检索结果-内蒙古大学图书馆

Adaptive Scaling Filter Pruning Method for Vision Networks With Embedded Devices

IEEE ACCESS 2024年 12卷 123771-123781页

作者： Ko, Hyunjun Kang, Jin-Ku Kim, Yongwoo Inha Univ Dept Elect & Comp Engn Incheon 22212 South Korea Korea Natl Univ Educ Dept Technol Educ Cheongju 28173 South Korea

Owing to improvements in computing power, deep learning technology using convolutional neural networks (CNNs) has recently been used in various fields. However, using CNNs on edge devices is challenging because of the large computation required to achieve high performance. To solve this problem, pruning, which reduces redundant parameters and computations, has been widely studied. However, a conventional pruning method requires two learning processes, which are time-consuming and resource-intensive, and it is difficult to reflect the redundancy in the pruned network because it only performs pruning once on the unpruned network. Therefore, in this paper, we utilize a single learning process and propose an adaptive scaling method that dynamically adjusts the size of the network to reflect the changing redundancy in the pruned network. To verify the performance of each method, we compare the performance of the proposed methods by conducting experiments on various datasets and networks. In our experiments using the ImageNet dataset on ResNet-50, pruning FLOPs by 50.1% and 74.0% resulted in a decrease in top-1 accuracy by 0.92% and 3.38%, respectively, and improved inference time by 26.4% and 58.9%, respectively. In addition, pruning FLOPs by 27.37%, 36.84% and 46.41% using the COCO dataset on YOLOv7, reduced mAP(0.5-0.95) by 1.2%, 2.2% and 2.9%, respectively, and improved inference time by 12.9%, 16.9% and19.3%.

关键词： Information filters Adaptive systems Adaptive filters Training filtering algorithms Quantization (signal) Batch normalization Computer vision Convolutional neural networks Deep learning convolutional neural network inference time network compression pruning

来源：评论

学校读者我要写书评

暂无评论

Joint Detection, Tracking, and Classification of Multiple Maneuvering Star-Convex Extended Targets

引用

IEEE SENSORS JOURNAL 2024年第4期24卷 5004-5024页

作者： Wang, Liping Zhan, Ronghui Peoples Publ Secur Univ China Sch Criminal Invest Beijing 100038 Peoples R China Natl Univ Def Technol Coll Elect Sci & Technol Changsha 410073 Peoples R China

The extended target (ET) joint detection, tracking, and classification (JDTC) algorithms based on elliptical shape utilize rough target size information for classification, which makes it difficult to effectively solve the problem of accurate classification of similarly sized targets. Therefore, this article proposes a multiple maneuvering star-convex ET JDTC algorithm called JDTC-MM-CBMeMBer filter. First, the target extent state is modeled as a star-convex shape via the star-convex random hypersurface model (RHM). By modeling the target class-related prior information with vector form, we construct its relationship with the simultaneous extent state and integrate it into the Bayesian filter framework for joint processing. Second, to solve the implementation difficulty due to the high-dimensional target state and strong nonlinear observation model in the star-convex RHM, we model the target kinematic by two separate vectors and use the particle filter (PF) to update target class probability. Next, we use the multiple model (MM) cardinality balanced multitarget multi-Bernoulli (CBMeMBer) filter to derive the JDTC recursion process of the multiple maneuvering star-convex ETs. At last, the Gamma-Gaussian-Gaussian mixture implementation is present. The simulation results show that: 1) compared with the ET JDTC algorithm based on an elliptical shape, the proposed filter can accurately classify targets with similar sizes but different shapes and 2) compared with the multiple ET tracking algorithm based on star-convex RHM, the proposed filter can significantly improve the target state estimation results without significantly increasing the running time.

关键词： filtering algorithms filtering theory Target tracking Classification algorithms Shape Information filters Kinematics Cardinality balanced multitarget multi-Bernoulli (CBMeMBer) extended targets (ETs) ioint tracking and classification star-convex random hypersurface model (RHM)

来源：评论

学校读者我要写书评

暂无评论

Efficient Skyline Keyword-Based Tree Retrieval on Attributed Graphs

引用

IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING 2024年第11期36卷 6056-6070页

作者： Wu, Dingming Zhang, Zhaofen Jensen, Christian S. Lu, Kezhong Shenzhen Univ Coll Comp Sci & Software Engn Shenzhen 518060 Peoples R China Aalborg Univ Dept Comp Sci DK-9220 Aalborg Denmark

Attributed graphs are graphs, where the vertices have attributes. Such graphs encompass, e.g., social network graph, citation graphs, and knowledge graphs, which have numerous real-world applications. Keyword-based search is a prominent and user-friendly way of querying attributed graphs. One widely used approach to keyword search adopts tree-based query semantics that relies on scoring functions that aggregate distances from a root to keyword-matched vertices. However, it is non-trivial to design scoring functions that capture different users' keyword preferences. This study defines and solves the skyline KTree retrieval problem that combines keyword querying with skyline functionality on attributed graphs. The result of a skyline KTree query is independent of scoring functions. Hence, no matter which keywords are preferred, users can always find their favorite KTrees in a result. To enable efficient skyline KTree retrieval, we propose algorithmFilterRefine that first identifies candidate results and then uses them for search space pruning. Computing distances between keywords and vertices is expensive and dominates the computational cost ofFilterRefine. Inspired by subspace skyline query techniques, we convert the skyline KTree retrieval problem into a multi-dimensional subspace skyline problem and propose algorithm MultiDiSkylineOpt. This algorithm is able to reuse skylines in subspaces and uses bounds on all dimensions to accelerate distance computation. Experimental results on real datasets show that a baseline algorithm cannot report results within a 500 second cut-off time, while the proposed algorithms are able to compute results in reasonable time. In particular, MultiDiSkylineOpt is able to efficiently retrieve skyline KTrees on large graphs with millions of nodes and hundreds of millions of edges.

关键词： filtering algorithms Europe Semantics Indexes Social networking (online) Keyword search Query processing Skyline query keyword search attributed graph query processing

来源：评论

学校读者我要写书评

暂无评论

Supervised Feature Selection via Multi-Center and Local Structure Learning

引用

IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING 2024年第9期36卷 4930-4942页

作者： Zhang, Canyu Nie, Feiping Wang, Rong Li, Xuelong Northwestern Polytech Univ Sch Comp Sci Minist Ind & Informat Technol Key Lab Intelligent Interact & Applicat Xian 710072 Peoples R China Northwestern Polytech Univ Sch Artificial Intelligence Opt & Elect iOPEN Key Lab Intelligent Interact & Applicat Minist Ind & Informat Technol Xian 710072 Peoples R China

Feature selection has achieved unprecedented success in obtaining sparse discriminative features. However, the existing methods almost use the & ell;(2,p)-norm constraint on transformation matrix to obtain sparse features, which introduces extra parameters and cannot obtain the features directly. In addition, existing algorithms only focused on the global structure and ignored the local structure, leading to poor performance when solving data with non-Gaussian distributions which a single center point cannot describe precisely. Based on above considerations, we propose a supervised feature selection via multi-center and local structure learning. We further introduce trace ratio criterion into our model in favor of improving the discriminant of features selected. In order to address the overlap problem, we use multiple center points to match the distribution of data and construct a k-Nearest Neighbor graph to explore the local structure of the data. In addition, we also propose an efficient method to optimize the transformation matrix with the & ell;(2,0)-norm constraint and can directly obtain the sparse features. We evaluate our method on Toy datasets and several real-world datasets, show improvement over state-of-the-art feature selection methods, and demonstrate the effectiveness of our model in dealing with non-Gaussian distributed data problems.

关键词： Feature extraction Sparse matrices Optical filters Information filters Signal processing algorithms Reviews filtering algorithms Supervised feature selection, sparse discriminant features multi-center local structure & ell (2,0 )-norm constraint

来源：评论

学校读者我要写书评

暂无评论

Adaptive Spatial Regularization Correlation Filters for UAV Tracking

引用

IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING 2024年 17卷 7867-7877页

作者： Cao, Yulin Dong, Shihao Zhang, Jiawei Xu, Han Zhang, Yan Zheng, Yuhui Qinghai Normal Univ Sch Comp Xining 810000 Peoples R China Nanjing Univ Informat Sci & Technol Sch Comp Nanjing 210044 Peoples R China

As a tool for near-earth remote sensing, unmanned aerial vehicle (UAV) can be used to acquire images and data of the earth's surface. This provides a powerful support for Earth observation and resource management. Object tracking in UAV videos has been a topic of much interest in recent years. A large number of algorithms have been proposed. Among these algorithms, deep learning has achieved a high accuracy rate. However, it is difficult to carry hardware devices for UAV, which makes it difficult to be practically applied. The correlation filter does not require a graphic processing unit to accelerate the computation, but it uses only manual features, which makes it difficult to achieve satisfactory performance. In order to solve the above problems, we proposed adaptive spatial regularization correlation filters, called DTSRT. Specifically, we first introduce deep features in the correlation filter instead of the original manual features, which can greatly improve the discriminative ability of the model. At the same time, in order to prevent affecting the real-time performance of the algorithm, we use histogram of oriented gradients features to determine the target scale and deep features to determine the target location. In addition, considering the large inter-frame distance between targets in UAV videos, we use a saliency detection method to dynamically generate spatially constrained templates. The proposed DTSRT outperforms other state-of-the-art algorithms with area under curve of 0.481, 0.431, and 0.474 on UAV123@10FPS, UAVDT, and DTB70 datasets, respectively.

关键词： Correlation Target tracking filtering algorithms Autonomous aerial vehicles Feature extraction Videos Heuristic algorithms Adaptive spatial regularization correlation filter deep features UAV tracking

来源：评论

学校读者我要写书评

暂无评论

Division of Regulatory Power: Collaborative Regulation for Privacy-Preserving Blockchains

引用

IEEE TRANSACTIONS ON INFORMATION FORENSICS AND SECURITY 2024年 19卷 2533-2548页

作者： Zhaolu, Tianyu Wan, Zhiguo Wang, Huaqun Nanjing Univ Posts & Telecommun Sch Comp Sci Nanjing 210003 Peoples R China Zhejiang Lab Hangzhou 310000 Peoples R China Guizhou Univ State Key Lab Publ Big Data Guiyang 550025 Guizhou Peoples R China

Decentralized anonymous payment schemes may be exploited for illicit activities, such as money laundering, bribery and blackmail. To address this issue, several regulatory-friendly decentralized anonymous payment schemes have been proposed. However, most of these solutions lack restrictions on the regulator's authority, which could potentially result in power abuse and privacy breaches. In this paper, we present a decentralized anonymous payment scheme with collaborative regulation (DAPCR). Unlike existing solutions, DAPCR reduces the risk of power abuse by distributing regulatory authority to two entities: Filter and Supervisor, neither of which can decode transactions to access transaction privacy without the assistance of the other one. Our scheme enjoys three major advantages over others: 1) Universality, achieved by using zk-SNARK to extend privacy-preserving transactions for regulation. 2) Collaborative regulation, attained by adding the ring signature with controllable linkability to the transaction. 3) Efficient aggregation of payment amounts, achieved through amount tags. As a key technology for realizing collaborative regulation in DAPCR, the ring signature with controllable linkability (CLRS) is proposed, where a user needs to specify a linker and an opener to generate a signature. The linker can extract pseudonyms from signatures and link signatures submitted by the same signer based on pseudonyms, without leaking the signer's identity. The opener can recover the signer's identity from a given pseudonym. The experimental results reflect the efficiency of DAPCR. The time overhead for transaction generation is 1231.2ms, representing an increase of less than 50% compared to ZETH. Additionally, the time overhead for transaction verification is only 1.2ms.

关键词： Regulation filtering algorithms Collaboration Blockchains Security Regulators Information filters Ring signature blockchain cryptocurrency regulation decentralized finance

来源：评论

学校读者我要写书评

暂无评论

A Synthesis Framework for Finding All Solutions of Microwave Filter Topologies Based on Homotopy Continuation

引用

IEEE TRANSACTIONS ON MICROWAVE THEORY AND TECHNIQUES 2025年

作者： Zeng, Yi Xiao, Rui Zhuo Huang, Huijin Zhang, Peng Yu, Ming Southern Univ Sci & Technol Dept Elect & Elect Engn Shenzhen 518055 Peoples R China

This article proposes a numerical filter synthesis framework based on homotopy continuation (HC) to find all solutions of target topologies reliably. First, the "base model" is defined and exhaustively solved via perturbation starting from one solution that can be calculated analytically. Although the base model features nonideal filtering characteristics, it can act as the starting system of the following homotopy path-tracking process. In particular, the base model can bring the computation burden that is less sensitive to the increase of filter orders, which makes the synthesis of high-order topologies feasible. Afterward, the "target models" with satisfactory filtering responses can be synthesized exhaustively and efficiently via the adaptive-step path tracking from the stored base model. All the solved base models and target models constitute a database, in which the accumulated pre-calculated information can significantly accelerate the synthesis of filters with the same topology but different return loss and transmission zeros (TZs). In addition, the parallel computation technique is exploited to further speed up the presented synthesis framework. Multiple synthesis examples, including higher order filters, are presented to verify the proposed theories. One combline filter is fabricated, and the measurement results exhibit a good agreement with the theoretical response.

关键词： Microwave filters filtering theory Topology Transversal filters Mathematical models Computational modeling Target tracking filtering algorithms Databases Adaptation models Coupling matrix (CM) exhaustive filter synthesis homotopy perturbation

来源：评论

学校读者我要写书评

暂无评论

Adaptive Low-Rank DOA Estimation Using Complex Kronecker Product Decomposition

引用

IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY 2024年第7期73卷 10726-10731页

作者： Joel, S. Yadav, Shekhar Kumar George, Nithin V. Indian Inst Technol Gandhinagar Dept Elect Engn Gandhinagar 382355 India

Traditional non-adaptive subspace-based direction of arrival (DOA) estimation algorithms require a lot of computation and are not suitable for power efficient implementation which is a necessity in battery-operated smart vehicles. Least mean square (LMS) based adaptive DOA estimation methods are computationally efficient for smaller sensor arrays but as the length of the array increases, the rate of convergence of these methods starts decreasing. In this correspondence, we propose two adaptive DOA estimation methods that decompose the large weights of the DOA estimating filter into smaller weights using a complex Kronecker product based low-rank decomposition scheme. The smaller weights of the two proposed algorithms are updated using the normalized LMS (NLMS) and recursive least squares (RLS) principles, respectively. Updating the smaller weights parallelly instead of one larger filter results in significantly lower computations, faster convergence along with competitive steady-state performance. We derive the update rules for the smaller weights and study the computational complexities of our methods. Various simulation validates the low-rank approximation and showcases the effectiveness of the proposed methods in estimating DOAs adaptively.

关键词： Direction-of-arrival estimation Estimation Adaptive arrays Signal processing algorithms Sensor arrays Convergence filtering algorithms Adaptive direction of arrival (DOA) estimation array signal processing Kronecker product complex least mean square (LMS) complex recursive least squares (RLS)

来源：评论

学校读者我要写书评

暂无评论

An Improved Lane Detection and Lane Departure Warning Framework for ADAS

引用

IEEE TRANSACTIONS ON CONSUMER ELECTRONICS 2024年第2期70卷 4793-4803页

作者： Sukumar, N. Sumathi, P. Indian Inst Technol Roorkee Dept Elect Engn Roorkee 247667 India

An improved vision-based framework is proposed for lane detection and departure warning system. It consists of an edge filter developed based on sliding discrete cosine transform for lane features extraction, RANdom SAmple Consensus (RANSAC) method for model fitting of the lanes, and a lateral offset ratio computation on lanes for warning system. A finding of edge detection kernel extracted from 2-D SDCT is reported with its evaluation on road images. FPGA implemented edge detector's performance is proved to be better compared to existing edge detectors in terms of Pratt's figure of merit, average gradient, and entropy for road images. Moreover, the edge detection capability under noisy condition of 10 dB SNR is proved to be the best. The lane detection and warning system are implemented in real-time processor of compact reconfigurable input output system using vision development module. Lane departure warning is issued based on the computed LOR using X-coordinates of detected bottom end points of lane lines in an image plane. The lane detection performance is evaluated on the Caltech lane dataset, KIST lane dataset, and LaneVisionIITR dataset with diversity of 4329 images altogether. The proposed framework shows the performance improvement on urban and highway driving in complex scenarios.

关键词： Image edge detection Feature extraction Lane detection Kernel Roads Noise filtering algorithms Edge detection kernel image edge feature sliding discrete cosine transform lane detection lane feature extraction road images lane departure warning

来源：评论

学校读者我要写书评

暂无评论

Direct Filter Learning From Iterative Reconstructed Images for High-Quality Analytical CBCT Reconstruction Using FDK-Based Neural Network

引用

IEEE ACCESS 2024年 12卷 121495-121506页

作者： Qi, Hongliang Long, Chao Li, Hanwei Huang, Shuang Hu, Debin Xu, Yuan Chen, Hongwen Southern Med Univ Nanfang Hosp Dept Clin Engn Guangzhou 510515 Peoples R China La Consolacion Univ Philippines Sch Grad Studies Malolos 3000 Philippines Foshan Shunde Heyou Hosp Foshan 528399 Peoples R China Southern Med Univ Sch Biomed Engn Guangzhou 510515 Peoples R China

Purpose: We propose an FDK-based neural network to directly learn the filter from an iterative reconstruction (IR) algorithm and apply the learned filter in the FDK algorithm to obtain a high-quality CBCT reconstruction. Methods: The FDK algorithm is transformed into a linear expression of several matrix multipliers and embedded into neural network layers. Then, the FDK-based neural network framework is built including two fundamental modules and four core network layers. This network model can learn a filter directly from the iteratively reconstructed CBCT images by cascading the network layers of cosine weighting, filtering, backprojection, and leaky rectified linear unit and setting filter as the only trainable parameter. Preliminary and simulation studies performed on abdominal CT datasets are conducted to explore the correctness and effectiveness of the learned filter. Then, the head and neck CT data and Catphan phantom are utilized to demonstrate the generalization performance of the learned filter. Results: Preliminary study shows that the learned filter is consistent with the target filter, and the mean absolute difference is around 0.001. Compared with conventional FDK, the FDK-based neural network shows a better image quality with the peak signal-to-noise ratio (PSNR) and structural similarity (SSIM) increasing by 67% and 6%, respectively. In terms of the line-pair slice in the Catphan phantom, the SSIM and PSNR are improved by 13.75% and 42.78%, respectively. Conclusions: The FDK-based neural network can reconstruct high-quality images by directly learning a filter from the label images and provides a new perspective on solving the time-consuming problem of IR methods.

关键词： Image reconstruction filtering algorithms Neural networks Computed tomography Training Nonlinear filters Maximum likelihood detection Algorithm design and analysis CBCT iterative reconstruction algorithm CBCT FDK algorithm neural network learnable filter

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：