检索结果-内蒙古大学图书馆

EMOST: A dual-branch hybrid network for medical image fusion via efficient model module and sparse transformer

Computers in Biology and Medicine 2024年 179卷 108771-108771页

作者： Wang, Wenqing He, Ji Liu, Han School of Automation and Information Engineering Xi'an University of Technology Xi'an710048 China Shaanxi Key Laboratory of Complex System Control and Intelligent Information Processing Xi'an University of Technology Xi'an710048 China

Multimodal medical image fusion fuses images with different modalities and provides more comprehensive and integrated diagnostic information. However, current multimodal image fusion methods cannot effectively model non-local contextual feature relationships, and due to direct aggregation of the extracted features, they introduce unnecessary implicit noise into the fused images. To solve the above problems, this paper proposes a novel dual-branch hybrid fusion network called EMOST for medical image fusion that combines a convolutional neural network (CNN) and a transformer. First, to extract more comprehensive feature information, an effective feature extraction module is proposed, which consists of an efficient dense block (EDB), an attention module (AM), a multiscale convolution block (MCB), and three sparse transformer blocks (STB). Meanwhile, a lightweight efficient model (EMO) is used in the feature extraction module to exploit the efficiency of the CNN with the dynamic modeling capability of the transformer. Additionally, the STB is incorporated to adaptively maintain the most useful self-attention values and remove as much redundant noise as possible by developing the top-k selection operator. Moreover, a novel feature fusion rule is designed to efficiently integrate the features. Experiments are conducted on four types of multimodal medical images. The proposed method shows higher performance than the art-of-the-state methods in terms of quantitative and qualitative evaluations. The code of the proposed method is available at https://***/XUTauto/EMOST. © 2024 Elsevier Ltd

关键词： Feature extraction

来源：评论

学校读者我要写书评

暂无评论

Robust adaptive neural network consensus tracking control of multi-robot systems

Robust adaptive neural network consensus tracking control of...

引用

第33届中国控制与决策会议

作者： Huijun Guo Jintao Liang School of Automation and Information Engineering Xi'an University of Technology Key Laboratory of Shaanxi Province for Complex System Control and Intelligent Information Processing Guilin University of Electronic Technology

This paper studies the problem of distributed consensus tracking control for multi-robot systems with different friction coefficients and external disturbances based on the multi-agent *** the case where the communication topology is a weighted directed graph and leader node is the neighbor of a small portion of the follower *** follower nodes are modeled as two second-order nonlinear dynamical system with coupling parts.A distributed robust adaptive control law based on the neural networks(NN) is designed for each follower node,where the controller requires only relative state information from its adjacent *** such control scheme,the states of all follower nodes ultimately synchronize to the leader node with bounded residual *** results are provided to validate the effectiveness of the algorithm.

关键词： Consensus control Multi-robot systems Agent systems Neural networks Sliding mode control

来源：评论

学校读者我要写书评

暂无评论

Event-triggered consensus for linear continuous-time multi-agent systems based on a predictor

引用

Information Sciences 2018年 459卷 278-289页

作者： Jian Sun Qiuling Yang Xiaoyu Liu Jie Chen School of Automation Beijing Institute of Technology Beijing 100081 PR China Key Laboratory of Intelligent Control and Decision of Complex System Beijing Institute of Technology Beijing 100081 PR China China North Industrial Group Corporation Limited China Ordnance Industry Navigation and Control Technology Institute Beijing 100089 PR China

In this paper, the problem of an event-triggered consensus for a linear continuous-time multi-agent system is investigated. A new event-triggered consensus protocol based on a predictor is proposed to achieve consensus while not requiring continuous communication among agents. The predictor utilizes an artificial closed-loop system to predict the future state of each agent. With the proposed consensus protocol, each agent only needs to monitor its own states to determine its event-triggered instants. When an event of an agent is triggered, the agent immediately updates its consensus protocol and sends its state information to its neighbors. When an agent receives state information from its neighbors, the agent immediately updates its consensus protocol and predictor. A necessary and sufficient condition that solves the consensus problem is derived. Moreover, it is proved that Zeno behaviors are excluded. Finally, some numerical examples are given to illustrate that, with the proposed protocol, a multi-agent system can achieve consensus while greatly reducing event-triggered times.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Mobile robot's electronic compass calibration based on modified Fourier Neural Network

Mobile robot's electronic compass calibration based on modif...

引用

IEEE Conference on Robotics, automation and Mechatronics

作者： Gong Kun Deng Fang Ma Tao Gong Kun is with School of Automation Beijing Institute of Technology and Key Laboratory of Advanced Control of Iron and Steel Process (Ministry of Education) Beijing China Deng Fang is with School of Automation Beijing Institute of Technology and Key Laboratory of Advanced Control of Iron and Steel Process (Ministry of Education) Beijing China Ma Tao is with School of Automation Beijing Institute of Technology and Key Laboratory of Complex System Intelligent Control and Decision Ministry of Education Beijing China

In order to improve the precision of the azimuth measured by mobile robot's electronic compass, this paper proposes a new calibration method based on Fourier Neural Network trained by Modified Particle Swarm Optimization (MPSO-FNN). This method makes use of Fourier Neural Network (FNN) to establish the error compensation model of electronic compass's azimuth, and introduces Modified Particle Swarm Optimization (MPSO) algorithm to optimize the weights of neural network. Thus the comparatively accurate error model of azimuth is obtained to compensate the output of electronic compass. This method not only has strong nonlinear approximation capability, but also overcomes the neural networks' shortcomings which are too slow convergence speed, oscillation, and easy to fall into local optimum and sensitive to the initial values. Experimental results demonstrate that after calibrated by this method, the range of azimuth error reduces to -0.35°~0.70° from -3.4°~25.2°, and the average value of absolute error is only 0.30°.

关键词： Compass Azimuth Biological neural networks Particle swarm optimization Training Accuracy automation

来源：评论

学校读者我要写书评

暂无评论

Adaptive Back-stepping Based control for Active Suspension systems with Prescribed Performance

Adaptive Back-stepping Based Control for Active Suspension S...

引用

第40届中国控制会议

This paper proposes an adaptive back-stepping tracking control with prescribed performance function for vehicle suspension systems with uncertain parameters. The prescribed performance function can lead the tracking error from a large set to an arbitrarily small residual set. Firstly, the constrained tracking error system incorporating the prescribed performance functions is transformed into unconstrained form. Then an adaptive updating scheme based on the projection algorithm is designed to estimate the uncertain parameter. Following that an adaptive nonlinear controller is designed using the back-stepping technique, and the stability of whole controlled systems is proved by the Lyapunov theory. Meanwhile, the state convergence of the internal dynamical system is also guaranteed. Finally, simulation results are provided to verify the proposed approach.

关键词： Active Suspension system Adaptive control Back-stepping technique prescribed performance

来源：评论

学校读者我要写书评

暂无评论

A 3D Display Parallel system: Light Field Re-rendering and Depth Sense Optimization

A 3D Display Parallel System: Light Field Re-rendering and D...

引用

International Conference on Display Technology, ICDT 2018

作者： Pei, Renjing Ma, Kui Wang, Feiyue State Key Laboratory of Management and Control for Complex Systems Institute of Automation Chinese Academy of Sciences East Zhongguanchun Road No. 95 Haidian District Beijing China School of Computer and Control Engineering University of Chinese Academy of Sciences Beijing China Parallel Optics Technology Innovation Center Qingdao Academy of Intelligent Industries Qingdao China

ISBN: (纸本)9781510867659

The main benefit of 3D display over 2D display is the obvious ability to create a more lifelike character with high depth sense. However, the limitation of human eye’s visual mechanism, unartful 3D scene structure design, or bad viewing condition always emerges poor depth perception experience or even physiological discomfort during the watching time, which is often sub-optimal for mass high-quality 3D display productions. To solve this problem, we propose a novel 3D display parallel system for depth sense optimization and it empirically guides how the light field should be re-rendered. Structurally, the parallel system consists of an artificial perception measurement system, a display evaluation model and a light field display rendering system, which includes the display calibration, scene capture, light field data processing and display. Particularly, the system can systematically analyze and model various factors affecting the depth sense which learned through the measurement system, like scene structure, objects’ speeds in 3D video and so on. And those sense factors can be personally modified or increased according to the viewer’s demands or technical improvement. Moreover, the light field could be real-time re-rendered, based on some image processing technology, optical flow analysis and object segmentation (or tracking) (especially the one-shot video segmentation). Theory and algorithms are developed and experimental validation results show a superior performance. © 2018 SID.

关键词： Three dimensional displays

来源：评论

学校读者我要写书评

暂无评论

Local entropy difference-fast marching method for leak diagnosis and localization

Local entropy difference-fast marching method for leak diagn...

引用

IEEE/ASME (AIM) International Conference on Advanced intelligent Mechatronics

作者： Ling Zhao Tao Wang Wei Fan Nan Ge Tong Zhao Guangzheng Peng School of Automation Complicated System of Intelligent Control and Decision Key Laboratory of Education Ministry Beijing Institute of Technology Beijing China Department of Engineering Physics Tsinghua University Beijing China

In this paper, an effective technique for leak diagnosis and localization which couples the local entropy difference method (LEDM) to the fast marching method (FMM) is proposed. A computer-based system is described to combine a specific pneumatic circuit with an on-line infrared camera which is used to test the air-tightness of a test vessel. The test object is charged with cryogenic compressed air and its temperature field is acquired by infrared thermography in this system. Then, both the leakage point and its edge can be identified by means of the designed LED-FM method. An experiment is conducted that confirms that the leak diagnosis and localization method presented in this paper is effective in diagnosing leakage points and in extracting accurate leakage edges, in spite of the presence of surface defects.

关键词： Entropy Pixel Image edge detection Level set Cameras Equations Probes

来源：评论

学校读者我要写书评

暂无评论

引用

IEEE International Conference on Mechatronics and automation

作者： Ying Wu Qiang Huang Xuechao Chen Zhangguo Yu Libo Meng Gan Ma Peisen Zhang Weimin Zhang Intelligent Robotics Institute Beijing Institute of Technology Beijing China State Key Laboratory of Intelligent Control and Decision of Complex System Beijing China International Joint Research Laboratory of Biomimetic Robots and Systems Ministry of Education Beijing China

ISBN: (纸本)9781479970995

This paper introduces the similarity evaluation on humanoid facial expression. For the purpose of understanding easily, firstly, the process of designing humanoid robot's facial expressions is explored in brief. Then, a similarity function is proposed to evaluate the similarity of the robot facial motion compared with the human being quantitatively based on both the kinematics and dynamics factors, also including how to select the key value of parameter in similarity function. Finally, the effectiveness of the similarity function is confirmed by an experiment of facial expressions based on BHR-4 robot.

关键词： Joints Kinematics Dynamics Trajectory Humanoid robots Rhythm

来源：评论

学校读者我要写书评

暂无评论

Power system Events Classification Technology Based on Deep-Learning

Power System Events Classification Technology Based on Deep-...

引用

IEEE International Symposium on Circuits and systems (ISCAS)

作者： Xin Lei Hongwei Ma Bin Liu Zhen Li School of Automation and Key Laboratory for Intelligent Control & Decision on Complex Systems Beijing Institute of Technology Beijing China College of Information Science and Engineering China University of Petroleum (Beijing) Beijing China

ISBN: (数字)9798350330991

ISBN: (纸本)9798350331004

As the complexity of the power system continues to increase, the frequency of the power system anomalies is on the rise. These anomalies have significant and widespread impacts on the stability of the power grid. Therefore, the rapid and accurate classification of these anomalies is crucial in preventing their further propagation and mitigating potential economic losses. This study presents an algorithm based on Phasor Measurement Unit (PMU) data for monitoring the state of power systems and identifying the types of anomalies. First, a dataset for anomaly event classification is created based on PMU data, which is used to train and validate the anomaly event classification model. Subsequently, a robust anomaly event classification model is constructed, consisting of a residual module with one-dimensional Convolutional Neural Networks (CNN) and a cascaded fully connected neural network classifier. This algorithm has undergone rigorous testing in the IEEE New England 39 bus test system, demonstrating exceptional event recognition accuracy.

关键词： Accuracy Power system stability Propagation losses Phasor measurement units Data models Stability analysis Classification algorithms

来源：评论

学校读者我要写书评

暂无评论

Real-time moving object detection and tracking based on improved mean shift

引用

Zhongnan Daxue Xuebao (Ziran Kexue Ban)/Journal of Central South University (Science and Technology) 2011年第SUPPL. 1期42卷 823-828页

作者： Lei, Ying-Hui Zhang, Juan Yang, Wen-Jia Key Laboratory of Complex System Intelligent Control and Decision Ministry of Education School of Automation Beijing Institute of Technology Beijing 100081 China

An improved real-time target detection and tracking method was proposed based on moving foreground object in the servo monitoring system. This method extracts moving object based on adaptive mixture of Gaussian when the object comes into the video scene, then tracks the moving object using improved MeanShift algorithm, and makes it in the center of the scene. The algorithm not only ensures the real-timing of the detection and tracking, but also enlarges the sight of the camera when the object is tracked. The experiment results show that this method can automatically detect moving object and do servo tracking.

关键词： Bandpass filters

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：