检索结果-内蒙古大学图书馆

A post-processing method for Time-Reassigned Multisynchrosqueezing Transform and Its Application in processing the Strong Frequency-Varying Signal

引用

IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT 2021年 70卷 1页

作者： Yu, Kun Wang, Xuesong Cheng, Yuhu China Univ Min & Technol Engn Res Ctr Intelligent Control Underground Spac Minist Educ Xuzhou 221116 Jiangsu Peoples R China China Univ Min & Technol Sch Informat & Control Engn Xuzhou 221116 Jiangsu Peoples R China

Strong frequency-varying signal always undergoes a rapid frequency change within a short time duration. Converting its 1-D waveform into the 2-D time-frequency (TF) plane is convenient for us to understand the structure of the strong frequency-varying signal. Time-reassigned multisynchrosqueezing transform (TMSST) is a recently proposed time-frequency analysis (TFA) method, which owns the capability to provide an energy concentrated time-frequency representation (TFR) result for the strong frequency-varying signal. However, the fixed-point iterative estimation manner of the group delay in the TMSST method would lead to some non-reassigned TF points and further affect the readability of the TFR result. To address this issue, a post-processing method for the TMSST is proposed in this study. In the proposed method, the group delay of the strong frequency-varying signal is estimated by local maximizing the TFR result provided by TMSST in the time direction, and then the TF coefficients from the TMSST are rearranged among the time direction on the basis of the new estimated group delay. Results from simulation and experimental case studies verify that the proposed method can provide a more energy-concentrated TFR result in comparison with the other existing TFA methods.

关键词： Group delay local maximizing post-processing method strong frequency-varying signal time-reassigned multi-synchrosqueezing transform (TMSST)

来源：评论

学校读者我要写书评

暂无评论

post-processing method for removing low-frequency bias in glottal inverse filtering

引用

ELECTRONICS LETTERS 2015年第1期51卷 110-U123页

作者： Koc, Turgay SDU Isparta Turkey

Estimation of glottal flow waveform from speech requires recording of low-frequency sounds down to DC level. It causes a low-frequency noise or bias in the reconstructed glottal flow waveform. Removing this bias by linear highpass filtering (HPF) degrades the shape of glottal flow waveform. A nonlinear method based on empirical mode decomposition is proposed for removing the bias without using HPF while preserving the shape of the glottal flow waveform. The biased glottal flow waveform is decomposed into its intrinsic modes, then the low-frequency bias is estimated by using the higher modes. Glottal flow waveform is reconstructed by subtracting the bias in the time domain. The results show that the proposed method accurately estimates the bias and yields significantly better glottal flow waveforms than the conventional HPF method.

关键词： time-domain analysis high-pass filters waveform analysis speech processing glottal flow waveform post-processing method low-frequency bias linear highpass filtering low-frequency noise empirical mode decomposition glottal inverse filtering time domain HPF

来源：评论

学校读者我要写书评

暂无评论

Research on a New Exponential Function Weighted Averaging method Used for Full-Gradient Strain Measurement of DIC

引用

EXPERIMENTAL MECHANICS 2025年第2期65卷 177-194页

作者： Song, X. Zhou, C. Xiong, K. Nanjing Univ Aeronaut & Astronaut State Key Lab Mech & Control Aerosp Struct 29 Yudao St Nanjing 210016 Peoples R China Shanghai Inst Satellite Engn 3666 Yuanjiang Rd Shanghai 200240 Peoples R China

BackgroundIn the implementation of Digital Image Correlation (DIC), several post-processing methods have been developed to calculate reliable strain field. Nevertheless, achieving effective and easy-to-implement strain measurement for full-gradient strain fields continues to be a *** widely used pointwise least square (PLS) method is hard to get a balance between smoothing and accuracy when dealing with different deformation fields. A large strain calculation window may lead to over-smoothing, whereas a small strain calculation window may be insufficient to suppress *** new exponential function and the exponential function weighted averaging (EFWA) method are proposed. The shape of the exponential function can be either sharp-topped or flat-topped, allowing the EFWA method to either preserve or smooth the original strain results. A straightforward and effective selection strategy for parameters of the exponential function is also provided, enabling the EFWA method to achieve self-adaptive *** calculation examples of synthetic images indicate that, the proposed EFWA method can consistently yield high measurement accuracy for unidirectional and multi-directional complex deformation fields and exhibits superior spatial resolution compared to the PLS method. The minimum Metrological Efficiency Indicator (MEI) value for the EFWA method is 1.72, compared to 4.67 for original results and 5.10 for the PLS method. The results of a tensile experiment carried out on an open-hole specimen indicate that, after the EFWA method is implemented, the strain results in areas away from the hole are effectively smoothed and the strain results in areas around the hole are *** proposed EFWA method can achieve effective and easy-to-implement strain measurement for full-gradient strain fields.

关键词： Digital image correlation post-processing method Weighted averaging method Exponential function Full-gradient strain measurement

来源：评论

学校读者我要写书评

暂无评论

A low-cost granular-medium hot quasi-isostatic pressing method for enhancing compressive properties of 3D printed structures

引用

JOURNAL OF MANUFACTURING PROCESSES 2024年 115卷 441-451页

作者： Wang, Kui Tan, Qianbing Wang, Jin Liu, Yisen Zhai, Zhanyu Yao, Song Peng, Yong Cent South Univ Sch Traff & Transportat Engn Key Lab Traff Safety Track Minist Educ Changsha 410075 Peoples R China Cent South Univ Coll Mech & Elect Engn Changsha 410083 Peoples R China

3D printed structures suffered weak bead-to-bead interfacial bonding due to the layer-by-layer manufacturing process. The hot isostatic pressing (HIP) method was effective in improving the micro-structures and enhancing the mechanical properties of 3D printed structures, but it was limited due to the high-cost device and process. Inspired by the HIP method, and aimed at improving the compressive properties of 3D printed thin-walled structures, this work proposed a low-cost granular-medium hot quasi-isostatic pressing (GMHQIP) method realized by the homemade equipment. The hot quasi-isostatic ambient pressure could be produced by pressing the granular medium in the high-temperature environment. The thin-walled structures treated by the GMHQIP method showed a more regular progressive deformation mode than non-treated structures. Compared to the nontreated structure, the specific energy absorption of thin-walled structures treated at 200degree celsius with 2 MPa for 8 h significantly increased by 170.4 % to the value of 4.95 J/g. The results indicated that the strength and toughness of 3D printed materials improved significantly after the GMHQIP treatment. The interfacial bonding of 3D printed structures was enhanced by using the GMHQIP method. This work offered a low-cost and efficient postprocessing method for 3D printed structures to improve compressive properties.

关键词： 3D printing fiber-reinforced composites post-processing method

来源：评论

学校读者我要写书评

暂无评论

Accelerating Algorithm for Total Focusing method Imaging Based on Optimization of Full Matrix Data

引用

RUSSIAN JOURNAL OF NONDESTRUCTIVE TESTING 2023年第2期59卷 161-170页

作者： Li, Junhan Wang, Shaofeng Mao, Xin Liu, Wenjing Wang, Hailing Zhou, Shengrong Inner Mongolia Univ Sci & Technol Inner Mongolia Key Lab Intelligent Diag & Control Baotou 014010 Inner Mongolia Peoples R China Shanghai Aerosp Equipment Manufacturer Co Ltd Shanghai 201101 Peoples R China Inner Mongolia North Heavy Ind Grp Co Ltd Baotou 014033 Inner Mongolia Peoples R China Inner Mongolia Baotou Steel Xichuang Grp Co Ltd Baotou 014033 Inner Mongolia Peoples R China

Total focusing method (TFM) for ultrasonic phased array imaging has the advantages of high imaging resolution, full-scale dynamic focusing and low undetected error rate in ultrasonic detection. However, the problem of long imaging time usually limits its practically applications. The imaging speed can be raised sharply by directly upgrading the detection hardware, but this solution is unavoidable to increase the detection cost. This study is interested in how to improve the TFM imaging speed based on the existing hardware. Thus, an accelerating algorithm combining the asymmetrical optimization mechanism for the full matrix data of ultrasonic array with the sparse one was proposed to simplify the imaging process aiming at reducing the imaging time. Furthermore, TFM imaging experiments of the transverse through hole flaw were finally carried out to evaluate the performance of the proposed algorithm. The results show that in sparse optimization algorithms, the imaging speed of 2 elements sparse is faster and the signal-to-noise ratio of image is higher. The asymmetric optimization and 2 elements sparse optimization are combined in proposed algorithm. The imaging speed of proposed algorithm is improved by 34.56%. Similarly, it is 10.50 and 24.94%, respectively faster than using the asymmetric and 2 elements sparse optimization algorithm. Meanwhile, it is also founded that the signal-to-noise ratio (SNR) of TFM imaging after amplitude optimization is greatly increased by 55.99% without losing the information on the defect characterization. The proposed accelerating algorithm provides an important reference for the practical application of TFM in the ultrasonic phased array flaw detection system.

关键词： ultrasonic phased array total focusing method post-processing method imaging speed

来源：评论

学校读者我要写书评

暂无评论

Influence of post-processing methods on bond-slip behavior of nonlinear Fe-SMA lap-shear joints

引用

CASE STUDIES IN CONSTRUCTION MATERIALS 2024年 21卷

作者： Li, Lingzhen Martinelli, Enzo Wang, Wandong Chatzi, Eleni Ghafoori, Elyas Swiss Fed Labs Mat Sci & Technol Empa Uberlandstr 129 CH-8600 Dubendorf Switzerland Swiss Fed Inst Technol Inst Struct Engn IBK Dept Civil Environm & Geomatic Engn D BAUG CH-8093 Zurich Switzerland Univ Salerno Dept Civil Engn I-84084 Fisciano Sa Italy Northwestern Polytech Univ Sch Aeronaut Xian 710072 Shaanxi Peoples R China Leibniz Univ Hannover Inst Steel Construct Fac Civil Engn & Geodet Sci D-30167 Hannover Germany Hong Kong Polytech Univ Dept Civil & Environm Engn Hong Kong Peoples R China

Prestressed bonded strengthening for structures employing iron-based shape memory alloy (FeSMA) has been proven promising. Analyzing adhesively bonded joints necessitates a thorough understanding of the bond-slip behavior. However, when examining the bond-slip behavior of FeSMA-to-steel joints comprising nonlinear adhesives, the "forward" and "backward" postprocessing methods, representing the current state-of-the-art, produce a trilinear and a trapezoidal bond-slip pattern, respectively, which is inconsistent. To address this inconsistency, the current study investigates the bond behavior of Fe-SMA-to-steel joints, with a particular focus on the bond-slip behavior. Two finite element (FE) joints, one featuring a linear adhesive and the other comprising a nonlinear adhesive, are modeled and compared against physical tests from literature. The "forward" and "backward" processing methods are used to analyze the bond behavior of the two FE joints. Eventually, the aforementioned inconsistency is resolved;a triangular and a trapezoidal bond-slip pattern are characterized for Fe-SMA-to-steel lap-shear joints with linear and nonlinear adhesives, respectively. The trilinear bond-slip behavior is concluded as a result of error accumulation and propagation during the "forward" processing. A hybrid post-processing method, which takes the advantages of both the "forward" and "backward" processing methods, is further proposed for inferring the full-range behavior;the resulting experimental behaviors closely align with simulations using a trapezoidal bond-slip model as input. A comparison against carbon fiber reinforced polymer (CFRP) lap-shear joints demonstrates similar bond-slip characteristics between Fe-SMA and CFRP lap-shear joints.

关键词： Fe-SMA (memory steel) Cohesive zone model (CZM) Bond-slip behavior post-processing method Full-range behavior

来源：评论

学校读者我要写书评

暂无评论

CompNET: Boosting image recognition and writer identification via complementary neural network post-processing

引用

PATTERN RECOGNITION 2025年 157卷

作者： Zhao, Bocheng Cao, Xuan Zhang, Wenxing Liu, Xujie Miao, Qiguang Li, Yunan Xidian Univ Sch Comp Sci & Technol Xian 710071 Shaanxi Peoples R China Xidian Univ Guangzhou Inst Technol Guangzhou 510555 Guangdong Peoples R China Xidian Univ Xian Key Lab Big Data & Intelligent Vis Xian 710071 Shaanxi Peoples R China Xidian Univ Key Lab Collaborat Intelligence Syst Minist Educ Xian 710071 Shaanxi Peoples R China

In current classification tasks, an important method to improve accuracy is to pre-train the model using a large-scale domain-specific dataset. However, many tasks such as writer identification (writerID) lack suitable large-scale datasets in practical scenarios. To address this issue, this paper proposes a method that can improve prediction accuracy without relying on significant pre-training but leveraging the diversity of probability distributions predicted by multiple networks, and enhancing the top-1 accuracy through complementary post-processing. Specifically, top-k distributions are sampled from the multiple probability mass functions separately. When the distribution differences of top-k are maximized, the intersection other than the correct category can be narrowed down. Finally, the correct target with suboptimal probability can be rectified by the only intersection. Furthermore, our method has exhibited an intriguing trait during experimentation. Its prediction accuracy enhances concurrently with the incorporation of novel SOTA methods, ultimately surpassing the performance of these new methods.

关键词： Classification task Probability mass function Model integration post-processing method

来源：评论

学校读者我要写书评

暂无评论

Efficiency of post-processing in PMU based state estimation of renewable energy microgrids

Efficiency of post-processing in PMU based state estimation ...

引用

IEEE Belgrade PowerTech Conference

作者： Abed, Amir Alwan, Natheer Abed, Munther H. Dobric, Goran Univ Belgrade Sch Elect Engn Belgrade Serbia Univ Malaysia Pahang Coll Comp & Appl Sci Pahang Malaysia

ISBN: (纸本)9781665487788

Power System State Estimation (SE) is a process for determining the state of all the buses in a power system (voltage magnitude and angle) based on measurements taken at a selection of a few buses. Traditionally, the only information that measurement devices could provide was the magnitude of the measured signal. On the other hand, the Phasor Measurement Unit (PMU) can measure the current phasors of the directly linked lines as well as the voltage phasors (both angle and magnitude) of the bus at which it is located. However, achieving observability of the system using only PMU devices is very expensive. In order to determine the condition of a power system, phasor measurements are employed in addition to conventional measurements. In this paper, the use of PMU measurements to estimate the state of a renewable energy microgrid (REM) has been explained and the proposed method has been verified on IEEE 21 bus microgrid. The method makes use of PMU voltage and current data after post-processing, as well as a separate linear state estimator model that makes use of the state estimate from Weighted Least Square (WLS). Using the WLS state estimation approach from conventional data, the model first estimates the state in polar coordinates. This state is then combined with PMU measurements in rectangular coordinates, to predict the system's final state.

关键词： Phasor Measurement Unit State Estimation post-processing method Weighted Least Square

来源：评论

学校读者我要写书评

暂无评论

An Efficient method for Video Object Detection in Automatic Scoring of Physical Experimental Operations

引用

INTERNATIONAL JOURNAL ON ARTIFICIAL INTELLIGENCE TOOLS 2023年第8期32卷 2350040-2350040页

作者： Zeng, Wenbin Guo, Jichang Hao, Luguo Liu, Jianfei Tianjin Univ Sch Elect & Informat Engn Tianjin 300072 Peoples R China Guangdong Univ Technol Sch Informat Engn Guangzhou 510006 Guangdong Peoples R China Hebei Univ Technol Sch Elect & Informat Engn Tianjin Peoples R China

Automatic scoring of students' physical experimental operations is a very practical application which has not been researched deeply. The common method for automatic scoring of students' experimental operations is to infer the behavior of experimental operations through the state of experimental instruments. Video object detection is the basic task of detecting the state of experimental instruments, and the problem of missed detection or false detection in video multi-object detection is one of the main reasons leading to the error of automatic scoring results. However, existing methods of video object detection mainly improve the accuracy of the model in public datasets, which has the disadvantage of not correcting false detection while improving accuracy. Therefore, an efficient video object detection method composed of YOLOv5 and a logical reasoning post-processing method was proposed to fill this gap. We compared our method with other state-of-the-art methods on three independent datasets of physical experimental instruments. We established a pipeline for automatic scoring of students' experimental operations, designed flow charts and state score tables of three physics experiments, and compared the automatic scoring results with the average scores of six experimental teachers. The results show that our method is more robust and efficient in this application scenario. We hope this report can promote the application of logical reasoning methods in video object detection.

关键词： Video object detection post-processing method automatic scoring YOLOv5

来源：评论

学校读者我要写书评

暂无评论

Flexible margins and multiple samples learning to enhance lexical semantic similarity

引用

ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE 2024年第PartC期133卷

作者： Pan, Jeng-Shyang Wang, Xiao Yang, Dongqiang Li, Ning Huang, Kevin Chu, Shu-Chuan Shandong Univ Sci & Technol Coll Comp Sci & Engn Qingdao 266000 Peoples R China Chaoyang Univ Technol Dept Informat Management Taichung 41349 Taiwan Shandong Jianzhu Univ Sch Comp Sci & Technol Jinan 250000 Peoples R China Chung Yuan Christian Univ Dept Ind & Syst Engn Taoyuan 320314 Taiwan

The advancement of deep learning and neural networks has led to the widespread adoption of neural word embeddings as a prominent lexical representation method in natural language processing. With the help of the neural language model trained by the contextual information of large scale text, the neural word embedding obtained by the neural language model captures more semantic correlation in the semantic space, while ignoring the semantic similarity. It will incur high computational cost and time costs during the training process of the model. To better inject semantic similarity into the distribution space and reduce time cost, we perform post processing learning of neural word embeddings using deep metric learning. This paper proposes a lexical enhancement method based on flexible margins and multiple samples learning. In this method, we embed the lexical entailment constraint relations into neural word embeddings. By categorizing the set of lexical constraints and penalizing the negative samples to different degrees according to the gap between categories, and allowing the positive and negative samples to learn from each other in the distributed space. The method we propose significantly improves neural word embeddings. By evaluating neural word embedded vocabulary similarity, the benchmark accuracy is improved to 75%. The method shows great competitiveness in text similarity tasks and text categorization tasks. These findings summarize research results and provide strong support for further applications.

关键词： Semantic similarity Deep metric learning post-processing method Enhanced word embedding

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：