检索结果-内蒙古大学图书馆

arXiv 2019年

作者： Cui, Yuqi Wu, Dongrui Key Laboratory of the Ministry of Education for Image Processing and Intelligent Control School of Artificial Intelligence and Automation Huazhong University of Science and Technology Wuhan430074 China

Takagi-Sugeno-Kang (TSK) fuzzy systems are flexible and interpretable machine learning models;however, they may not be easily applicable to big data problems, especially when the size and the dimensionality of the data are both large. This paper proposes a mini-batch gradient descent (MBGD) based algorithm to efficiently and effectively train TSK fuzzy systems for big data classification problems. It integrates three novel techniques: 1) uniform regularization (UR), which is a regularization term added to the loss function to make sure the rules have similar average firing levels, and hence better generalization performance;2) random percentile initialization (RPI), which initializes the membership function parameters efficiently and reliably;and, 3) batch normalization (BN), which extends BN from deep neural networks to TSK fuzzy systems to speedup the convergence and improve generalization. Experiments on nine datasets from various application domains, with varying size and feature dimensionality, demonstrated that each of UR, RPI and BN has its own unique advantages, and integrating all three together can achieve the best classification performance. Copyright © 2019, The Authors. All rights reserved.

关键词： Gradient methods

来源：评论

学校读者我要写书评

暂无评论

3D Ultrasound Data Acqusition System Based on Back End Scan Mode

3D Ultrasound Data Acqusition System Based on Back End Scan ...

引用

International Conference on intelligent Computation and Bio-Medical Instrumentation (ICBMI)

作者： Jian Qi Mingyue Ding Ming Yuchi Key Laboratory of Image Processing and Intelligent Control of Education Ministry of China School of Life Science and Technology Huazhong University of Science and Technology Wuhan China

The data acquisition of 3D-Ultrasound includes array scan and mechanical scan, and the later one is more easy to realize. Currently, the traditional probe scanning mode is Front-end scanning. Under the above scanning mode, when it scans over the breast, 2D-Ultrasound probe slides through the surface of the patients' bodies, the image will be influenced strongly by the human *** this paper we propose a new scanning mode to solve the above problem: Back-End scan, the back end rotates while the front end contacts the patients' skin without slide. The device designed using dual stepper motors which are under the synchronization control. Experiment results show the effectiveness of the proposed device.

关键词： Probes Ultrasonic imaging Synchronous motors Three dimensional displays Gears Ribs Phantoms

来源：评论

学校读者我要写书评

暂无评论

MULTI-ROBOT TRACKING OF MOBILE TARGET BASED ON COMMUNICATION

引用

IFAC Proceedings Volumes 2008年第2期41卷 10397-10402页

作者： Lei Liu Yongji Wang Dept. of Control Science and Engineering Huazhong Univ. of Sci. and Tech. Wuhan Hubei China 430074 Key Laboratory of Image Processing and Intelligent Control Ministry of Education Wuhan Hubei China 430074

Multi-robot tracking of mobile target is studied in the paper, which is based on the communication and sensors. For an independent tracking robot, the processes are separated into three layers and four tasks, and allocated to different robots for distinct roles in tracking, which is named the Distributed Decision control System (DDCS). After that, two tracking models, centralized and distributed models, are designed for multi-robot tracking. Furthermore, a Proportional Navigation Guidance Law (PNGL) and l-ϕ formation control algorithm are mentioned to realize the robot motion control. At last the simulation has shown the feasibility and validity of both models.

关键词： Multi-robot tracking DDCS PNGL Communication based l-ϕ formation control algorithm

来源：评论

学校读者我要写书评

暂无评论

Transmit Aperture Function for Large Depth Focusing Combined with Phase Coherence Imaging for Interference Suppressing

Transmit Aperture Function for Large Depth Focusing Combined...

引用

International Conference on intelligent Computation and Bio-Medical Instrumentation (ICBMI)

作者： Mengling Xu Yimin Chen Ming Yuchi Mingyue Ding School of Life Science and Technology Key Laboratory of Image Processing and Intelligent Control of Education Ministry of China Huazhong University of Science & Technology Wuhan China

Dynamic focus and dynamic aperture applied on both transmit and receive aperture can provide the best possible quality of the B-mode ultrasound image, which is called a "gold standard" image. However, they are implemented only on the receive beam in most medical ultrasound systems because the application of dynamic focus on the transmit beam will reduce the frame rate severely, which has great medical importance especially when tissue motion needs to be considered. Recently, a specific transmit aperture function has been proposed to achieve uniform transmit energy distribution on the axis over a large depth without reducing the frame rate. Although it can maintain the uniformity of the transmit energy on the axis, the energy distribution off the axis of the transmit beam is not satisfactory, meaning a wider main lobe width and higher side lobe levels compared with some general aperture functions such as rectangular and hamming functions. To suppress the interference signals originating from the targets on the side and grating lobes, we processed the received echo signals using the phase coherence imaging method (PCI), which is an adaptive receive aperture weighting technique presented in the recent literatures. Experimental results show that the uniform sensitivity along the axis range has been achieved and image quality on spatial and contrast resolution improved when we use the transmit aperture function combined with PCI.

关键词： Apertures Ultrasonic imaging Coherence Wires Biomedical imaging image resolution

来源：评论

学校读者我要写书评

暂无评论

image Retrieval Using the Modified Intersecting Cortical Model

Image Retrieval Using the Modified Intersecting Cortical Mod...

引用

International Conference on intelligent Computation and Bio-Medical Instrumentation (ICBMI)

作者： Wenjin Yuan Xuming Zhang School of Life Science and Technology Key Laboratory of Image Processing and Intelligent Control of Education Ministry of China Huazhong University of Science and Technology Wuhan China

The intersecting cortical model(ICM) is driven from the pulse-coupled neural networks (PCNN), and a lot of works have been done on it. In this paper, we have proposed a method, which is modified from the PCNN, to retrieval the image. This method uses the iteration times of the ICM as the feature, and we do it repeatedly. So we will get the feature vectors and utilize the features to retrieval the image. And the experimental results have proved the effectiveness of the proposed method. While the proposed method shows some advantages like the feature curves are linear and steady.

关键词： Neurons Mathematical model Feature extraction Computational modeling Equations Time series analysis Firing

来源：评论

学校读者我要写书评

暂无评论

Single upper limb pose estimation method based on improved stacked hourglass network

arXiv

引用

arXiv 2020年

作者： Peng, Gang Zheng, Yuezhi Li, Jianfeng Yang, Jin Deng, Zhonghua Key Laboratory of Image Processing and Intelligent Control Ministry of Education School of Artificial Intelligence and Automation Huazhong University of Science and Technology Wuhan430074 China

At present, most high-accuracy single-person pose estimation methods have high computational complexity and insufficient real-time performance due to the complex structure of the network model. However, a single-person pose estimation method with high real-time performance also needs to improve its accuracy due to the simple structure of the network model. It is currently difficult to achieve both high accuracy and real-time performance in single-person pose estimation. For use in human–machine cooperative operations, this paper proposes a single-person upper limb pose estimation method based on an end-to-end approach for accurate and real-time limb pose estimation. Using the stacked hourglass network model, a single-person upper limb skeleton key point detection model was designed. Deconvolution was employed to replace the up-sampling operation of the hourglass module in the original model, solving the problem of rough feature maps. Integral regression was used to calculate the position coordinates of key points of the skeleton, reducing quantization errors and calculations. Experiments showed that the developed single-person upper limb skeleton key point detection model achieves high accuracy and that the pose estimation method based on the end-to-end approach provides high accuracy and real-time performance. Copyright © 2020, The Authors. All rights reserved.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Late Better than Early: A Decision-Level Information Fusion Approach for Rgb-Thermal Crowd Counting with Illumination Awareness

SSRN

引用

SSRN 2024年

作者： Cheng, Jian Feng, Chen Xiao, Yang Cao, Zhiguo Key Laboratory of Image Processing and Intelligent Control Ministry of Education School of Artificial Intelligence and Automation Huazhong University of Science and Technology Wuhan430074 China

In this paper, we make the first research effort to address the RGB-Thermal (RGB-T) crowd counting problem with decision-level late fusion manner. Being different from the existing pixel-level or feature-level fusion methods, our proposition chooses to fuse the density maps yielded by RGB and thermal counterparts via spatially adaptive weighting with RGB illumination-aware attention. Our key intuition to conduct RGB-T density map fusion lies in 3 main folders. First, compared with the raw RGB-T images or convolutional feature maps, RGB-T density maps contain stronger counting-wise semantic meanings. Secondly, they are also of high spatial resolution for revealing fine local details. Thirdly, RGB-T density maps are complementary to each other for effective counting pattern characterization. To fuse them adaptively, a spatial weighting map is generated concerning the issues of RGB illumination awareness and local counting pattern characterization ability jointly. To the best of our knowledge, we are the first to leverage RGB-T crowd counting via concerning these 2 issues in a unified way. Meanwhile, cross-modality feature interaction is conducted between RGB and thermal modalities to facilitate spatial weighting map generation. The experiments on 2 well-established RGB-T crowd counting datasets, i.e., RGBT-CC and DroneRGBT) verify the superiority of our proposition. The source code and pretrained models will be released upon acceptance at https://***/hustaia/DLF-IA. © 2024, The Authors. All rights reserved.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

BokehMe: When Neural Rendering Meets Classical Rendering

arXiv

引用

arXiv 2022年

作者： Peng, Juewen Cao, Zhiguo Luo, Xianrui Lu, Hao Xian, Ke Zhang, Jianming Key Laboratory of Image Processing and Intelligent Control Ministry of Education School of Artificial Intelligence and Automation Huazhong University of Science and Technology China Adobe Research

We propose BokehMe, a hybrid bokeh rendering framework that marries a neural renderer with a classical physically motivated renderer. Given a single image and a potentially imperfect disparity map, BokehMe generates high-resolution photo-realistic bokeh effects with adjustable blur size, focal plane, and aperture shape. To this end, we analyze the errors from the classical scattering-based method and derive a formulation to calculate an error map. Based on this formulation, we implement the classical renderer by a scattering-based method and propose a two-stage neural renderer to fix the erroneous areas from the classical renderer. The neural renderer employs a dynamic multi-scale scheme to efficiently handle arbitrary blur sizes, and it is trained to handle imperfect disparity input. Experiments show that our method compares favorably against previous methods on both synthetic image data and real image data with predicted disparity. A user study is further conducted to validate the advantage of our method. Copyright © 2022, The Authors. All rights reserved.

关键词： Rendering (computer graphics)

来源：评论

学校读者我要写书评

暂无评论

Monocular visual–inertial SLAM algorithm combined with wheel speed anomaly detection

arXiv

引用

arXiv 2020年

作者： Gang, Peng Zezao, Lu Shanliang, Chen Bocheng, Chen Dingxin, He Key Laboratory of Image Processing and Intelligent Control Ministry of Education School of Artificial Intelligence and Automation Huazhong University of Science and Technology Wuhan430074 China

To address the weak observability of monocular visual–inertial odometers on ground-based mobile robots, this paper proposes a monocular inertial SLAM algorithm combined with wheel speed anomaly detection. The algorithm uses a wheel speed odometer pre-integration method to add the wheel speed measurement to the least-squares problem in a tightly coupled manner. For abnormal motion situations, such as skidding and abduction, this paper adopts the Mecanum mobile chassis control method, based on torque control. This method use the motion constraint error to estimate the reliability of the wheel speed measurement. At the same time, in order to prevent incorrect chassis speed measurements from negatively influencing robot pose estimation, this paper uses three methods to detect abnormal chassis movement and analyze chassis movement status in real time. When the chassis movement is determined to be abnormal, the wheel odometer pre-integration measurement of the current frame is removed from the state estimation equation, thereby ensuring the accuracy and robustness of the state estimation. Experimental results show that the accuracy and robustness of the method in this paper are better than those of a monocular visual–inertial odometer. Copyright © 2020, The Authors. All rights reserved.

关键词： Anomaly detection

来源：评论

学校读者我要写书评

暂无评论

Multiple Unmanned Surface Vehicle Coverage control in Dynamic Environments

Multiple Unmanned Surface Vehicle Coverage Control in Dynami...

引用

Neuromorphic Computing (ICNC), International Conference on

作者： Hong Chen Dingxin He Huantao Mao Yong Yang Zhi-Wei Liu School of Artificial Intelligence and Automation Key Laboratory of Image Processing and Intelligent Control of Education Ministry of China Huazhong University of Science and Technology Wuhan China

This paper investigates the area coverage problem for multiple Unmanned Surface Vehicles (USVs) in dynamic environments. Multiple USVs are employed to execute the coverage task using the Voronoi Partition method for task allocation. Parameter adaptive laws are employed to estimate the parameters of the sensing function and the Voronoi cell centroids for updating the target positions in dynamic environments. Additionally, considering the influence of water surface disturbances, a nonholonomic underactuated model analysis is conducted for the USV under non-full constraints, resulting in a motion strategy for USVs. This enables the multiple USVs to effectively cover the task area. Finally, simulation results are presented to validate the effectiveness of the coverage scheme designed in this paper and the ability of USVs to effectively cover dynamic environments.

关键词：

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：