检索结果-内蒙古大学图书馆

IEEE International Conference on Image processing

作者： Zihan Ni Jia Chen Nong Sang Changxin Gao Leyuan Liu Key Laboratory of Ministry of Education for Image Processing and Intelligent Control Huazhong University of Science and Technology Wuhan China

This paper proposes an efficient model named Light YOLO for hand gesture recognition on the embedded platforms. Light YOLO improves accuracy, speed, and model size, in three aspects. To deal with the small scale gestures in practical applications, we strengthen the YOLOv2 with a spatial refinement module to obtain fine-grained features. To accelerate the refined network, we propose a selective-dropout channel pruning approach to prune the redundancy convolution kernels in the network. Moreover, we introduce a dataset for hand gesture recognition in complex scenes. The experimental results on this dataset show that the proposed Light YOLO significantly improve the YOLOv2 network, i.e., accuracy from 96.80% to 98.06%, speed form 40PFS to 125FPS, and size form 250M to 4MB.

关键词： Gesture recognition Feature extraction Detectors Acceleration Taylor series Kernel Cost function

来源：评论

学校读者我要写书评

暂无评论

Multi-Scale YOLOv2 for Hand Detection in Complex Scenes

Multi-Scale YOLOv2 for Hand Detection in Complex Scenes

引用

International Conference on Control, Automation, Robotics and Vision (ICARCV)

作者： Jia Chen Zihan Ni Nong Sang Key Laboratory of Ministry of Education for Image Processing and Intelligent Control Huazhong University of Science and Technology Wuhan China

This paper presents a model named Multi-Scale YOLOv2 (MS-YOLOv2) for hand detection in complex scenes. The proposed MS-YOLOv2 is implemented by introducing three modules to YOLOv2, including a Multi-Scale Feature Refinement Module to acquire fine-grained features, a Channel Importance Evaluation Module to recalibrate feature channels and a Hard Example Punishment Module to get rid of hand interference areas. Experiment results show that the proposed MS-YOLOv2 makes much performance improvement to YOLOv2, but with little computational complexity gain. On our dataset, the proposed MS-YOLOv2 can achieve 98.2% of AP and 97.9% of AR. Moreover, on the VIVA challenge, the proposed MS-YOLOv2 achieves AP/AR of 85.1%/45.8% at Level-1 and 80.1%/45.9% at Level-2.

关键词： Feature extraction Detectors Object detection Skin Color Gesture recognition

来源：评论

学校读者我要写书评

暂无评论

Effects of Control Parameters on Oxygen Distribution in Czochralski Crystal Growth Process

Effects of Control Parameters on Oxygen Distribution in Czoc...

引用

Chinese Automation Congress

作者： Jun-Chao Ren Ding Liu Yin Wan Shaanxi Key Laboratory of Complex System Control and Intelligent Information Processing Xi‘an University of Technology Xi'an China

The application of a quadrupole magnetic field structure with low power consumption and high field strength is very effective in solving the problem of non-uniformity of oxygen concentration distribution in the process of industrial drawing single crystal silicon by MCZ method. In this paper, a 3D model is used to simulate the crystal growth process under the action of quadrupole magnetic field for accurately describe the effect of control parameters on the oxygen concentration distribution at the crystal-melt interface. The results show that the radial oxygen concentration gradient at the crystal-melt interface is positively correlated increases with the radial temperature gradient when the magnetic induction strength is increased, while the positive correlation decreases with the increase of the crucible rotation speed or the crystal rotation speed within the appropriate range. Moreover, the uniformity of the radial oxygen concentration distribution at the crystal-melt interface can be improved by about 19.11% when the crystal rotation speed is higher and the crucible rotation speed is lower, which is more favorable for improving the quality of the silicon single crystal.

关键词： Quadrupole magnetic field Czochralski silicon single crystals Numerical simulation Crystal-melt interface Oxygen concentration distribution

来源：评论

学校读者我要写书评

暂无评论

Deep Learning for Sleep Stage Classification

Deep Learning for Sleep Stage Classification

引用

Chinese Automation Congress (CAC)

作者： Yang Wang Dongrui Wu Key Laboratory of the Ministry of Education for Image Processing and Intelligent Control Huazhong University of Science and Technology Wuhan China

Scoring of sleep stages plays an important role in the diagnosis of sleep-related diseases. Scoring by visual inspection is time-consuming and heavily depends on the experience of experts. Thus, there is an urgent need for an automatic sleep stage classification system. This paper proposes a novel compact convolutional neural network (C-CNN) using only single-channel EEG signal. Compared with traditional machine learning approaches based on hand-engineered features, our approach provides an end-to-end solution that requires almost no prior knowledge and preprocessing while achieving better performance. Experiments on the expanded Sleep-EDF database verified its effectiveness and efficiency. In addition, we notice the issue of class imbalance in sleep stages, and a class-imbalance metric, the balanced classification accuracy (BCA), is introduced. At the cost of a little drop in accuracy, which is still higher than existing classification methods, the introduction of class-imbalance weights can significantly increase the BCA metric and result in a higher recall for each sleep stage. This paper also proposes a recurrent neural network based on the attention mechanism and bidirectional long short-term memory (LSTM), which provides better performance than C-CNN but requires more training time.

关键词： Feature extraction Sleep Electroencephalography Training Convolution Brain modeling Deep learning

来源：评论

学校读者我要写书评

暂无评论

Fast S-transform for fault line selection in distribution network system

Fast S-transform for fault line selection in distribution ne...

引用

IEEE Conference on Industrial Electronics and Applications (ICIEA)

作者： Jiao Shang-bin Yang Yun-peng Shi Yan-jin Shi Wen-wen Guo Jing-wen Shaanxi Key Laboratory of Complex System Control and Intelligent Information Processing Xi'an University of Technology Xi'an China

The current of the distribution networks under 66kV is widely operating in small current grounding mode. The problem of extracting the fault feature in each line would be quite difficult due to the short duration of the transient zero sequence current with the non-stationary and non-linear characteristics when the single-phase-to-ground fault occurs in the system. Therefore, a novel method of single-phase-to-ground fault line selection for small current based on Fast S-transform (FST) is developed in this paper. Firstly, FST is presented to transform the transient electric current signal in each feeder, after that the one-dimension modulus and phase angle value can been obtained. Then, the feeder with the largest energy value calculated by the modulus obtained above is selected as the alternative possible fault line. Finally, line failure or bus failure can been determined by evaluating the difference of the phase angle between the fault line and the comparison line that has been selected from the non-alternative feeders before. The results of the simulative and experimental circuits reveal that the proposed method using FST can avoid the influence of neural point grounding type. Additionally, the method of line selection is quite accurate and real-time.

关键词： Transient analysis Circuit faults Grounding Time-frequency analysis Feature extraction Fourier transforms

来源：评论

学校读者我要写书评

暂无评论

Color Space Volume and Superpixel Based Leukocyte Image Segmentation

Color Space Volume and Superpixel Based Leukocyte Image Segm...

引用

IEEE International Symposium on Information (IT) in Medicine and Education, ITME

作者： Xiaogen Zhou Zuoyong Li Chuansheng Wang Fujian Provincial Key Laboratory of Information Processing and Intelligent Control Minjiang University Fuzhou P.R. China College of Mathematics and Computer Science Fuzhou University Fuzhou P.R. China School of Computer Science and Technology Harbin University of Science and Technology Harbin P.R. China

In this paper, we propose a novel method based on color space volume and multimodal histogram thresholding (MHT) for leukocyte segmentation in human blood smear images under the rapid staining condition. The proposed method first defines a new color space volume to highlight the leukocyte's nucleus region, and it then uses the MHT method to extract the nucleus. Then, the proposed method takes the central region in the superpixel segmentation result as an initial leukocyte region, and it then uses the MHT method to extract the leukocyte region. Finally, the cytoplasm is obtained by subtracting the nucleus region from the leukocyte region. Experimental results on a dataset containing 300 leukocyte images under rapid staining condition show that the proposed method provides a higher segmentation accuracy than the counterparts.

关键词： Image segmentation Image color analysis Histograms White blood cells Clustering algorithms Shape

来源：评论

学校读者我要写书评

暂无评论

Research on Grounding Fault Detection Method of united adjustment power source system Based on VMD and Comprehensive Energy Relative Entropy

Research on Grounding Fault Detection Method of united adjus...

引用

Chinese Automation Congress (CAC)

作者： Shangbin Jiao Yunpeng Yang Jingwen Guo Sen Li Shaanxi Key Laboratory of Complex System Control and Intelligent Information Processing Xi‘an University of Technology Xi'an China

The paper present a ground fault detection method of united adjustment power source system based on variational mode decomposition (VMD) and comprehensive energy relative entropy. First, the analysis of the system revealed the fault feature extraction is more difficult and the line selection is inaccurate. Secondly, we summarize the fault characteristics of different operation modes. Different operation modes select line by the same selection line method. The proposed method can accurately extract feature in the weak zero sequence current of each feeder, and distinguish the bus fault and feeder fault for each mode of operation. The simulation results show that the proposed method is also suitable for the traditional small current grounding system, and distribution networks with different voltage levels and grounding modes. The results of laboratory simulation experiments show that the method is reliable and practical.

关键词： Entropy Fans Grounding Feature extraction Transient analysis Power generation Generators

来源：评论

学校读者我要写书评

暂无评论

The Novel Sensor Network Structure for Classification processing Based on the Machine Learning Method of the ACGAN

引用

Sensors (Switzerland) 2019年第14期19卷 3145页

作者： Chen, Yuantao Tao, Jiajun Wang, Jin Chen, Xi Yang, Kai School of Computer and Communication Engineering & Hunan Provincial Key Laboratory of Intelligent Processing of Big Data on Transportatio Changsha University of Science and Technology Changsha410114 China School of Information Science and Engineering Fujian University of Technology Fuzhou350118 China Hunan Institute of Scientific and Technical Information Changsha410001 China Electronics & Information School Yangtze University Jingzhou434023 China Technical Quality Department Hunan ZOOMLION Heavy Industry Intelligent Technology Corporation Limited Changsha410005 China

To address the problem of unstable training and poor accuracy in image classification algorithms based on generative adversarial networks (GAN), a novel sensor network structure for classification processing using auxiliary classifier generative adversarial networks (ACGAN) is proposed in this paper. Firstly, the real/fake discrimination of sensor samples in the network has been canceled at the output layer of the discriminative network and only the posterior probability estimation of the sample tag is outputted. Secondly, by regarding the real sensor samples as supervised data and the generative sensor samples as labeled fake data, we have reconstructed the loss function of the generator and discriminator by using the real/fake attributes of sensor samples and the cross-entropy loss function of the label. Thirdly, the pooling and caching method has been introduced into the discriminator to enable more ekective extraction of the classification features. Finally, feature matching has been added to the discriminative network to ensure the diversity of the generative sensor samples. Experimental results have shown that the proposed algorithm (CP-ACGAN) achieves better classification accuracy on the MNIST dataset, CIFAR10 dataset and CIFAR100 dataset than other solutions. Moreover, when compared with the ACGAN and CNN classification algorithms, which have the same deep network structure as CP-ACGAN, the proposed method continues to achieve better classification effects and stability than other main existing sensor solutions. © Sensors (Switzerland) 2019. All Rights Reserved.

关键词： Generative adversarial networks

来源：评论

学校读者我要写书评

暂无评论

PLC based variable structure control of hydraulic position servo system considering controller saturation

PLC based variable structure control of hydraulic position s...

引用

IEEE Conference on Industrial Electronics and Applications (ICIEA)

作者： Hai-Peng Ren Xuan Wang Shaanxi Key Laboratory of Complex System Control and Intelligent Information Processing Xian University of Technology Xian China

Due to the advantages of strong carrying capacity and fast response speed, hydraulic servo system has been widely used in industrial motion control field. The hydraulic servo system is susceptible to the nonlinearity, parameter uncertainty and unknown disturbance, which makes the high performance position control of the hydraulic servo system become a challenge. In this paper, considering the hydraulic position servo control system with input saturation, a sliding mode variable structure controller with shrinking boundary layer is designed. The proposed method is used to experimentally track three kinds of reference signals based on Programmable Logic Controller (PLC), i.e., S7-200PLC, and hydraulic control system provided by Festo company. The experiment results indicate that the proposed method is effective and achieves smaller steady state tracking error as compared with sign function robust item and the conventional PID control.

关键词： Servomotors Cavity resonators Valves Force Touch sensitive screens

来源：评论

学校读者我要写书评

暂无评论

Reversible Natural Language Watermarking Using Synonym Substitution and Arithmetic Coding

引用

Computers, Materials & Continua 2018年第6期55卷 541-559页

作者： Lingyun Xiang Yan Li Wei Hao Peng Yang Xiaobo Shen Hunan Provincial Key Laboratory of Intelligent Processing of Big Data on Transportation Changsha University of Science and TechnologyChangsha 410114HunanChina School of Computer and Communication Engineering Changsha University of Science and TechnologyChangsha 410114HunanChina School of Traffic&Transportation Engineering Changsha University of Science&TechnologyChangsha 410114HunanChina Hunan Branch of CNCERT/CC Changsha 410004HunanChina School of Computer Science and Engineering Nanyang Technological University639798Singapore

For protecting the copyright of a text and recovering its original content harmlessly,this paper proposes a novel reversible natural language watermarking method that combines arithmetic coding and synonym substitution *** analyzing relative frequencies of synonymous words,synonyms employed for carrying payload are quantized into an unbalanced and redundant binary *** quantized binary sequence is compressed by adaptive binary arithmetic coding losslessly to provide a spare for accommodating additional ***,the compressed data appended with the watermark are embedded into the cover text via synonym substitutions in an invertible *** the receiver side,the watermark and compressed data can be extracted by decoding the values of synonyms in the watermarked text,as a result of which the original context can be perfectly recovered by decompressing the extracted compressed data and substituting the replaced synonyms with their original *** results demonstrate that the proposed method can extract the watermark successfully and achieve a lossless recovery of the original ***,it achieves a high embedding capacity.

关键词： Arithmetic coding synonym substitution lossless compression reversible watermarking.

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：