检索结果-内蒙古大学图书馆

Facial expression recognition based on fusion of extended LDP and Gabor features

The Journal of China Universities of Posts and Telecommunications 2018年第1期25卷 48-53页

作者： Luo Yuan Yu Chaojing Zhang Yi Wang Boyu Key Laboratory of Optoelectronic Information Sensing and Transmission Technology Chongqing University of Posts and Telecommunications Engineering Research Center for Information Accessibility and Service Robots Chongqing University of Posts and Telecommunications

The local directional pattern （LDP） is unsusceptible to random noise which is widely used in texture extraction of face region. LDP cannot encode the central pixel thus the important information will be lost. Thus a new feature descriptor called extended local directional pattern （ELDP） is proposed for face extraction. First, the mean value of the eight directional edge response values and the gray value of center pixel are calculated. Second, the mean value is taken as the threshold. Then, the expression image is encoded using nine encoded values. In order to reduce redundant information and get more effective information, the Gabor filter is used to obtain the multi- direction Gabor magnitude maps （GMMs） , and then the ELDP is used to encode the GMMs. Finally, support vector machine （SVM） is applied to classify and recognize facial expression. The experimental results show that the feature dimensions is greatly reduced and the rate of facial expression recognition is improved.

关键词： facial expression recognition local directional pattern ELDP Gabor

来源：评论

学校读者我要写书评

暂无评论

A novel fusion method of improved adaptive LTP and two-directional two-dimensional PCA for face feature extraction

引用

Optoelectronics Letters 2018年第2期14卷 143-147页

作者：罗元王薄宇张毅赵立明 Key Laboratory of Optoelectronic Information Sensing and Technology Chongqing University of Posts and Telecommunications Chongqing 400065 China Engineering Research Center for Information Accessibility and Service Robots Chongqing University of Posts and Telecommunications Chongqing 400065 China

In this paper, under different illuminations and random noises, focusing on the local texture feature's defects of a face image that cannot be completely described because the threshold of local ternary pattern(LTP) cannot be calculated adaptively, a local three-value model of improved adaptive local ternary pattern(IALTP) is proposed. Firstly, the difference function between the center pixel and the neighborhood pixel weight is established to obtain the statistical characteristics of the central pixel and the neighborhood pixel. Secondly, the adaptively gradient descent iterative function is established to calculate the difference coefficient which is defined to be the threshold of the IALTP operator. Finally, the mean and standard deviation of the pixel weight of the local region are used as the coding mode of IALTP. In order to reflect the overall properties of the face and reduce the dimension of features, the two-directional two-dimensional PCA((2D)~2 PCA) is adopted. The IALTP is used to extract local texture features of eyes and mouth area. After combining the global features and local features, the fusion features(IALTP+) are obtained. The experimental results on the Extended Yale B and AR standard face databases indicate that under different illuminations and random noises, the algorithm proposed in this paper is more robust than others, and the feature's dimension is smaller. The shortest running time reaches 0.329 6 s, and the highest recognition rate reaches 97.39%.

关键词： PCA LTP

来源：评论

学校读者我要写书评

暂无评论

A Multi-AGV scheduling planning method based on improved GA

引用

Journal of Physics: Conference Series 2020年第2期1550卷

作者： Yi Zhang Mengsi Li China Chongqing Information Accessibility and Service Robot Technology Research Center Chongqing University of Posts and Telecommunications Chongqing 400065 China

The path planning of Multi-AGV does not especially consider the power consumption. Therefore, a new Multi-AGV scheduling planning method based on improved GA(Genetic Algorithm) is proposed. The improvement of this algorithm is showed in three aspects: first, in order to ensure that AGVs have the power to perform the task, the operation of power evaluation is added after each iteration is completed; second, the objective function will have two constraints: first, the minimum total power consumption for all AGV; second, the maximum power consumption of a single AGV is minimal; finally, by changing the mutation operator of the GA, the convergence speed and the convergence of the population are improved. The simulation results show that the improved GA scheduling results are more reasonable, and under the same conditions, AGVs power consumption will be smaller.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Edge Detection Algorithm of Image Fusion Based on Improved Sobel Operator 3

Edge Detection Algorithm of Image Fusion Based on Improved S...

引用

2017 IEEE 第3届信息技术与机电一体化工程国际学术会议(ITOEC 2017)

作者： Yi Zhang Xiaoyuan Han Han Zhang Liming Zhao Chongqing Information Accessibility and Service Robot Technology Research Center Chongqing University of Post and Telecommunication

The traditional Sobel operator has low edge localization accuracy,poor noise and edge detection *** order to solve these problems,this paper proposes an image fusion algorithm combining fusi

ISBN: (纸本)9781509053643;9781509053636

The traditional Sobel operator has low edge localization accuracy,poor noise and edge detection *** order to solve these problems,this paper proposes an image fusion algorithm combining fusi

关键词： Image edge detection Image fusion Smoothing methods Manganese

来源：评论

学校读者我要写书评

暂无评论

research on Power-Assisted Strategy and Device Based on Muscle Synergy

Research on Power-Assisted Strategy and Device Based on Musc...

引用

IEEE International Conference on Computational Intelligence and Virtual Environments for Measurement Systems and Applications

作者： Lintao Hu Xiuying Luo Shangjie Tang Xiaoying Wu Lin Chen Xiaolin Zheng Wensheng Hou Key Laboratory of Biorheological Science and Technology of Ministry of Education Collaborative Innovation Center for Brain Science Chongqing University Chongqing Key Laboratory of Artificial Intelligence and Service Robot Control Technology Chongqing China Chongqing Medical Electronics Engineering Technology Research Center Chongqing University

ISBN: (数字)9781538683446

ISBN: (纸本)9781538683453

Surface EMG (sEMG) signals are non-invasive means of recording muscle activity that reflect the activation of human skeletal muscles. The purpose of this research is to study a dynamic assisted strategy based on muscle synergy, and to design an upper limb motion assisted device to achieve different assist tasks. Eleven healthy participants were recruited for the study. The participants were asked to perform grasping tasks at 9 target locations in the space and sEMG signals of the eight involved muscles were recorded. The non-negative matrix factorization (NMF) algorithm was applied to extract muscle coordination information during each corresponding experimental task. According to the muscle coordination information of all the participants' sEMG signals extracted by NMF decomposition, the corresponding upper limb motion tasks were decoded. The proposed method can decode the movement pattern of the human arm by considering the mapping relationship between the muscle coordination information and the joint motion, which may provide less effortful control of the robotic exoskeleton for rehabilitation training of individuals with neurological disorders or arm impairment.

关键词： Exoskeletons Muscles Task analysis Motion control Human computer interaction robot sensing systems Man-machine systems

来源：评论

学校读者我要写书评

暂无评论

Design of Visual Gaze Target Locating Device Based on Depth Camera

Design of Visual Gaze Target Locating Device Based on Depth ...

引用

IEEE International Conference on Computational Intelligence and Virtual Environments for Measurement Systems and Applications

作者： Yuxin. Liu Lintao. Hu Bowen. Xiao Xiao Y. Wu Yanhan. Chen Donghai. Ye Wen S. Hou Xiaolin. Zheng Key Laboratory of Biorheological Science and Technology of Ministry of Education & Collaborative Innovation Center for Brain Science Chongqing University Chongqing Key Laboratory of Artificial Intelligence and Service Robot Control Technology College of art Chongqing University Chongqing China Chongqing Medical Electronics Engineering Technology Research Center

ISBN: (数字)9781538683446

ISBN: (纸本)9781538683453

Based on the research of machine vision positioning application in recent years, this paper discusses the common machine vision positioning system and its principle, and designs a visual target positioning device to detect people's target intention. It mainly includes dynamic following and target positioning functions. Dynamic following can determine the fixation direction of the tester, and target positioning can obtain the spatial position information of the fixation target. The device can recognize the intention of people by observing the visual fixation direction of the subject and obtain the spatial position information of the target, which is a human-computer interaction path with a widely used scene.

关键词： Cameras Visualization Neck Microprocessors Target recognition Head

来源：评论

学校读者我要写书评

暂无评论

research on multi-base depth neural network speech recognition 2

Research on multi-base depth neural network speech recogniti...

引用

2nd IEEE Advanced information technology, Electronic and Automation Control Conference, IAEAC 2017

作者： Jun, Cai Fei, Li Yi, Zhang Yu, Liu College of Automation Chongqing University of Posts and Telecommunications Chongqing400065 China Chongqing Information Accessibility and Service Robot Technology Research Center Chongqing400065 China School of Advanced Manufacturing Engineering Chongqing University of Posts and Telecommunications Chongqing400065 China

ISBN: (纸本)9781467389778

In speech recognition system, an improved multi-base neural network speech recognition model is proposed to solve the problem of long learning time and slow convergence rate of deep neural network. However, the improved model introduces a large number of parameters in the training process to make the model over-fitted in the test set, resulting in the deterioration of generalization ability and the decrease of speech recognition rate. In order to solve this problem, an improved BP algorithm is introduced in the last layer of the depth neural network model. The experimental results show that the multi-base neural network model not only reduces the training time and calculation amount, but also improves the recognition accuracy compared with other recognition models, and it is an excellent speech recognition model. © 2017 IEEE.

关键词： Speech recognition

来源：评论

学校读者我要写书评

暂无评论

research on Multi-base Depth Neural Network Speech Recognition

Research on Multi-base Depth Neural Network Speech Recogniti...

引用

2017 IEEE 第2届先进信息技术,电子与自动化控制国际会议(IAEAC 2017)

作者： Cai Jun Li Fei Zhang Yi Liu Yu College of Automation Chongqing University of Posts and Telecommunications Chongqing Information Accessibility and Service Robot Technology Research Center School of Advanced Manufacturing Engineering Chongqing University of Posts and Telecommunications

In speech recognition system, an improved multibase neural network speech recognition model is proposed to solve the problem of long learning time and slow convergence rate of deep neural network. How

ISBN: (纸本)9781467389808

关键词： Speech recognition Depth neural network Over-fitting BP algorithm Accuracy Speech recognition TEST SETS Neural network speech recognition system Neural Networks (Computer)

来源：评论

学校读者我要写书评

暂无评论

An improved online SVM algorithm for overcoming the influence of muscle fatigue in sEMG based human machine interaction

An improved online SVM algorithm for overcoming the influenc...

引用

2014 2nd International Conference on Mechatronics, robotics and Automation, ICMRA 2014

作者： Cai, Jun Liu, Rui National Engineering Research and Development Center for Information Accessibility Research Center of Intelligent System and Robot Chongqing University of Posts and Telecommunications Chongqing 400065 China

ISBN: (纸本)9783038350781

For the problem that the stability of surface Electromyographic EMG(sEMG) based human-machine interface(HMI) declines as the muscle fatigue takes place, an improved incremental training algorithm for online support vector machine(SVM) is proposed. This paper study the changes of sEMG when muscle fatigue occurs by the method of continuous wavelet transform, and then apply the improved online SVM for sEMG classification. The novel method adjusts the parameters of SVM model to adapt itself based on the changes of sEMG signals and the training data is conditionally selected and forgot. Experiment results show that the presented algorithm performs high modeling precision and training speed is increased. Furthermore, this method effectively overcomes the influence of muscle fatigue during long-term operating sEMG based HMI. © (2014) Trans Tech Publications, Switzerland.

关键词： Support vector machines

来源：评论

学校读者我要写书评

暂无评论

EEG Feature Extraction Method Fused Wavelet Transform and Sample Entropy

EEG Feature Extraction Method Fused Wavelet Transform and Sa...

引用

2012 International Conference on Medical Physics and Biomedical engineering(ICMPBE 2012)

作者： Yi Zhang Mingwei Luo Yuan Luo Engineering Research & Development Center of Information Accessibility Research Center of Intelligent System and RobotChongqing university of posts and telecommunications

Considering the issue of low recognition rate for motor imagery EEG signal by using current single feature extraction method,a feature extraction method fused wavelet transform and sample entropy is presented in this paper.A method which is used to measure the discrimination of the left-right hands motor imagery EEG signals is proposed and the extracted feature values using wavelet transform and sample entropy are scaled by average feature value of training *** fusion feature value is defined by weighting the scaled feature values using discrimination and summing of ***,the left-right hands motor imagery EEG signals are classified by support vector machine *** experimental results show that the feature extraction method fused wavelet transform and sample entropy outperforms the way only using wavelet transform or sample entropy in the highest recognition rate,mean recognition rate and standard deviation,which verify the effectiveness and superiority of this method.

关键词： EEG,Wavelet transform,Sample entropy,Support vector machine

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：