检索结果-内蒙古大学图书馆

IEEE Transactions on Artificial Intelligence 2024年第6期5卷 3072-3082页

作者： Zhang, Chunjie Cao, Ping Bai, Huihui Zheng, Xiaolong Beijing Jiaotong University Institute of Information Science Beijing100044 China Beijing Jiaotong University Beijing Key Laboratory of Advanced Information Science and Network Technology Beijing100044 China University of Chinese Academy of Sciences State Key Laboratory of Multimodal Artificial Intelligence Systems The State of Key Laboratory of Management and Control for Complex System Institute of Automation Beijing100190 China University of Chinese Academy of Sciences School of Artificial Intelligence Beijing100190 China

Pairwise similarity has been widely used for image classification by propagating the class information from labeled images to unlabeled images and predicting the classes of unlabeled images accordingly. Although widely used, pairwise similarity based classification methods have two drawbacks. On one hand, visual-semantic similarity consistency does not always hold due to semantic gap. On the other hand, the reliability of pairwise similarity is also influenced by the number and underlying distribution of labeled images. A well-designed classification model cannot be easily generalized to few-shot classification tasks. To solve these two problems, in this article, we propose a semantic coherence guided multiview similarity (SCG-MV) for image classification method with varied supervision levels. We measure the semantic coherence between image pairs for classifier learning. This is achieved by first conducting line search along the line segment between each image pair, and then measure the semantics of each interval. The similarity of one image pair can then be obtained by aggregating the semantic coherence of all the intervals. In this way, we can model image similarities more finely and discriminatively. To make use of multiview information, we also use the semantic coherence guided image similarities to learn multiview classifiers. The proposed semantic coherence guided method can be combined with other similarity based methods for image classification with various levels of supervision. We conduct both few-shot and fully supervised classification experiments on four datasets with the results well prove the discriminative power and generalization ability of the proposed method. © 2020 IEEE.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

Distributed Multi-Sensor Multi-View Fusion of PHD Filter for Maneuvering Targets

Distributed Multi-Sensor Multi-View Fusion of PHD Filter for...

引用

IEEE International Conference on Radar

作者： Cong Peng Lei Chai Wei Yi Linxiu Chen School of Information and Communication Engineering University of Electronic Science and Technology of China Chengdu China Science and Technology on Complex System Control and Intelligent Agent Cooperation Laboratory Beijing China

ISBN: (纸本)9781665468893

This paper addresses the problem of maneuvering multi-target tracking by a network of sensors having different and limited fields of view (FoV s). Each local sensor runs the Gaussian Mixture Probability Hypothetical Density (GMPHD) filter. Due to the target maneuver, based on a single motion model, severe tracking performance degradation can be observed. We propose to use multi-model (MM) method to realize the adaptation to motion characteristic so as to overcome maneuverability of targets. Then considering FoV s of sensors in the network are different and limited, the standard weighted arithmetic average (WAA) fusion is no longer applicable and leads to an underestimation of the target number. Therefore, we use state-dependent WAA (SD-WAA) fusion rule, which performs the WAA in a more robust way by calculating a set of state-related fusion weights. Numerical experiment is designed to demonstrate the efficacy of the proposed method.

关键词： Degradation Adaptation models Target tracking Fuses Radar Sensor fusion Sensor phenomena and characterization

来源：评论

学校读者我要写书评

暂无评论

A GPR-Aided Method for Collaborative Relative Navigation for Multiple Unmanned Aerial Vehicles During Ranging Measurement Outages

A GPR-Aided Method for Collaborative Relative Navigation for...

引用

International Symposium on Autonomous systems (ISAS)

作者： Xin Ding Mingrui Hao Baichun Gong Linxiu Chen Zhixing Zhuang Nanjing University of Aeronautics and Astronautics Nanjing China Science and Technology on Complex System Control and Intelligent Agent Cooperation Laboratory Beijing China

To maintain reliable relative navigation accuracy of unmanned aerial vehicles (UAVs) during range measurement outages, a GPR-aided ranging method is proposed to assist the extended Kalman filter for measurement updates. Specifically, when the on-board ranging measurement is available, a Gaussian process regression (GPR) algorithm is used to train the mapping model between acceleration, angular velocity of the inertial navigation system (INS), relative position, and the ranging distance. When ranging measurement is unavailable, the trained GPR algorithm is used for distance prediction, replacing the ranging measurement and being used for consensus extended Kalman filter (CEKF) measurement update to reduce the degree of relative navigation error divergence. Numerical simulation was conducted with three UAVs performing cooperative flight, the results show that the proposed algorithm can provide reliable distance prediction during ranging outage and effectively reduce the divergence of relative state estimation error.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A Novel Relocating Recognition Algorithm in Overlapping Region for Multi-UAVs Cluster

A Novel Relocating Recognition Algorithm in Overlapping Regi...

引用

2022 IEEE International Conference on Unmanned systems, ICUS 2022

作者： Li, Chen Qi, Xuelei Zhao, Juntao Liao, Xuanhong Lin, Zitian Ma, Hongjun Liang, Xinkai College of Information Science and Engineering Northeastern University Shenyang China Hong Kong University of Science and Technology Department of Electronic and Computer Engineering Hong Kong Hong Kong High School Attached to Northeast Normal University International Curriculum Center Changchun China School of Automation Science and Engineering South China University of Technology Guangzhou China Beijing Electro-Mechanical Engineering Institute Science and Technology on Complex System Control and Intelligent Agent Cooperation Laboratory Beijing China

ISBN: (数字)9781665484565

ISBN: (纸本)9781665484565

How improve the mapping efficiency and location accuracy of the multi-UAV cluster based on the distributed SLAM technology is a significant problem in overlapping regions. Therefore, this paper mainly proposes a novel relocation recognition algorithm for the multi-UAV cluster. First, the autonomous localization algorithm based on visual-inertial fusion is proposed. Then, the algorithm for vocabulary updating and relocating is designed. Finally, practical testing is implemented to verify and improve the efficiency of exploration ability and the accuracy of the autonomous navigation and positioning in the complex overlapping regions. © 2022 IEEE.

关键词： Unmanned aerial vehicles (UAV)

来源：评论

学校读者我要写书评

暂无评论

Monocular Visual-Inertial Odometry Based on Point and Line Features

Monocular Visual-Inertial Odometry Based on Point and Line F...

引用

International Conference on Computer and Communications (ICCC)

作者： Gongyan Chen Hai Zhang Wenxing Fu Yiqun Qiao School of Automation Science and Electrical Engineering Beihang University Beijing China Science and Technology on Complex System Control and Intelligent Agent Cooperation Laboratory Beijing China

ISBN: (数字)9781728186351

ISBN: (纸本)9781728186368

In order to solve the problem that the point feature tracking is not robust enough to reduce the accuracy of the system in a low-texture environment, this paper proposes a visual inertial odometry system based on point-line features (PL-VINS). Firstly, the system improves the extraction and matching algorithm of line features to increase the speed and accuracy of the algorithm; then the line features are added to the front-end tracking and back-end optimization. Finally, the EuRoc public data set is used as the experimental object to prove the effectiveness of the algorithm.

关键词： Visualization Conferences Feature extraction Cameras Real-time systems Optimization Optical flow

来源：评论

学校读者我要写书评

暂无评论

An Improved Robust Adaptive Kalman Filtering Algorithm

An Improved Robust Adaptive Kalman Filtering Algorithm

引用

第三十八届中国控制会议

作者： Liuyang Jiang Wenxing Fu Hai Zhang Zheng Li Longyun Chi School of Automation Science and Electrical Engineering Beihang University Science and Technology on Complex System Control and Intelligent Agent Cooperation Laboratory National Airspace Management Center

Robust adaptive Kalman filter(RAKF) is a method that can not only compensate for the model errors but also resist the measurement outliers. The method reduces the influence of the measurement outliers on the filtering system by reducing the weight of the abnormal observation. After the processing of robust estimation, all the errors are attributed to the inaccuracy of the model. Since the robust estimation is not theoretically quantitative that it can only reduce the observation error to a certain extent, the remaining observation errors are compensated as the model errors, resulting in new estimation errors. To address this problem, a method for detecting the model error is proposed as a trigger condition for RAKF to compensate for the model error,and this constitutes an improved robust adaptive Kalman filter(IRAKF) algorithm. The new algorithm performs the model errors compensation stricter than the original method. The efficacy of the improved algorithm is demonstrated via a traffic running testing with GPS/INS tight integration navigation system. Simulation results indicate that the improved algorithm can effectively reduce the inaccurate adaptive procedure and obtain relatively accurate and stable filtering results.

关键词： Adaptive Kalman Filter Robust Estimation Improved Model Errors Tight Integrated Navigation

来源：评论

学校读者我要写书评

暂无评论

Line-point feature based structure-preserving image stitching

Line-point feature based structure-preserving image stitchin...

引用

Chinese Automation Congress (CAC)

作者： Longyun Chi Xujun Guan Xiaorong Shen Hai Zhang School of Automation Science and Electrical Engineering Beihang University Beijing China Science and Technology on Complex System Control and Intelligent Agent Cooperation Laboratory Beijing China

The naturalness of the panoramic image is an important criterion for evaluating the result of image stitching. When the overlapping region of the image has a low-texture area, the alignment is hard to accomplish due to incorrect and inadequate feature point matching pairs, and the salient structures are easily destroyed due to the unreasonable position of the optimal seam. To solve the problem, we use line segment matching information to assist the alignment of low-texture regions and use superpixel segmentation regions to filter the unreliable point correspondences. An improved seam-finding algorithm is finally used to blend images together. The experimental results demonstrate that the proposed method can get more natural stitching results.

关键词： Image segmentation Strain Feature extraction Information filters Matched filters Optimization

来源：评论

学校读者我要写书评

暂无评论

Improved Res2Net Based on Multi-scale Attention Mechanism for Breast Cancer Image Classification 12th

Improved Res2Net Based on Multi-scale Attention Mechanism fo...

引用

12th International Conference on Computer Engineering and Networks, CENet 2022

作者： Lei, Suqin Deng, Xing Shao, Haijian Xia, Wang Department of Computer Jiangsu University of Science and Technology Jiangsu Zhenjiang212003 China College of Automation Key Laboratory of Ministry of Education of Complex Engineering System Measurement and Control Southeast University Jiangsu Nanjing210009 China School of Information Science and Technology Nantong University Nantong Jiangsu226019 China Department of Electrical and Computer Engineering National University of Singapore Singapore117583 Singapore

ISBN: (纸本)9789811969003

Breast cancer is one of the most common cancers among female diseases. Since the classification accuracy of pathological images is crucial to the diagnosis of breast cancer, in order to reduce the error of manual diagnosis and improve the automatic classification accuracy of pathological images, this paper proposes a high-accuracy breast cancer classification method based on multi-scale attention mechanism enhanced Res2Net. By constructing a multi-channel residual connection in the residual block, and introducing a convolutional attention mechanism to enhance the feature extraction ability of Res2Net, and using group convolution to replace part of the ordinary convolution. In addition, since the channel information interaction across groups cannot be realized in the group convolution, which reduces the feature extraction ability. In this paper, the channel shuffling operation is used to reorganize the feature maps after the group convolution to achieve the information interaction between different groups. The evaluation results on the BreaKHis dataset show that, the accuracy and precision of the method in this paper have reached 98.5% and 98.9% respectively, in binary classification of histopathological images compared with the benchmark method, and the advantages of recall rate and F1 value are also more obvious. © 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Image classification

来源：评论

学校读者我要写书评

暂无评论

Multi-Stage Marine Ship Recognition Based on Stackable Residual Network

Multi-Stage Marine Ship Recognition Based on Stackable Resid...

引用

International Conference on Fuzzy systems and Knowledge Discovery (FSKD)

作者： Wanwan Yu Junping Zhang Linlin Wang Yunqiao Xi Hongfeng Xu Linxiu Chen Mingrui Hao School of Electronics and Information Engineering Harbin Institute of Technology Harbin China Science and Technology on Complex System Control and Intelligent Agent Cooperation Laboratory Beijing China

Marine ship recognition has always been an important research field. The ships can be usually recognized by analyzing their attribute characteristics. However, in the actual process of recognition, many ship recognition methods using a single model cannot meet the requirements of this task due to the complexity of ship types, diversity of interference factors and different information obtained from distance differences. Therefore, we propose a multi-stage ship recognition scheme. First, the single radiation emitter feature recognition network model is designed for long-distance recognition. Second, the fusion feature network model for short distance recognition. The proposed network is mainly constructed with the residual units, which not only simplify the training of deep networks and facilitate the spread of information, but also can solve the vanishing gradient and exploding gradient problem. The simulation experimental results show that the proposed network model has better performance than typical methods and has great potential in applications.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Gradient-direction-based Rectangles and Triangles Traffic Signs Detection Algorithm in Natural Scenes*

Gradient-direction-based Rectangles and Triangles Traffic Si...

引用

Chinese Automation Congress (CAC)

作者： YanQiong Jin XuJun Guan Hai Zhang School of Automation Science and Electrical Engineering Beihang University Beijing China Science and Technology on Complex System Control and Intelligent Agent Cooperation Laboratory Beijing China Science and Technology on Aircraft Control Laboratory Beihang University Beijing China

In this paper, we propose a novel method to detect rectangular and triangular traffic signs in natural scenes according to the geometric features of the rectangle and triangle traffic signs. It is a Gradient-direction-based method that mainly includes four stages: gradient information classification, filter templates-based line segments with different orientation extraction, structural corners detection and rectangle/triangle detection. The extracted corners are categorized into different sets according to the orientation of the line segments, by which we can quickly extract rectangle and triangle traffic signs in image. And the experimental results demonstrate under different conditions are shown that the proposed method can detect rectangular and triangular traffic signs in natural scenes efficiently and robustness.

关键词： Image segmentation Feature extraction Image edge detection Image color analysis Detection algorithms Shape Gray-scale

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：