检索结果-内蒙古大学图书馆

47th IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

作者： Liu, Yiming Zhang, Yanni Li, Qiang Kong, Jun Qi, Miao Wang, Jianzhong Northeast Normal Univ Coll Informat Sci & Technol Changchun Peoples R China Hunan Normal Univ Coll Informat Sci & Engn Changsha Peoples R China Changchun Humanities & Sci Coll Inst Intelligent Elderly Care Changchun Peoples R China

ISBN: (纸本)9781665405409

In this paper, a deep learning network with double closed-loop structure is introduced to tackle the image deblurring problem. The first closed-loop in our model is composed of two networks which learn a pair of opposite mappings between the blurry and sharp images. By this way, the solution spaces of possible functions that map a blurry image to its sharp counterpart can be effectively reduced. Furthermore, the first closed-loop also helps our model to deal with the unpaired samples in the training set. The second closed-loop in the proposed approach employed a self-supervision mechanism to constrain the features of intermedia layers in the network, so that the detailed information of sharp images can be well exploited. Through combining the two closed-loops together, our model can address the limitations of existing methods and improve the deblurring performance. Extensive experiments on both benchmark and real-world datasets show that the proposed network achieves state-of-the-art performance. The code will be released in: https://***/LiQiang0307/DCLNet.

关键词： Image deblurring closed-loop structure cycle consistency solution space

来源：评论

学校读者我要写书评

暂无评论

Innovative design of a separatable coil for tube frame structure in magnetic pulse crimping

引用

JOURNAL OF MANUFACTURING PROCESSES 2024年 125卷 538-551页

作者： Jiang, Hao Du, Weidong Wang, Shaoluo Sun, Hao Luo, Kang Li, Guangyao Cui, Junjia Hunan Univ State Key Lab Adv Design & Mfg Technol Vehicle Changsha 410082 Peoples R China Shenzhen Automot Res Inst Beijing Inst Technol Shenzhen 518057 Peoples R China

Magnetic pulse crimping (MPC) employs magnetic forces for high-speed metal forming, offering superior reliability compared to bolted joints and conventional fusion welding. However, the widely utilized devices often feature a monolithic structure and tend to be larger in volume, resulting in notable limitations regarding flexibility. This leads to a degree of complexity when attempting to join the closed-loop structure. This study proposed a highly flexible coil, characterized by a separatable working area, which could be employed to join complex structures. Numerical simulations were performed to analyze the feasibility of the separatable coil structure and the crimping parameters of the workpiece. The results revealed that the coil was reduced in size and reliable in strength. The velocity and displacement parameters displayed by the tube were well within the intended crimping range, thus confirming that the coil had basic functionality. The crimping tests on AA5052 tubes and rods confirmed the operational performance of the coil. Subsequently, the mechanical properties of the joints were evaluated through quasi-static tensile tests. Meanwhile, laser microscopy provided a detailed characterization of the interface morphology. The specimens developed two distinct types of joint mechanisms. At 24 kJ and below, the joints were joined by embedded mechanical interlocking. At exceeding 26 kJ, the observation of continuous interfacial waves highlighted the coil's capability in magnetic pulse welding. Finally, the typical engineering samples of AA5052-ZS304 dissimilar materials were successfully fabricated by using the separatable coil. The coil had demonstrated its flexibility in joining complex structures, effectively overcoming the constraints of traditional crimping devices.

关键词： Magnetic pulse crimping Coil design Flexible manufacturing closed-loop structure

来源：评论

学校读者我要写书评

暂无评论

Bioinspired Jumping Soft Actuators of the Liquid Crystal Elastomer Enabled by Photo-Mechanical Coupling

引用

ACS APPLIED MATERIALS & INTERFACES 2023年第1期16卷 1596-1604页

作者： Lei, Bing Wen, Zhi-Yuan Wang, Hua-Kun Gao, Jing Chen, Lu-Jian Xiamen Univ Sch Elect Sci & Engn Dept Elect Engn Fujian Key Lab Ultrafast Laser Technol & Applicat Xiamen 361005 Peoples R China Xiamen Univ Sch Architecture & Civil Engn Dept Civil Engn Fujian Key Lab Digital Simulat Coastal Civil Engn Xiamen 361005 Peoples R China

Jumping, a fundamental survival behavior observed in organisms, serves as a vital mechanism for adapting to the surrounding environment and overcoming significant obstacles within a given terrain. Here, we present a light-controlled soft jumping actuator inspired by asphondylia, which employs a closed-loop structure and utilizes a liquid crystal elastomer (LCE). Photo-mechanical coupling highlights the significant influence of the light source on the actuator's jumping behavior. Manipulating the light intensity, the relative position of stimulus and light lock, and the concentration of disperse red 1 (DR1) allows precise control over both the maximum take-off velocity and jump height. Furthermore, tailoring the size of the LCE actuator offers a means of regulating jumping behavior. Upon exposure to 460 nm LED irradiation, our actuator achieves remarkable performance, with a maximum jumping height of 10 body length (BL) and take-off velocity of 62 BL/s. These actuators accumulate and rapidly release energy, enabling the effective transportation of microcargos across substantial distances. Our research yields valuable insights into the realm of soft robotics, underscoring the pivotal importance of photo-mechanical coupling in the field of soft robotics, thereby serving as a catalyst for inspiring continued exploration of agile and capable systems by prestoring elastic energy.

关键词： liquid crystal elastomer photo-mechanicalcoupling closed-loop structure elastic energy jumpingsoft actuator

来源：评论

学校读者我要写书评

暂无评论

AVO Inversion Based on closed-loop Multitask Conditional Wasserstein Generative Adversarial Network

引用

IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING 2023年 61卷

作者： Wang, Zixu Wang, Shoudong Zhou, Chen Cheng, Wanli China Univ Petr State Key Lab Petr Resources & Prospecting Natl Engn Lab Offshore Oil Explorat Beijing 102249 Peoples R China

Neural networks are commonly used for poststack and prestack seismic inversion. With sufficient labeled data, the neural network-based seismic inversion results are more accurate than that use traditional seismic inversion methods. However, in the case of insufficient labeled data, the accuracy of neural networks-based seismic inversion results decreases and is even lower than those based on the traditional inversion methods. In addition, the seismic inversion results based on neural networks generally suffer from lateral discontinuity. It further reduces the accuracy of the inversion results. To tackle these problems, we propose a prestack seismic amplitude variation with offset (AVO) inversion method based on closed-loop multitask conditional Wasserstein generative adversarial network (CMcWGAN), which is a generative adversarial network (GAN)-based AVO inversion method. CMcWGAN enables simultaneous and accurate inversion of P-wave velocity (V p), S-wave velocity (Vs), and density (rho). Moreover, it uses the low-frequency information of elastic parameters as a conditional input to alleviate the problem of lateral discontinuity in inversion results. Experimental results of simulated data show that the inversion results based on CMcWGAN have higher accuracy than those based on the traditional AVO inversion methods. In addition, when the seismic angle gather is noisy, CMcWGAN has better robustness than the traditional methods. CMcWGAN can also obtain reasonable AVO inversion results in field seismic angle gather data. Experimental results of simulated data show that the inversion results based on CMcWGAN have higher accuracy than those based on the traditional AVO inversion method. In addition, when the seismic angle gather is noisy, CMcWGAN has better robustness than the traditional method. CMcWGAN can also get reasonable AVO inversion results in field seismic angle gather data.

关键词： Generative adversarial networks Multitasking Feature extraction Robustness Convolutional neural networks Impedance Convolution Amplitude variation with offset (AVO) inversion closed-loop structure conditional generative adversarial network (cGAN) Wasserstein generative adversarial network (WGAN)

来源：评论

学校读者我要写书评

暂无评论

A closed-loop Network for Single Infrared Remote Sensing Image Super-Resolution in Real World

引用

REMOTE SENSING 2023年第4期15卷

作者： Zhang, Haopeng Zhang, Cong Xie, Fengying Jiang, Zhiguo Beihang Univ Sch Astronaut Dept Aerosp Informat Engn Beijing 102206 Peoples R China AVIC DIGITAL Beijing 100028 Peoples R China Beijing Key Lab Digital Media Beijing 102206 Peoples R China Minist Educ Key Lab Spacecraft Design Optimizat & Dynam Simula Beijing 102206 Peoples R China

Single image super-resolution (SISR) is to reconstruct a high-resolution (HR) image from a corresponding low-resolution (LR) input. It is an effective way to solve the problem that infrared remote sensing images are usually suffering low resolution due to hardware limitations. Most previous learning-based SISR methods just use synthetic HR-LR image pairs (obtained by bicubic kernels) to learn the mapping from LR images to HR images. However, the underlying degradation in the real world is often different from the synthetic method, i.e., the real LR images are obtained through a more complex degradation kernel, which leads to the adaptation problem and poor SR performance. To handle this problem, we propose a novel closed-loop framework that can not only make full use of the learning ability of the channel attention module but also introduce the information of real images as much as possible through a closed-loop structure. Our network includes two independent generative networks for down-sampling and super-resolution, respectively, and they are connected to each other to get more information from real images. We make a comprehensive analysis of the training data, resolution level and imaging spectrum to validate the performance of our network for infrared remote sensing image super-resolution. Experiments on real infrared remote sensing images show that our method achieves superior performance in various training strategies of supervised learning, weakly supervised learning and unsupervised learning. Especially, our peak signal-to-noise ratio (PSNR) is 0.9 dB better than the second-best unsupervised super-resolution model on PROBA-V dataset.

关键词： super-resolution closed-loop structure channel attention infrared remote sensing image deep learning

来源：评论

学校读者我要写书评

暂无评论

A PVT Compensated Resistance to Frequency Converter for Sensor Array Read-Out

引用

IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS II-EXPRESS BRIEFS 2022年第5期69卷 2418-2422页

作者： Yan, Ming Colombo, Dalton Martini Freund, Michael S. El-Sankary, Kamal Dalhousie Univ Dept Elect & Comp Engn Halifax NS B3H 4R2 Canada Univ Fed Minas Gerais Dept Elect Engn BR-31270901 Belo Horizonte MG Brazil

This brief presents a PVT (process, voltage, temperature) compensated CMOS resistor to frequency read-out circuit for resistive sensing applications. Resistive sensing is used in different applications, such as olfactory sensing. An accurate read-out circuit is needed to detect the resistance values for on-chip sensor arrays. Existing techniques suffer from PVT variations and require a large silicon chip area. The proposed design offers a simple, accurate, PVT insensitive sensor read-out circuit where a closed-loop configuration, a switched capacitor resistor, operational amplifier integrator, and a voltage-controlled oscillator are used. The design is implemented and fabricated using TSMC 180 nm CMOS technology. The value for the measured sensor's resistance ranges between 10k and 100M Omega with an accuracy of 0.1% while the chip area is 0.0455mm(2) with 98.1 mu W power consumption. Also, the settling time for the step resistance change is about 2 mu s.

关键词： Resistive sensor sensor array read-out circuit PVT variation robust closed-loop structure VCO relaxation oscillator

来源：评论

学校读者我要写书评

暂无评论

Well-Logging Constrained Seismic Inversion Based on closed-loop Convolutional Neural Network

引用

IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING 2020年第8期58卷 5564-5574页

作者： Wang, Yuqing Ge, Qiang Lu, Wenkai Yan, Xinfei Tsinghua Univ Beijing Natl Res Ctr Informat Sci & Technol BNRis State Key Lab Intelligent Technol & Syst Inst Artificial Intelligence THUAI Beijing 100084 Peoples R China Tsinghua Univ Dept Automat Beijing 100084 Peoples R China China Natl Petr Corp CNPC Res Inst Petr Explorat & Dev Beijing 100083 Peoples R China

Seismic inversion is a process of predicting high-resolution stratigraphic parameters from low-resolution seismic data. Traditional inversion methods tend to impose human prior knowledge, such as sparsity, to the modeling of the seismic inversion process. Nowadays, with the development of deep learning, the idea of modeling by learning from data has gained great attention in varieties of research fields. As a data-driven method, an artificial neural network (ANN) has already been explored by many researchers in the field of seismic inversion. Compared to ANN, a convolutional neural network (CNN) has a stronger learning ability attributing to its sophisticated structures. However, the development of CNN is limited by the amount of labeled data in many industrial fields including the field of seismic inversion. In order to mitigate the dependence of CNN on the amount of labeled data, we propose a closed-loop CNN structure in this article. The proposed closed-loop CNN can model the seismic forward and inversion process simultaneously from the training data set. Compared to traditional CNN, which is in an open-loop form, closed-loop CNN can not only learn from labeled data but also extract information contained in unlabeled data. The experimental results show that the closed-loop CNN has a better performance than both traditional methods and other deep learning-based methods on the synthetic data set and also can be efficiently applied on the real seismic data set.

关键词： closed-loop structure convolutional neural network (CNN) seismic inversion

来源：评论

学校读者我要写书评

暂无评论

Image segmentation encryption algorithm with chaotic sequence generation participated by cipher and multi-feedback loops

引用

MULTIMEDIA TOOLS AND APPLICATIONS 2021年第9期80卷 13821-13840页

作者： Deng, Jie Zhou, Minjun Wang, Chunhua Wang, Sicheng Xu, Cong Hunan Univ Coll Comp Sci & Elect Engn Changsha 410082 Peoples R China

The existing chaotic image encryption algorithms have common defects: (i) ciphertext does not participate in the generation processes of chaotic pseudo-random sequences and key sequences;(ii) the entire encryption process does not have a closed-loop structure. In order to solve above problems, in this paper, an image segmentation encryption algorithm based on hyperchaotic system is proposed. We decompose the scrambled sequence into three sequences of different lengths: S-1, S-2 and S-3. Then, the initial values of the chaotic system are updated by the sequences S-2 and S-3 and using the updated initial value iterates the chaotic system and generates the key sequence K-3, and the sequence S-1 is encrypted by the sequence K-3 to obtain the cipher sequence C-1, using the sequences C-1 and S-3 updates the initial value of the chaotic system, and using the updated initial value iterates the chaotic system and generates the key sequence K-4, and using the sequence K-4 encrypts the sequence S-2 to obtain the cipher sequence C-2. Thus, ciphertext participates in the generation processes of chaotic pseudo-random sequences and key sequences, and the entire encryption process has a closed-loop structure. The experimental results show that the encryption algorithm has high security and sensitivity.

关键词： Chaotic image encryption Segmentation encryption Hyperchaotic system Ciphertext feedback closed-loop structure

来源：评论

学校读者我要写书评

暂无评论

Covariance regulation based invariant Kalman filtering for attitude estimation on matrix Lie groups

引用

IET CONTROL THEORY AND APPLICATIONS 2021年第15期15卷 2017-2025页

作者： Wang, Jiaolong Li, Minzhe Jiangnan Univ Inst Automat Key Lab Adv Proc Control Light Ind 1800 Lihu Ave Wuxi 214122 Jiangsu Peoples R China Shanghai Jiao Tong Univ Sch Aeronaut & Astronaut Shanghai Peoples R China

For matrix Lie groups attitude estimation problems with the trouble of unknown/inaccurate process noise covariance, by elaborating the proportion based covariance regulation scheme, this work proposes a novel version of adaptive invariant Kalman filter (AIKF). Invariant Kalman filter (IKF) takes into account the group geometry and can give better results than Euclidean Kalman filters, but it still heavily depends on the accuracy of noise statistics parameters. To ease this constraint, IKF's covariance propagation step is removed and a proportional regulation scheme is elaborated for the proposed AIKF: the feedback of posterior sequence is introduced to construct a closed-loop structure of covariance propagation, and then a proportional regulator is employed to amplify the feedback and accelerate the convergence of covariance calibration. As the main benefit, implementation of new AIKF does not require the accurate knowledge of noise statistics, which is also the main advantage over IKF. The mathematical derivation of proposed covariance regulation scheme is presented and the numerical simulations of the Lie groups attitude estimation problem are used to certify the filtering performance of the new approach.

关键词： Lie groups proposed covariance regulation scheme statistical analysis proportional regulation scheme adaptive invariant Kalman filter closed-loop structure Euclidean Kalman filters Lie groups attitude estimation problem covariance matrices Kalman filters noise statistics parameters covariance calibration group geometry IKF covariance propagation step filtering performance matrix Lie groups attitude estimation problems proportion based covariance regulation scheme AIKF proportional regulator

来源：评论

学校读者我要写书评

暂无评论

UB-Net: Improved Seismic Inversion Based on Uncertainty Backpropagation

引用

IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING 2022年 60卷

作者： Ma, Qiming Wang, Yuqing Ao, Yile Wang, Qi Lu, Wenkai Tsinghua Univ Beijing Natl Res Ctr Informat Sci & Technol BNRIS Inst Artificial Intelligence THUAI State Key Lab Intelligent Technol & Syst Beijing 100084 Peoples R China Tsinghua Univ Dept Automat Beijing 100084 Peoples R China

Seismic inversion is aimed at building a mapping from low-resolution seismic data to high-resolution impedance data. Most of the traditional methods have satisfactory interpretability, and most parameters tend to have specific physical definitions. On the other hand, deep learning-based methods present poor interpretability as their prediction performance is not always clearly explainable. One of the significant challenges of deep learning-based methods is to quantify the uncertainty of the model. The uncertainty includes aleatoric uncertainty and epistemic uncertainty, and epistemic uncertainty can be used to evaluate the predicted accuracy of the trained model. In this article, we propose a new deep learning model called uncertainty backpropagation network (UB-Net) to perform impedance inversion. The proposed UB-Net is based on a closed-loop framework and can predict the impedance and the epistemic uncertainty simultaneously. UB-Net has three closed-loop data flows, whereby the predicted uncertainty is utilized as the weight of loss functions to improve the inversion accuracy. Experimental analyses demonstrate that UB-Net presents advanced inversion accuracy on both synthetic and real examples. In particular, the mean absolute error (MAE) on synthetic examples drops by 40%, and the Pearson correlation coefficient (PCC) on real examples increases by 2%. Besides, compared with existing approaches, UB-Net presents superior spatial continuity and preserves more geological structures such as little faults in real examples.

关键词： Uncertainty Impedance Convolutional neural networks Training Bayes methods Data models Predictive models closed-loop structure deep evidential regression seismic inversion uncertainty

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：