检索结果-内蒙古大学图书馆

Adversarial Attack on Object Detection via Object Feature-Wise Attention and Perturbation Extraction

清华大学学报自然科学版（英文版） 2025年第3期30卷 1174-1189页

作者： Wei Xue Xiaoyan Xia Pengcheng Wan Ping Zhong Xiao Zheng School of Computer Science and Technology Anhui University of TechnologyMaanshan 243032China National Key Laboratory of Science and Technology on Automatic Target Recognition National University of Defense TechnologyChangsha 410073China

Deep neural networks are commonly used in computer vision tasks,but they are vulnerable to adversarial samples,resulting in poor recognition *** traditional algorithms that craft adversarial samples have been effective in attacking classification models,the attacking performance degrades when facing object detection models with more complex *** address this issue better,in this paper we first analyze the mechanism of multi-scale feature extraction of object detection models,and then by constructing the object feature-wise attention module and the perturbation extraction module,a novel adversarial sample generation algorithm for attacking detection models is ***,in the first module,based on the multi-scale feature map,we reduce the range of perturbation and improve the stealthiness of adversarial samples by computing the noise distribution in the object *** in the second module,we feed the noise distribution into the generative adversarial networks to generate adversarial perturbation with strong attack *** doing so,the proposed approach possesses the ability to better confuse the judgment of detection *** carried out on the DroneVehicle dataset show that our method is computationally efficient and works well in attacking detection models measured by qualitative analysis and quantitative analysis.

关键词： adversarial attack transfer attack object detection generative adversarial networks multi-scale feature map

来源：评论

学校读者我要写书评

暂无评论

Dual Space SVDD Based Radar target recognition from HRRP

引用

Chinese Journal of Electronics 2025年第2期23卷 419-425页

作者： Huaitie Xiao Guoyu Feng Yongfeng Zhu Mengjun Huang Science and Technology on Automatic Target Recognition Laboratory National University of Defense Technology Changsha China

In radar target recognition based on kernel method, Support vector data description (SVDD) has been applied to High resolution range profiles (HRRPs) recognition. In this paper, first, three distribution models, i.e., Membership model, Cloud model, Gaussian mixture model, are developed to describe distribution characteristics of HRRPs in extended space. Secondly, test HRRPs in multi-target hypersphere spaces are classified, in accordance with their multi-space distributing characteristics, into two types, i.e., shrink sample and slack sample. Determination of the property of a test sample is achieved by using the minimum relative distance for shrink samples and three distribution models for slack samples, respectively. Thus three HRRP recognition methods based on dual space SVDD are formed. The extensive experiment results for HRRPs of four planes show that the proposed methods have better recognition performance than the recognition method based on single space SVDD.

关键词： target recognition Spaceborne radar Mixture models Vectors Data models Character recognition Kernel Gaussian mixture model Signal to noise ratio

来源：评论

学校读者我要写书评

暂无评论

FPGA implementation of a Bernoulli particle filter

引用

Xi Tong Gong Cheng Yu Dian Zi Ji Shu/Systems Engineering and Electronics 2025年第2期47卷 398-405页

作者： Lian, Hongfei Li, Dongsheng Jiang, Yanwen Fan, Hongqi Xiao, Huaitie Wang, Guoyan National Key Laboratory of Automatic Target Recognition College of Electronic Science and Technology National University of Defense Technology Changsha410073 China

Aiming at the high-speed and efficient computing problem of Bernoulli particle filters in embedded application environments, taking the Bernoulli particle filters for joint detection and estimation of radar weak targets as an example, a function modularized and particle size scalable field programmable gate array (FPGA) implementation architecture is proposed. The computing speed of filtering calculation is further improved through approaches as particle state pipelining, layered accumulation and sum, and parallel resampling, etc. Xilinx ZC706 evaluation board on-board testing experiments have demonstrated the good scalability and excellent acceleration ratio of the proposed architecture. When the number of particles is 1 024, the acceleration ratio is about 104 orders of magnitude compared to the Intel Core i3-4130 CPU computing environment. The results have important reference value for the application of Bernoulli particle filtering technology in radar, robotics, and navigation guidance fields. © 2025 Chinese Institute of Electronics. All rights reserved.

关键词： Electronic guidance systems

来源：评论

学校读者我要写书评

暂无评论

On Adversarial Robust Generalization of DNNs for Remote Sensing Image Classification

引用

IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing 2025年 18卷 11370-11385页

作者： Xue, Wei Wang, Yonghao Zhang, Shaoquan Zheng, Xiao Zhong, Ping Anhui University of Technology School of Computer Science and Technology Maanshan243032 China Nanchang Institute of Technology School of Information Engineering Nanchang330099 China National University of Defense Technology National Key Laboratory of Science and Technology on Automatic Target Recognition Changsha410073 China

Deep neural networks (DNNs)-based deep learning is an important technical support in the task of remote sensing image classification. But DNNs are susceptible to adversarial attacks. Adversarial training is one of the effective ways to obtain robust models that can against such attacks. However, many adversarial training strategies can only defend against known attacks and perform poorly when faced with unknown attacks, namely, although the adversarial robustness of the model is enhanced, the generalization is reduced. To alleviate this phenomenon, we propose a new method to improve the adversarial robust generalization of DNNs for remote sensing image classification within the framework of adversarial training. Specifically speaking, we impose a Euclidean regularization constraint on the gradient of the adversarial loss function in order to achieve a narrowing of the robust generalization gap by enhancing the gradient alignment between clean and adversarial examples. In addition, we incorporate a label smoothing strategy into the adversarial training process, aiming to further improve the adversarial robustness of the model by reducing its sensitivity to subtle fluctuations. The combination of the above two strategies can not only improve the adversarial robustness of the model but also improve its adversarial robust generalization. Finally, in the case of two classical attack approaches FGSM and PGD, we validate the effectiveness and feasibility of our method through extensive experiments on two commonly used remote sensing image classification datasets NWPU-RESSC45, RSSCN7, and WHU-RS19, demonstrating its superiority over previous methods, particularly in the context of improving the adversarial robust generalization of DNNs. © 2025 IEEE.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

Nearshore Underwater target Detection Meets UAV-borne Hyperspectral Remote Sensing: A Novel Hybrid-level Contrastive Learning Framework and Benchmark Dataset

arXiv

引用

arXiv 2025年

作者： Qi, Jiahao Zhou, Chuanhong Liu, Xingyue Chen, Chen Zhu, Dehui Bin, Kangcheng Zhong, Ping National Key Laboratory of Science and Technology on Automatic Target Recognition National University of Defense Technology Changsha410073 China

UAV-borne hyperspectral remote sensing has emerged as a promising approach for underwater target detection (UTD). However, its effectiveness is hindered by spectral distortions in nearshore environments, which compromise the accuracy of traditional hyperspectral UTD (HUTD) methods that rely on bathymetric model. These distortions lead to significant uncertainty in target and background spectra, challenging the detection process. To address this, we propose the Hyperspectral Underwater Contrastive Learning Network (HUCLNet), a novel framework that integrates contrastive learning with a self-paced learning paradigm for robust HUTD in nearshore regions. HUCLNet extracts discriminative features from distorted hyperspectral data through contrastive learning, while the self-paced learning strategy selectively prioritizes the most informative samples. Additionally, a reliability-guided clustering strategy enhances the robustness of learned representations. To evaluate the method effectiveness, we conduct a novel nearshore HUTD benchmark dataset, ATR2-HUTD, covering three diverse scenarios with varying water types and turbidity, and target types. Extensive experiments demonstrate that HUCLNet significantly outperforms state-of-the-art methods. The dataset and code will be publicly available at: https://***/qjh1996/HUTD. Copyright © 2025, The Authors. All rights reserved.

关键词： Unmanned aerial vehicles (UAV)

来源：评论

学校读者我要写书评

暂无评论

Active Object Detection for UAV Remote Sensing via Behavior Cloning and Enhanced Q-Network with Shallow Features

引用

IEEE Transactions on Geoscience and Remote Sensing 2025年

作者： Zou, Zhuocheng Hu, Xikun Zhong, Ping National University of Defense Technology National Key Laboratory of Automatic Target Recognition College of Electronic Science and Technology Changsha410073 China

Object detection in Unmanned Aerial Vehicle (UAV) remote sensing imagery faces two critical challenges: the inability to effectively handle multi-scale targets and the difficulty in addressing partial occlusions, which significantly impact detection accuracy in real-world applications. To overcome these limitations, we propose an active object detection (AOD) method that dynamically adjusts viewing angles and scales during the detection process. Our key innovation lies in a novel architecture that uniquely combines shallow Feature Pyramid Network features with detector output bounding boxes, combining with a self-supervised learning framework. The technical originality of our approach is further enhanced by our two-stage learning methodology, which initially employs behavior cloning to establish robust foundational performance, followed by Q-learning fine-tuning to optimize detection strategies. To facilitate comprehensive evaluation, we introduce CARLA-AOD, a new benchmark dataset that encompasses 18 diverse scenarios across hemispheric space above targets, specifically designed for UAV remote sensing AOD applications. Extensive experimental validation demonstrates the effectiveness of our approach, achieving substantial improvements over baseline detectors across multiple datasets: 7.9% on Small Airport, 14.0% on Virtual Park, and 10.2% on our CARLA-AOD dataset. The two-stage learning process proves particularly effective, with Q-learning fine-tuning providing an additional performance boost of up to 7.3% beyond the behavior cloning baseline. Moreover, our method achieves a 60.2% reduction in inference time compared to the state-of-the-art DCCL algorithm, making it particularly suitable for time-critical applications such as emergency response and real-time surveillance. The dataset is available at IEEE Dataport. IEEE Dataport. © 2025 IEEE.

关键词： Self-supervised learning

来源：评论

学校读者我要写书评

暂无评论

Joint calibration algorithm for gain-phase and mutual coupling errors in uniform linear array

引用

Chinese Journal of Aeronautics 2016年第4期29卷 1065-1073页

作者： Li Weixing Lin Jianzhi Zhang Yue Chen Zengping Science and Technology on Automatic Target Recognition Laboratory National University of Defense Technology

The effect of gain-phase perturbations and mutual coupling significantly degrades the performance of digital array radar (DAR). This paper investigates array calibration problems in the scenario where the true locations of auxiliary sources deviate from nominal values but the angle intervals are known. A practical algorithm is proposed to jointly calibrate gain-phase errors and mutual coupling errors. Firstly, a simplified model of the distortion matrix is developed based on its special structure in uniform linear array (ULA). Then the model is employed to derive the precise locations of the auxiliary sources by one-dimension search. Finally, the least-squares estimation of the distortion matrix is obtained. The algorithm has the potential of achieving considerable improvement in calibration accuracy due to the reduction of unknown parameters. In addition, the algorithm is feasible for practical applications, since it requires only one auxiliary source with the help of rotation platforms. Simulation results demonstrate the validity, robustness and high performance of the proposed algorithm. Experiments were carried out using an S-band DAR test-bed. The results of measured data show that the proposed algorithm is practical and effective in application. (C) 2016 Production and hosting by Elsevier Ltd. on behalf of Chinese Society of Aeronautics and Astronautics.

关键词： Calibration accuracy Digital array Gain and phase errors Least-squares Mutual coupling Uniform linear array

来源：评论

学校读者我要写书评

暂无评论

Bistatic weak target detection method using non-cooperative air surveillance radar

引用

Journal of Systems Engineering and Electronics 2015年第5期26卷 954-963页

作者： Dinghe Wang Qinglong Bao Ruiqi Tian Zengping Chen Science and Technology on Automatic Target Recognition Laboratory National University of Defense Technology

At present, most of the passive radar system researches utilize FM radios, TV broadcasts, navigation satellites,etc. as illuminators. The transmitted signals are not specifically designed radar waveforms. In this work, the frequency agile, phased array air surveillance radar(ASR) is used as the illuminator of opportunity to detect the weak target. The phased array technology can help realize beam agility to track targets from different aspects simultaneously. The frequency agility technology is widely employed in radar system design to increase the ability of anti-jamming and increase the detection probability. While the frequency bandwidth of radar signals is usually wide and the range resolution is high, the range cell migration effect is obvious during the long time integration of non-cooperative bistatic radar. In this context, coherent integration methods are not applicable. In this work, a parametric non-coherent integration algorithm based on task de-interweaving is proposed. Numerical experiments verify that this is effective in weak target detection.

关键词： air surveillance radar non-cooperative bistatic task de-interweave

来源：评论

学校读者我要写书评

暂无评论

Infrared modeling and imaging simulation of midcourse ballistic targets based on strap-down platform

引用

Journal of Systems Engineering and Electronics 2014年第5期25卷 776-785页

作者： Changzhen Qiu Zhiyong Zhang Huanzhang Lu Kaifeng Zhang Science and Technology on Automatic Target Recognition Laboratory National University of Defense Technology

An infrared （IR） imaging simulation framework based on the strap-down platform is proposed for midcourse ballistic targets. It overcomes the shortcoming of the existing algorithms, which cannot simulate IR imaging from the entire midcourse process. The proposed framework includes three steps, target characteristic modeling, motion modeling, and imaging modeling. In imaging modeling, the staring focal plane is taken into account due to its wide employment. In order to obtain IR images of high fidelity, especially that the fluctuation of the target signal-to-noise ratio （SNR） is reasonably similar to the actual one, this paper proposes an improved IR imaging simulation method. The proposed method considers two critical factors of the pixel plane, occupy-empty ratio and defect elements, which affect the imaging of targets markedly but are neglected in previous work. Finally, the IR image sequence of high fidelity is obtained. And the correlative parameters of simulation can be set according to the given scene. Thus the generated images can satisfy the needs of algorithms validation for tracking and recognition.

关键词： strap-down platform midceurse ballistic target infrared (IR) staring focal plane.

来源：评论

学校读者我要写书评

暂无评论

SMC-PHD based multi-target track-before-detect with nonstandard point observations model

引用

Journal of Central South University 2015年第1期22卷 232-240页

作者：占荣辉高彦钊胡杰民张军 Science and Technology on Automatic Target Recognition Laboratory National University of Defense Technology

Detection and tracking of multi-target with unknown and varying number is a challenging issue, especially under the condition of low signal-to-noise ratio(SNR). A modified multi-target track-before-detect(TBD) method was proposed to tackle this issue using a nonstandard point observation model. The method was developed from sequential Monte Carlo(SMC)-based probability hypothesis density(PHD) filter, and it was implemented by modifying the original calculation in update weights of the particles and by adopting an adaptive particle sampling strategy. To efficiently execute the SMC-PHD based TBD method, a fast implementation approach was also presented by partitioning the particles into multiple subsets according to their position coordinates in 2D resolution cells of the sensor. Simulation results show the effectiveness of the proposed method for time-varying multi-target tracking using raw observation data.

关键词： adaptive particle sampling multi-target track-before-detect probability hypothesis density(PHD) filter sequential Monte Carlo(SMC) method

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：