligned as much as possible while a low level of collisions between useful and interference signals is allowed. Specifically, the transmission starting time and transmission power are jointly optimized to maximize the total transmission capacity. Different from the conventional IA, the proposed mechanism intends to reduce the total reception time by tolerating slight collisions of useful signals while adapts the power allocation to compensate for the collisions, which could increase the transmission degrees of freedom (d.o.f.). To solve the optimization problem in a tractable way, we decompose the problem into two smaller sub-problems which optimize the transmission schedule and power allocation with a smaller feasible region. The simulation results show that, with an appropriate collision level, the proposed method could achieve decent throughput improvement, compared to the conventional IA and a benchmark method which solves the optimization problem by the genetic algorithm.
作者机构:
[Yan, Shiyu; Yang, Xiaohua; Li, Meng; Wen, Zhaoyu] Univ South China, Sch Comp Sci, Hengyang 421001, Peoples R China.
会议名称:
3rd International Conference on Electronics and Communication; Network and Computer Technology (ECNCT)
会议时间:
DEC 03-05, 2021
会议地点:
Xiamen, PEOPLES R CHINA
会议主办单位:
[Wen, Zhaoyu;Yang, Xiaohua;Yan, Shiyu;Li, Meng] Univ South China, Sch Comp Sci, Hengyang 421001, Peoples R China.
会议论文集名称:
Proceedings of SPIE
关键词:
metamorphic testing;metamorphic relation;likely metamorphic relation;GEP;neutron diffusion program
摘要:
The metamorphic test is a method to alleviate the unexpected value of the Oracle problem. The key point is the identification of the metamorphic relations. The identification of the metamorphic relations of scientific calculation programs is also a complex problem, and the likely of the metamorphic relation of the program can provide enlightening information for the identification of the metamorphic relations. The likely metamorphic relation can be regarded as the implicit expression of the input pattern and output pattern. This paper proposes an output pattern recognition technology based on the likely metamorphic relations of GEP. The technology is mainly aimed at the core neutron diffusion calculation program. The input pattern of the program, and then generate input data and run the program. Finally, in the corresponding output data results, through GEP data mining technology, the output pattern expressed in a variety of functional forms is obtained, which is further compared with analytical solutions and verified to be reliable likely metamorphic relations.