检索结果-内蒙古大学图书馆

Biological Eagle-Eye-Based Correlation Filter Learning for Fast UAV Tracking

IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT 2024年 73卷 1页

作者： Wu, Tongyan Duan, Haibin Zeng, Zhigang Beihang Univ Sch Automat Sci & Elect Engn State Key Lab Virtual Real Technol & Syst Beijing 100083 Peoples R China Huazhong Univ Sci & Technol Sch Artificial Intelligence & Automat Wuhan 430074 Peoples R China

As a cutting-edge tracking algorithm for unmanned aerial vehicles (UAVs), discriminative correlation filter (DCF)-based trackers have evolved by embracing diverse feature extraction and computational methods. However, many DCF-based trackers struggle to adapt to background variations encountered during UAV flights due to the lack of image feature enhancement and velocity-dependent scale adjustment. The high-speed movement of UAVs requires trackers that can leverage fast computational capabilities to keep pace. Taking inspiration from the impressive hunting skills of eagles, this article introduces a DCF-based tracker called eagle-eye-based correlation filter (EECF). The EECF tracker utilizes a "dual fovea" mechanism for image preprocessing, enhancing the utilization of target information in candidate regions. By integrating color opponent and contrast adaptation into the feature extraction process, the EECF tracker emulates the perceptual characteristics of eagle vision, enabling robust target recognition in complex backgrounds. We introduce masking and regularization techniques based on the "lateral inhibition" mechanism observed in eagle visual pathways. The EECF tracker showcases exceptional computational speed and surpasses other trackers in tracking success rates. Extensive experiments carried out on four publicly available UAV tracking benchmarks and custom drone datasets confirm the superiority of our tracker over other DCF-based options. Our algorithm achieves a maximum tracking speed of 86 FPS while maintaining higher accuracy than most comparative algorithms.

关键词： Target tracking Image color analysis Feature extraction Autonomous aerial vehicles filtering algorithms Tracking Correlation Bionic perception correlation filter dynamic tracking eagle eye unmanned aerial vehicle (UAV)

来源：评论

学校读者我要写书评

暂无评论

A fault diagnosis method for ship shaft system based on empirical wavelet transform and particle filtering

A fault diagnosis method for ship shaft system based on empi...

引用

Electrical Engineering, Big Data and algorithms (EEBDA), IEEE International Conference on

作者： Yangyang He Chao Ni Science and Technology on Electromagnetic Compatibility Laboratory Wuhan China China Ship Development and Design Center Wuhan China

In order to solve the problems of excessive redundant information and low fault feature extraction rate of ship shafting vibration signal, a fault feature extraction method based on empirical wavelet transform and particle filter is proposed. Firstly, the original signal is processed by empirical wavelet transform to separate the redundant vibration components, and solve the defects such as end effect and modal aliasing inherent in empirical mode decomposition. Secondly, intrinsic mode functions are filtered and reconstructed based on kurtosis, correlation coefficient and energy ratio to highlight fault information and improve signal-to-noise ratio. Finally, particle filter is applied to the reconstructed fault feature components to eliminate the residual non-linear and non-Gaussian noise in EWT, and the fault type is analyzed by the envelope spectrum. Through the analysis of the experimental bearing signal, the feature extraction ability and noise suppression ability of the method are verified, which can be effectively applied to the fault diagnosis of the ship shaft system.

关键词： Wavelet transforms Vibrations Shafts Fault diagnosis filtering algorithms Feature extraction Information filters

来源：评论

学校读者我要写书评

暂无评论

A Frequency-Domain Recursive Least-Squares Adaptive filtering Algorithm Based On A Kronecker Product Decomposition

A Frequency-Domain Recursive Least-Squares Adaptive Filterin...

引用

International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

作者： Hongsen He Jingdong Chen Jacob Benesty Yi Yu School of Information Engineering Southwest University of Science and Technology Mianyang China CIAIC and Shaanxi Provincial Key Laboratory of Artificial Intelligence Northwestern Polytechnical University Xi’an China INRS-EMT University of Quebec Montreal Canada

This paper proposes a frequency-domain recursive least-squares (RLS) adaptive filtering algorithm for identifying time-varying acoustic systems in noisy environments. The Kronecker product (KP) is employed to decompose the model filter of the acoustic channel impulse response into two sets of short sub-filters, based on which a generalized frequency-domain signal model and the associated cost function are established. A KP based RLS algorithm is subsequently deduced. In comparison with the conventional frequency-domain RLS adaptive filter, the presented algorithm is not only computationally more efficient, but also has a faster convergence rate for the identification of acoustic systems regardless of whether the excitation is a white sequence or a speech signal.

关键词： Adaptation models Adaptive systems Frequency-domain analysis Signal processing algorithms Adaptive filters filtering algorithms Cost function

来源：评论

学校读者我要写书评

暂无评论

A Single-pass Noise Covariance Estimation Algorithm in Multiple-model Adaptive Kalman filtering

A Single-pass Noise Covariance Estimation Algorithm in Multi...

引用

IEEE Conference on Aerospace

作者： Hee–Seung Kim Adam Bienkowski Krishna R. Pattipati Department of ECE University of Connecticut Storrs CT

Estimation of unknown noise covariances in a Kalman filter is a problem of significant practical interest in a wide array of industrial applications. The reliable algorithms for their estimation are scarce, and the necessary and sufficient conditions for the identifiability of the covariances were in dispute until very recently. In this paper, we present a single-pass stochastic gradient descent (SGD) algorithm for noise covariance estimation for use in multiple-model adaptive Kalman filters. The multiple-model estimation algorithm assumes that the system obeys one of a finite number of models, and each model has its own dynamics. The overall estimate of the multiple-model approach is a convex combination of the estimates from the filters running in parallel based on the individual models, where the weights of the convex combination are the concomitant posterior model probabilities. Here, the identifiability conditions of the multiple-model approach depend on each candidate model since the corresponding Kalman filters are non-interacting. The proposed streaming algorithm reads measurement data exactly once and has an acceptable root mean square error (RMSE) of state estimates. The proposed estimating algorithm is implemented based on its single-pass nature, recursive fading memory estimation of the sample cross-correlations of the innovations, and the root mean square propagation (RMSprop) accelerated SGD algorithm. Our approach enforces the structural assumptions on unknown noise covariances, and ensures the symmetry and positive definiteness of the estimated covariance matrices. The proposed method is applicable to non-stationary systems where the noise covariances continuously change. The validation of the proposed method on several numerical examples demonstrates its accuracy and filter consistency.

关键词： Adaptation models Technological innovation Sufficient conditions Heuristic algorithms Estimation Stochastic processes filtering algorithms

来源：评论

学校读者我要写书评

暂无评论

On Effect of Adversaries in a Cooperative MCvD Network and filtering Mechanism

On Effect of Adversaries in a Cooperative MCvD Network and F...

引用

IEEE Conference on Wireless Communications and Networking

作者： Tamoghno Nath Adrish Banerjee Department of Electrical Engineering India Institute of Technology Kanpur

In this paper, we have considered a cooperative Molecular Communication via Diffusion (MCvD) network in the presence of adversaries. We have mathematically modelled the adversarial receiver distribution and analyzed its effect on the system performance. The fusion center (FC) gathers all the local information from the receivers and applies the OR or AND rule to the local decisions to arrive at a global decision. We have computed the bit error rate (BER) performance and proposed "η-min filtering" and "η-max filtering" algorithms to improve the system reliability by reducing the BER. It is also shown that the FC chooses the proper filtering algorithm depending on the adversarial receiver distribution and the logical operation to get the best performance.

关键词： filtering Computational modeling System performance Bit error rate Molecular communication Receivers filtering algorithms

来源：评论

学校读者我要写书评

暂无评论

Switching Kronecker Product Linear filtering for Multispeaker Adaptive Speech Dereverberation

Switching Kronecker Product Linear Filtering for Multispeake...

引用

International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

作者： Gongping Huang Jacob Benesty Israel Cohen Emil Winebrand Jingdong Chen Walter Kellermann LMS University Erlangen-Nuremberg Erlangen Germany INRS-EMT University of Quebec Montreal QC Canada Technion–Israel Institute of Technology Haifa Israel Insoundz Ltd. Tel Aviv Israel Ciaic Northwestern Polytechnical University Xi’an China

Dereverberation, a process to mitigate or eliminate the reverberation effect, plays an important role in hands-free speech communication and human-machine interfaces. Tremendous efforts have been devoted to this problem and various methods have been developed over the last three decades. Those methods generally assume that there is only a single speaker in the acoustic environment and, consequently, they suffer from significant performance degradation if multiple speakers participate in the conversation. How to deal with reverberation in multiple-speaker scenarios is still a challenging problem, which is studied in this work. We present a switching multichannel linear prediction filtering method, which designs multiple linear filters with each tracking one speaker. When some speaker is active, the corresponding filter and the weighted cross-correlation matrix are updated while the other filters are kept unchanged. To further improve the performance and reduce complexity, we apply the Kronecker product to decompose every linear prediction filter into a Kronecker product of two shorter filters: one is time-invariant and the other is time-varying. The former is estimated with a batch method (using only a few seconds of speech signal when the corresponding speaker starts to talk in the entire conversation) while a recursive least-squares algorithm is derived for identifying the time-varying set of Kronecker filters.

关键词： Maximum likelihood detection filtering Signal processing algorithms Nonlinear filters Switches Oral communication filtering algorithms

来源：评论

学校读者我要写书评

暂无评论

Least Mean Square Adaptive Filter Detection Based on Wavelet Transform in Spectrum Monitoring

Least Mean Square Adaptive Filter Detection Based on Wavelet...

引用

Photonics and Electromagnetics Research Symposium (PIERS)

作者： He, Xin Zhang, Yonghui Chen, Zhenjia Wang, Lihui Hainan Univ Haikou Sch Informat & Commun Engn Haikou Hainan Peoples R China

ISBN: (纸本)9781665460231

The difficulty of spectrum detection and signal source perceivement is growing because of the increase in electromagnetic interference. Due to the complexity of the electromagnetic environment and the diversity of system components, micro or even weak signals sometimes submerge in noise and interference, difficult to detect. Aiming at a problem that the signal is not easy to detect, a signal detection algorithm based on electromagnetic spectrum monitoring combined with adaptive filtering method is proposed. In terms of theoretical verification, simulation analysis using analog simulation platform, and the algorithm performs performance testing. The main research contents of this paper are as follows: LMS algorithm simulation experiments are performed on the sinusoidal cycle signal. In response to the limitations of the fixed step LMS algorithm on convergence speed, the wavelet transform is used to perform frequency band separation, and the wavelet transform method is combined with the LMS algorithm to perform signal detection simulation. The simulation results show that the proposition combined with the wavelet transform and the LMS algorithm can detect useful signals. The convergence speed of the adaptive ltering algorithm is effectively improved compared to the LMS algorithm before the improvement. And the scheme can reduce steady state disorders, while effective elimination of interference on the spectrum, enhanced noise elimination capability.

关键词： Wavelet transforms Adaptation models Analytical models Wavelet domain filtering algorithms Steady-state Electromagnetics

来源：评论

学校读者我要写书评

暂无评论

Adaptive Combination of Two Multi-Sample Multiband-Structured Subband Adaptive Filters

Adaptive Combination of Two Multi-Sample Multiband-Structure...

引用

IEEE International Symposium on Circuits and Systems (ISCAS)

作者： Peng, Yishu Zhang, Sheng Zheng, Wei Xing Southwest Jiaotong Univ Sch Informat Sci & Technol Chengdu 611756 Sichuan Peoples R China Western Sydney Univ Sch Comp Data & Math Sci Sydney NSW 2751 Australia

ISBN: (纸本)9781665484855

To address the conflict caused by the fixed sampled period in the multi-sampled multiband-structured subband adaptive filter (MS-MSAF), the adaptive convex combination of two MS-MSAFs is proposed in this paper, in which the individual filters independently run with different sampled periods. Moreover, the mean behavior of the convex-combined two MS-MSAF algorithm is also studied. In addition, the convex-combined two MS-MSAF algorithm with periodic feedback is also developed to further enhance the convergence performance. Finally, the validity of the proposed adaptive filtering algorithms together with their theoretical analyses is supported by computer simulations.

关键词： Circuits and systems Computer simulation Computational modeling Adaptive filters filtering algorithms System identification Behavioral sciences

来源：评论

学校读者我要写书评

暂无评论

Fusing WiFi Signals and Camera for Driver Activity Recognition based on Deep Learning 19

Fusing WiFi Signals and Camera for Driver Activity Recogniti...

引用

19th IEEE International Conference on Mobile Ad Hoc and Smart Systems (MASS)

作者： Bai, Yunhao Chen, Guoyu Wang, Xiaorui Ohio State Univ Columbus OH 43210 USA

ISBN: (数字)9781665471800

ISBN: (纸本)9781665471800

In-vehicle camera has been widely adopted for driver activity recognition. Unfortunately, camera-based solutions have some limitations, like brightness condition, blurriness caused by bumpy roads, and angle of view. To this end, some recent studies have resorted to other emerging recognition technologies that show great promise, such as the Channel State Information (CSI) of WiFi signals. However, it is well-known that WiFi signals can be interfered significantly by other nearby WiFi systems, such as neighboring vehicles on the same road. Hence, it is important to address this challenge for WiFi-based solutions to become more practical. In this paper, we propose DriveFusion that overcomes the limitations of both WiFi-based solutions and in-vehicle camera by fusing their inputs based on deep learning. Specifically, DriveFusion flexibly extracts useful information from both systems by having 1) a novel non-reference CSI signal quality estimator based on network analysis, 2) a learning-based signal filter to exclude low-quality features, and 3) an online feature combination module that assigns weights to the two inputs dynamically, based on their signal qualities that can change rapidly. Our results in a real car with commercial WiFi devices show that DriveFusion achieves a recognition accuracy of 91.3% when the deployed WiFi interference is relatively strong, which outperforms state-of-the-art baselines by 13.4% on average.

关键词： Deep learning Roads Interference Activity recognition filtering algorithms Cameras Feature extraction

来源：评论

学校读者我要写书评

暂无评论

Subspace-based Feature Alignment for Unsupervised Domain Adaptation

Subspace-based Feature Alignment for Unsupervised Domain Ada...

引用

IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)

作者： Yi, Eojindl Kim, Junmo Korea Adv Inst Sci & Technol KAIST Daejeon South Korea

ISBN: (数字)9781665479271

ISBN: (纸本)9781665479271

Autonomous agents need to perceive the world in a robust way, such that the shift in data distribution does not lead to faulty perception results. When agents cannot be trained with abundant data, agents may need to operate on real world environments while trained on simulated data, and suffer from domain shift. This paper proposes an effective and robust unsupervised domain adaptation (UDA) method that can resolve these situations. In the UDA setup, we are given a labeled source domain and an unlabeled target domain that share the same set of classes but are sampled from different distributions. This domain shift prevents agents which employ deep neural networks from generalizing well on the target domain. Recent methods adopt the strategy of self-training the networks with pseudo labeled target samples. However, falsely labeled samples cause negative transfer and deteriorate generalization of a network. To reduce negative transfer we propose an algorithm that can filter the pseudo labels, and use the filtered labels to align the domains in the feature space. The samples whose labels have not passed the filtering process can be used as an index to tune the hyperparameters of our method. Across various benchmarks, we validate the performance of our method. Especially, our method achieves strong performance on the synthetic-to-real adaptation scenario.

关键词： Deep learning filtering Neural networks Prototypes filtering algorithms Benchmark testing Autonomous agents

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：