检索结果-内蒙古大学图书馆

Simplified Set-Membership Augmented Affine Projection Generalized Maximum Complex Correntropy Criterion Algorithm

IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS II-EXPRESS BRIEFS 2024年第7期71卷 3623-3627页

作者： Zhao, Haiquan Long, Xiaoqiang Hou, Xinyan Wang, Chen Southwest Jiaotong Univ Sch Elect Engn Minist Educ Chengdu 610031 Peoples R China Southwest Jiaotong Univ Key Lab Magnet Suspens Technol & Maglev Vehicle Minist Educ Chengdu 610031 Peoples R China

In order to cope with the situation, the performance of the set-membership augmented affine projection algorithm deteriorates dramatically when the output signal contains impact noise, this brief applies the maximum complex correntropy criterion to the set-membership augmented affine projection algorithm. And since the maximum complex correntropy criterion is robust, it can enhance the ability of the algorithm to resist impact noise. Then to better deal with the problems caused by strongly correlated input signals, the simplified set-membership augmented affine projection maximum complex correntropy algorithm (SSM-AAPMC) is proposed in this brief. To extend the algorithm to a more general form, the simplified set-membership augmented affine projection generalized maximum complex correntropy algorithm (SSM-AAPGMC) is derived. Furthermore, the stability of the proposed algorithm as well as the computational complexity between the proposed and other algorithms are analyzed. Finally, the simulation results confirm that the proposed two algorithms can maintain good performance under impact noise.

关键词： Signal processing algorithms Cost function Computational complexity Random variables Kernel filtering algorithms Computational efficiency Affine projection maximum complex correntropy set-membership

来源：评论

学校读者我要写书评

暂无评论

Feature Subset Selection for High-Dimensional, Low Sampling Size Data Classification Using Ensemble Feature Selection With a Wrapper-Based Search

引用

IEEE ACCESS 2024年 12卷 62341-62357页

作者： Mandal, Ashis Kumar Nadim, MD. Saha, Hasi Sultana, Tangina Hossain, Md. Delowar Huh, Eui-Nam Univ Saskatchewan Dept Comp Sci Saskatoon SK S7N 5A2 Canada Hajee Mohammad Danesh Sci & Technol Univ Dept Comp Sci & Engn Dinajpur 5200 Bangladesh Hajee Mohammad Danesh Sci & Technol Univ Dept Elect & Commun Engn Dinajpur 5200 Bangladesh Kyung Hee Univ Dept Comp Sci & Engn Yongin 17104 South Korea

The identification of suitable feature subsets from High-Dimensional Low-Sample-Size (HDLSS) data is of paramount importance because this dataset often contains numerous redundant and irrelevant features, leading to poor classification performance. However, the selection of an optimal feature subset from a vast feature space creates a significant computational challenge. In the domain of HDLSS data, conventional feature selection methods often face challenges in achieving a balance between reducing the number of features and preserving high classification accuracy. Addressing these issues, the study introduces an effective framework that employs a filter and wrapper-based strategy specifically designed to address the classification challenges inherent in HDLSS data. The framework adopts a multi-step approach where ensemble feature selection integrates five filter ranking approaches: Chi-square ( chi(2) ), Gini index (GI), F-score, Mutual Information (MI), and Symmetric uncertainty (SU) to identify the top-ranking features. In the subsequent stage, a wrapper-based search method is utilized, which employs the Differential Evaluation (DE) metaheuristic algorithm as the search strategy. The fitness of feature subsets during this search is assessed based on a weighted combination of the error rate of the Support Vector Machine (SVM) classifier and the ratio of feature cardinality. The datasets, after undergoing dimensionality reduction, are then utilized to construct classification models using SVM, K-Nearest Neighbors (KNN), and Logistic Regression (LR). The approach was evaluated on 13 HDLSS datasets to assess its efficacy in selecting appropriate feature subsets and improving Classification Accuracy (ACC) analog with Area Under the Curve (AUC). Results show that the proposed ensemble with wrapper-based approach produces a smaller number of features (ranging between 2 and 9 for all datasets), while maintaining a commendable average AUC and ACC (between 98% and 100%). T

关键词： Feature extraction Support vector machines Metaheuristics Search problems filtering algorithms Information filters Classification algorithms Classification differential evaluation feature selection filter approach HDLSS data wrapper approach

来源：评论

学校读者我要写书评

暂无评论

EgeFusion: Towards Edge Gradient Enhancement in Infrared and Visible Image Fusion With Multi-Scale Transform

IEEE TRANSACTIONS ON COMPUTATIONAL IMAGING

引用

IEEE TRANSACTIONS ON COMPUTATIONAL IMAGING 2024年 10卷 385-398页

作者： Tang, Haojie Liu, Gang Qian, Yao Wang, Jiebang Xiong, Jinxin Shanghai Univ Elect Power Coll Automat Engn Shanghai 200090 Peoples R China

Existing image fusion methods focus on aggregate image features from different modalities into a clear and comprehensive image. However, these solutions ignore the importance of gradient features, which results in smooth performance of contrast information in the fused images. In this paper, an edge gradient enhancement method for infrared and visible image fusion is proposed, named EgeFusion. First, the source images are decomposed into a series of base and detail layers through a simple weighted least squares filter. Next, sub-window variance filter is proposed for the fusion of detail layers. For the base layer, a fusion strategy that combines visual saliency mapping with the idea of adaptive weight assignment is designed. The method effectively assigns the saliency features in source images globally, thus providing more comprehensive and valuable information about the region of interest in fused images. Finally, the fused base and detail layers are reconstructed in reverse to obtain the fusion results. The experimental results show that the proposed method has significantly enhanced the gradient features in source images, which makes it easier for the human eye system to focus on the region of interest. Compared with other state-of-the-art fusion methods, the proposed EgeFusion has superior visual quality and acceptable results in infrared and visible, multi-focus, as well as multi-modal medical image fusion. More importantly, our approach achieves performance improvements on object detection.

关键词： Image fusion Information filters Image edge detection Transforms Adaptive filters Imaging filtering algorithms Gradient feature weighted least squares filter image fusion saliency mapping adaptive weight assignment

来源：评论

学校读者我要写书评

暂无评论

Infrared Small Target Detection Based on Adaptive Size Estimation by Multidirectional Gradient Filter

引用

IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING 2024年 62卷

作者： Hao, Congyu Li, Zhengzhou Zhang, Yuting Chen, Wenhao Zou, Yong Chongqing Univ Sch Microelect & Commun Engn Chongqing 400044 Peoples R China

The strong edge of bright clutter would produce a large number of false alarms, and it seriously debases the detection performance of infrared (IR) small targets. Existing IR small target detection algorithms often lack adaptability in complex scenes, and their performance heavily relies on initial parameter configurations, including target size, which is closely related to the characteristics of the background and target. This article proposes an IR small target detection algorithm based on adaptive size estimation by a multidirectional gradient filter to address these limitations. First, based on the Gaussian-like distribution of a small target, the multidirectional gradient filter is constructed to enhance the target in order to get the target characteristic map (TCM). Second, the size of the small target is estimated from the enhanced TCMs by the criteria of sinusoidal curve best fitting. Subsequently, a multidirectional morphological filter with the optimal estimated size is proposed to suppress the strong background clutter to obtain the local strength difference map (LSDM). Finally, the corresponding enhanced TCM with the optimal estimated target size is fused with the LSDM for threshold segmentation to further suppress the background and detect the small targets. A large number of experimental results show that the proposed method can not only accurately estimate target size but also effectively detect small targets in diverse, complex scenes.

关键词： Clutter Object detection Image edge detection filtering algorithms Correlation Computational modeling Geoscience and remote sensing Fitting Target tracking Shape Adaptive size estimation complex background suppression infrared (IR) small target detection multidirectional gradient filter multidirectional morphological filter small target enhancement

来源：评论

学校读者我要写书评

暂无评论

Optimal Decentralized Federated Density Fusion of GM-PHD Filters

引用

IEEE SENSORS JOURNAL 2025年第8期25卷 13420-13432页

作者： Yu, Xue Feng, Xi-An Northwestern Polytech Univ Sch Marine Sci & Technol Xian 710072 Peoples R China

To achieve the optimal uncertain multitarget decentralized tracking fusion, a federated density fusion algorithm of Gaussian mixture-probability hypothesis density (GM-PHD) filters is proposed. Our fusion algorithm requires a hierarchical structure. Each sensor node runs a local GM-PHD filter to extract multitarget state estimates from sensor measurements. Besides running a global filter that only predicts the fusion result of the previous moment, the fusion node also collects, associates, and merges the GM-PHDs of the global and local filters after eliminating the correlations among all filters. The association results can be judged as survival target detection, missed detection, target new birth, or target vanishing. Each association situation employs a different optimal merging method. To eliminate correlations resulting from feedback and common process noise, the covariance upper bounding technique is introduced. This technique returns the cross-covariances among all filters to zeros by increasing their auto-covariances and the process noise covariance. It is proven that using the global filter, feedback, and covariance upper bounding technique, the fusion node can obtain the same optimal accuracy as the centralized fusion. Simulations demonstrate that our algorithm, with appropriate parameters, can obtain the optimal tracking accuracy and outperforms the existing arithmetic average (AA) fusion algorithm.

关键词： filtering algorithms Merging Target tracking Sensor fusion Information filters Data mining Training Intelligent sensors Correlation Artificial intelligence Covariance upper bounding technique decentralized fusion Gaussian mixture-probability hypothesis density (GM-PHD) filter uncertain multitarget

来源：评论

学校读者我要写书评

暂无评论

Meta-SonifiedDroid: Metaheuristics for Optimizing Sonified Android Malware Detection

引用

IEEE ACCESS 2024年 12卷 134779-134808页

作者： Tarwireyi, Paul Terzoli, Alfredo Adigun, Matthew O. Univ Zululand Dept Comp Sci ZA-3886 Kwa Dlangezwa South Africa

To mitigate the rising threat of Android malware, researchers have been actively looking for mechanisms that will enable rapid and accurate malware detection. Recently, attention has been paid to the use of audio-based features derived through the use of music information retrieval techniques. Since the exploration of these features is still in the early stages, there is a need to continue experimentation, especially with features that have yet to be used for this task. In this paper, we present the results of an ongoing investigation into the use of audio-based features for Android malware detection. In addition to extracting new audio-based features, this research aims to find the most discriminative subset of audio-based features through a comparative evaluation of Wrapper-based metaheuristic optimization algorithms on two separate datasets. First, we sonified the Android APK datasets, then extracted 191 static audio-based features from the resultant audio datasets. Fourteen different nature-inspired Wrapper-based metaheuristic optimization algorithms were evaluated for feature selection, and the selected features were then used to train the light gradient-boosting machine (LGBM) classification model. Experimental results demonstrate that the proposed approach exhibits high discriminative capabilities that can outperform other state-of-the-art techniques. The best outcome for Android malware detection was obtained using features selected by the Genetic Algorithm, which achieved 50.26% feature reduction and an improved classification accuracy of 99.72%.

关键词： Feature extraction Malware Metaheuristics Classification algorithms Accuracy filtering algorithms Machine learning algorithms Sonification Androids audio-based features android malware detection feature selection metaheuristic optimization genetic algorithm (GA) nature-inspired algorithms queuing search algorithm (QSA)

来源：评论

学校读者我要写书评

暂无评论

DCFP: Distribution Calibrated Filter Pruning for Lightweight and Accurate Long-Tail Semantic Segmentation

引用

IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY 2024年第7期34卷 6063-6076页

作者： Wang, Zixiao Xie, Hongtao Wang, Yuxin Xu, Hai Jin, Guoqing Univ Sci & Technol China Sch Informat Sci & Technol Hefei 230022 Peoples R China Peoples Daily Online State Key Lab Commun Content Cognit Beijing 100733 Peoples R China

Recently, semantic segmentation has made promising progress, but the high cost of processing still limits its application. With focusing on removing the parameters of the networks, filter pruning using the importance criterion is a straightforward and effective technique to obtain the lightweight sub-network. However, we argue that the long-tail distribution in segmentation datasets poses two significant problems which are ignored in existing pruning algorithms: 1) The importance criterion is dominated by head classes which contain numerous positive samples, where the knowledge of tail classes is easily degenerated. 2) The degenerated knowledge of tail classes is hard to recover as their samples are also insufficient during fine-tuning. To address these issues, we propose a Distribution Calibrated Filter Pruning (DCFP) framework for segmentation. Firstly, a gradient-based Equalization Importance Criterion (EIC) is designed to generate a class-balanced pruning procedure. It avoids the bias on head classes by discarding the imbalanced positive gradients. Secondly, we introduce a Geometric-Semantic Re-balanced Loss (GSRL) to emphasize the learning on tail classes during fine-tuning. The GSRL consists of two cooperative components to calibrate the imbalanced optimization on geometric and semantic domains dynamically. Compared with previous methods, DCFP explores a novel distribution-aware pruning framework to obtain lightweight architectures with accurate results. Extensive experiments proved that DCFP achieves impressive performance on four popular segmentation benchmarks.

关键词： Tail Semantics Task analysis Regulation Semantic segmentation Optimization filtering algorithms filter pruning imbalance

来源：评论

学校读者我要写书评

暂无评论

An FPGA-Based Hardware Architecture of Gaussian-Adaptive Bilateral Filter for Real-Time Image Denoising

引用

IEEE ACCESS 2024年 12卷 115277-115285页

作者： Xie, Ailin Zhang, Ao Mei, Guohui Northeastern Univ Coll Informat Sci & Engn Shenyang 110819 Peoples R China

The bilateral filter exhibits remarkable efficacy in noise suppression and edge preservation. This article proposes a hardware architecture based on field-programmable gate array (FPGA) for a modified bilateral filter. To enhance the efficacy of the bilateral filter, the Gaussian-adaptive Bilateral Filter (GABF)is employed as a modified filtering method. Approximating the filter weights using look-up tables (LUTs)results in reduced storage requirements and eliminates the need for complex exponential weight calculations. Moreover, the GABF is markedly accelerated by the highly parallel functional modules and LUTs. In addition to the aforementioned features, the GABF is implemented as a parallel architecture, which results in are duction in hardware resource utilization compared to previous works. The results of the image quality analysis demonstrate that this article can achieve superior image quality compared with state-of-the-artworks. The implementation results indicate that the proposed architecture is capable of performing real-time denoising at a frame rate of 95.65 fps for a 640 x 480 video with a power dissipation of 93.22 mW.

关键词： filtering algorithms Low-pass filters Kernel Information filters Hardware Field programmable gate arrays Computer architecture Bilateral filter edge-preserving field-programmable gate array (FPGA) image denoising

来源：评论

学校读者我要写书评

暂无评论

CPU Address-Leakage Transient Execution Attack Detection and Its Countermeasures

引用

IEEE TRANSACTIONS ON COMPUTER-AIDED DESIGN OF INTEGRATED CIRCUITS AND SYSTEMS 2024年第12期43卷 4463-4475页

作者： Yin, Yiyun Cui, Jinhua Zhang, Jiliang Hunan Univ Changsha 410082 Peoples R China

Modern advanced CPU designs are frequently exposed to transient execution vulnerabilities, which allow attackers to harness microarchitectural side effects for data exfiltration. The leaked data may encompass direct target data, such as RSA keys or indirect information, like physical page mappings. Thus, transient execution attacks can be divided into data-leakage and address-leakage, depending on the specific targets that are exposed. Existing studies have developed practical defenses and detection mechanisms against the microarchitectural attacks. However, almost all of them focus solely on data leakage and are thus unable to detect and counter address-leakage attacks, like Spoiler, due to their unique mechanisms. This article introduces AAlert, the first detection mechanism specifically designed for address-leakage transient execution attacks. AALERT integrates a Cuckoo filter module within the CPU's memory order buffer (MOB) to screen buffered addresses on the fly. We further optimize the filtering algorithm to minimize false positives. We discuss and implement several countermeasures to defeat the detected attacks. Finally, we evaluate the effectiveness and performance of AAlert based on prototype implementations, demonstrating a detection rate of 99.99% with negligible performance overhead.

关键词： Transient analysis Microarchitecture Hardware Integrated circuits Optimization filtering algorithms Design automation Countermeasures cuckoo filter detection mechanism information leakage transient execution attack

来源：评论

学校读者我要写书评

暂无评论

Swarm Optimization of 2-D Metasurface Structure for Low-Crosstalk LWIR Filter Arrays

引用

IEEE SENSORS JOURNAL 2024年第22期24卷 36971-36978页

作者： Zhang, Xun Wang, Jiaxuan Reda, Mohamed Zhao, Yongqiang Kong, Seong G. Northwestern Polytech Univ Sch Automat Xian 710072 Peoples R China Mil Tech Coll Cairo Egypt Sejong Univ Dept Comp Engn Seoul 05006 South Korea

Recent developments in division of focal plane (DoFP) sensor technology have led to the availability of compact snapshot multispectral imagers. However, many of these imagers based on micro-/nano-filter structures encounter issues, such as crosstalk among the units. To address this, we propose a 2-D metasurface structure that modulates light to capture spectral information in multiple bands in the long-wave infrared (LWIR) spectrum. We also present a multiobjective optimization method to optimize the 2-D metasurface structure as an LWIR multispectral filter array with low crosstalk. The proposed multiobjective optimization method, based on a particle swarm optimization algorithm, finds the optimal parameters for the 2-D metasurface structure to minimize the crosstalk between the units of the filter array. Since the crosstalk between two bands may not be measured directly, we compute spectral aliasing between two bands to estimate the amount of crosstalk. We transform the problem of reducing crosstalk into a multiobjective optimization problem to minimize spectral aliasing. Simulation results demonstrate that our proposed filter array design technique reduces crosstalk among filter array units, enabling high-fidelity spectral characteristics in snapshot LWIR multispectral imaging.

关键词： Metasurfaces Crosstalk Optical filters Sensors Information filters filtering algorithms Amplitude modulation filtering theory Boundary conditions Particle swarm optimization metasurface snapshot multispectral imagers spectral aliasing swarm optimization

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：