Image inpainting based on deep learning has been greatly improved.The original purpose of image inpainting was to repair some broken photos, suchas inpainting artifacts. However, it may also be used for malicious oper...Image inpainting based on deep learning has been greatly improved.The original purpose of image inpainting was to repair some broken photos, suchas inpainting artifacts. However, it may also be used for malicious operations,such as destroying evidence. Therefore, detection and localization of imageinpainting operations are essential. Recent research shows that high-pass filteringfull convolutional network (HPFCN) is applied to image inpainting detection andachieves good results. However, those methods did not consider the spatial location and channel information of the feature map. To solve these shortcomings, weintroduce the squeezed excitation blocks (SE) and propose a high-pass filter attention full convolutional network (HPACN). In feature extraction, we apply concurrent spatial and channel attention (scSE) to enhance feature extraction and obtainmore information. Channel attention (cSE) is introduced in upsampling toenhance detection and localization. The experimental results show that the proposed method can achieve improvement on ImageNet.展开更多
The combination of flexible sensors and bionic innovative design has become an important direction for the development of intelligent sensing technology.To this end,this paper systematically describes the latest resea...The combination of flexible sensors and bionic innovative design has become an important direction for the development of intelligent sensing technology.To this end,this paper systematically describes the latest research progress of bionic sensors inspired by the synergistic mechanism of“stress concentration-high pass filtering-omnidirectional localization”of scorpion slit receptors.First,it presents breakthroughs such as ultra-high sensitivity through gradient-cracked structures,dynamic signal decoupling mediated by viscoelastic materials,and omnidirectional localization accuracy supported by curvilinear array layouts.Aiming at the cross-interference and integration redundancy problems faced by traditional multimodal sensing systems,this paper introduces a vertically stacked heterogeneous integration strategy.Through the synergistic design of bionic stretchable conductive film and strain-isolated communication interfaces,a flexible multimodal sensing system with pressure-temperature bimodal sensing,multiaxial stress decoupling,and spatial distribution tracking capability is successfully constructed.Relevant research further confirms that the bionic architecture shows significant advantages in medical monitoring,industrial equipment health management and lunar rover terrain sensing scenarios.It provides a new paradigm of cross-scale structure-function synergistic optimization for the development of adaptive intelligent sensing systems in extreme environments,and marks an important leap in the integration of bionic flexible electronics from single-device innovation to systematic technology.展开更多
Active damping(AD)strategy is an economical and efficient method to solve the resonant problem of the permanent magnet synchronous motor(PMSM)drive system with inductor-capacitor(LC)sine wave filter.In this article,th...Active damping(AD)strategy is an economical and efficient method to solve the resonant problem of the permanent magnet synchronous motor(PMSM)drive system with inductor-capacitor(LC)sine wave filter.In this article,the AD methods used in PMSM drive system are classified as inherent damping(ID),state variable feedback,and digital filter.Based on this,the purpose of this article is to provide an overview and analysis of the AD methods on PMSM drive system in recent years,and to comprehensively review,compare,and summarize the stability,dynamic performance,robustness,and algorithm complexity.Furthermore,a new expansion of AD method based on capacitor current feedback with high-pass filter(HPF-CCF)is studied to ensure the effectiveness when the resonant frequency is around sixth of the sampling frequency.The simulation and experimental results validate the effectiveness of theoretical analysis.展开更多
基金supported by the National Natural Science Foundation of China under Grant 62172059,61972057 and 62072055Hunan Provincial Natural Science Foundations of China under Grant 2020JJ4626+1 种基金Scientific Research Fund of Hunan Provincial Education Department of China under Grant 19B004Postgraduate Scientific Research Innovation Project of Hunan Province under Grant CX20210811.
文摘Image inpainting based on deep learning has been greatly improved.The original purpose of image inpainting was to repair some broken photos, suchas inpainting artifacts. However, it may also be used for malicious operations,such as destroying evidence. Therefore, detection and localization of imageinpainting operations are essential. Recent research shows that high-pass filteringfull convolutional network (HPFCN) is applied to image inpainting detection andachieves good results. However, those methods did not consider the spatial location and channel information of the feature map. To solve these shortcomings, weintroduce the squeezed excitation blocks (SE) and propose a high-pass filter attention full convolutional network (HPACN). In feature extraction, we apply concurrent spatial and channel attention (scSE) to enhance feature extraction and obtainmore information. Channel attention (cSE) is introduced in upsampling toenhance detection and localization. The experimental results show that the proposed method can achieve improvement on ImageNet.
基金funded by the Foundation for Innovative Research Groups of the National Natural Science Foundation of China(No.52021003)“Fundamental Research Funds for the Central Universities”.
文摘The combination of flexible sensors and bionic innovative design has become an important direction for the development of intelligent sensing technology.To this end,this paper systematically describes the latest research progress of bionic sensors inspired by the synergistic mechanism of“stress concentration-high pass filtering-omnidirectional localization”of scorpion slit receptors.First,it presents breakthroughs such as ultra-high sensitivity through gradient-cracked structures,dynamic signal decoupling mediated by viscoelastic materials,and omnidirectional localization accuracy supported by curvilinear array layouts.Aiming at the cross-interference and integration redundancy problems faced by traditional multimodal sensing systems,this paper introduces a vertically stacked heterogeneous integration strategy.Through the synergistic design of bionic stretchable conductive film and strain-isolated communication interfaces,a flexible multimodal sensing system with pressure-temperature bimodal sensing,multiaxial stress decoupling,and spatial distribution tracking capability is successfully constructed.Relevant research further confirms that the bionic architecture shows significant advantages in medical monitoring,industrial equipment health management and lunar rover terrain sensing scenarios.It provides a new paradigm of cross-scale structure-function synergistic optimization for the development of adaptive intelligent sensing systems in extreme environments,and marks an important leap in the integration of bionic flexible electronics from single-device innovation to systematic technology.
基金supported in part by the National Natural Science Foundational of China under Grants 62373363 and 52007190
文摘Active damping(AD)strategy is an economical and efficient method to solve the resonant problem of the permanent magnet synchronous motor(PMSM)drive system with inductor-capacitor(LC)sine wave filter.In this article,the AD methods used in PMSM drive system are classified as inherent damping(ID),state variable feedback,and digital filter.Based on this,the purpose of this article is to provide an overview and analysis of the AD methods on PMSM drive system in recent years,and to comprehensively review,compare,and summarize the stability,dynamic performance,robustness,and algorithm complexity.Furthermore,a new expansion of AD method based on capacitor current feedback with high-pass filter(HPF-CCF)is studied to ensure the effectiveness when the resonant frequency is around sixth of the sampling frequency.The simulation and experimental results validate the effectiveness of theoretical analysis.
文摘矿井关闭后,其上覆岩层及地表会再次发生变形,影响建(构)筑物安全运营。由于关闭矿井缺乏监管,导致地表形变时空演化规律及预测预警模型研究不足。为此,提出了一种分布式目标雷达干涉测量(Distributed scatter Interferometric synthetic aperture radar,DS-InSAR)、时间域高通滤波(Temporal High Pass Filtering,THPF)以及长短期记忆网络(Long Short Term Memory Network,LSTM)相结合的关闭矿井地表形变预测模型。以98景Sentinel-1A升轨影像为数据源,首先利用DS-InSAR方法联合PS(Persistent Scatterer)和DS点获取徐州西部关闭矿井2019−11—2022−12的时序地表沉降信息;然后利用THPF分解原始沉降序列获取高频和低频序列沉降信息;之后采用LSTM完成高低频子序列的形变预测,将高低频子序列预测值叠加获取最终的预测结果。结果表明:DS-InSAR监测点密度分布均匀,监测结果水准实测形变间的决定系数达到0.95;相较于LSTM模型,THPF-LSTM模型在预测点位上的最大均方根误差(RMSE,Root Mean Square Error)为3.0,最大平均绝对误差(Mean Absolute Error,MAE)为2.4,最大校正决定系数(Adjusted R-square)为0.9,优于传统LSTM模型的4.5、3.9和0.6,模型综合预测精度提升20%以上,能够准确反映出关闭矿井地表形变的趋势和波动性,可有效提高短期内关闭矿井沉降预测精度,实现了关闭矿井地表形变监测和预测的一体化分析。
