检索结果-内蒙古大学图书馆

RJAN:Region-based joint attention network for 3D shape recognition

CAAI Transactions on Intelligence technology 2025年第2期10卷 460-473页

作者： Yue Zhao Weizhi Nie Jie Nie Yuyi Zhang Bo Wang Hangzhou Institute of Technology Xidian UniversityHangzhouChina School of Electronic Engineering Key Laboratory of Collaborative Intelligence SystemsMinistry of EducationXidian UniversityXi'anChina School of Electrical and Information Engineering Tianjin UniversityTianjinChina Ocean University of China QingdaoChina State Key Laboratory of Digital Multimedia Technology HisenseChina

As an essential field of multimedia and computer vision,3D shape recognition has attracted much research attention in recent ***-based approaches have demonstrated their superiority in generating effective 3D shape *** methods usually extract the multiview global features and aggregate them together to generate 3D shape ***,there exist two disadvantages:First,the mainstream methods ignore the comprehensive exploration of local information in each ***,many approaches roughly aggregate multiview features by adding or concatenating them *** information loss for some discriminative characteristics limits the representation *** address these problems,a novel architecture named region-based joint attention network(RJAN)was ***,the authors first design a hierarchical local information exploration module for view descriptor *** region-to-region and channel-to-channel relationships from different granularities can be comprehensively explored and utilised to provide more discriminative characteristics for view feature ***,a novel relation-aware view aggregation module is designed to aggregate the multiview features for shape descriptor generation,considering the view-to-view *** experiments were conducted on three public databases:ModelNet40,ModelNet10,and *** achieves state-of-the-art performance in the tasks of 3D shape classification and 3D shape retrieval,which demonstrates the effectiveness of *** code has been released on https://***/slurrpp/RJAN.

关键词： 3D shape recognition attention mechanism multiview

来源：评论

学校读者我要写书评

暂无评论

Code-Based Conjunction Obfuscation

引用

Chinese Journal of Electronics 2023年第2期32卷 237-247页

作者： ZHANG Zheng ZHANG Zhuoran ZHANG Fangguo School of Computer Science and Engineering Sun Yat-sen University Guangdong Province Key Laboratory of Information Security Technology

The pattern-matching problem with wildcards can be formulated as a conjunction where an accepting string is same as the pattern for all non-wildcards. A scheme of conjunction obfuscation is a algorithm that “encrypt” the pattern to prevent some adversary from forging any accepting string. Since 2013, there are abundant works about conjunction obfuscation which discussed with weak/strong functionality preservation and distributed black-box security. These works are based on generic group model, learning with error assumption,learning with noise assumption, etc. Our work proposes the first conjunction obfuscation with strong functionality preservation and distributed black-box security from a standard assumption. Our scheme with some parameter constraints can also resist some related attacks such as the information set decoding attack and the structured error arrack.

关键词： Computers Quantum computing Computational modeling Closed box Resists Linear codes Mathematical models

来源：评论

学校读者我要写书评

暂无评论

Adjustable patch and feature prior token-based transformer for weakly supervised semantic segmentation

引用

International Journal of Computers and Applications 2024年第7期46卷 445-454页

作者： Li, Linjuan Zhang, Haoxue Xie, Gang Bai, Yanhong School of Electronic and Information Engineering Taiyuan University of Science and Technology Taiyuan China Shanxi Key Laboratory of Advanced Control and Equipment Intelligence Taiyuan China

Weakly supervised semantic segmentation is a challenging task, utilizing only low-cost weak supervision to produce pixel-level predictions. Existing transformer-based methods for weakly supervised semantic segmentation have some limitations: (1) Fixing patch size might destroy the structured semantics, which is unfriendly to objects of different scales, and (2) Ignoring the prior features when using multi-head attention mechanisms might lead to inaccurate segmentation localization. To tackle these issues, we proposed an effective transformer framework coupled with the adjustable patch and prior feature tokens, termed as APFPformer, in which an adjustable patch module is developed to split the image according to the area of the salient object for preserving the structured semantics in patches. A prior feature token module is devised to exploit the edge and texture information as prior tokens, ensuring the gain of discriminative representation. Additionally, a single-stage scheme is applied to reduce the computation and rapid segmentation process. Our experiments demonstrate the superiority of our approach over early methods, gaining competitive mean Intersection-over-Union scores of 67.9% on the PASCAL VOC2012 dataset and 40.2% on the MS COCO dataset. © 2024 Informa UK Limited, trading as Taylor & Francis Group.

关键词： Semantic Segmentation

来源：评论

学校读者我要写书评

暂无评论

Layered double hydroxide-derived Mg_(2)Ni/TiH_(1.5)composite catalysts for enhancing hydrogen storage performance of MgH_(2)

引用

Journal of Magnesium and Alloys 2024年第12期12卷 4966-4975页

作者： Gang Huang Yao Lu Xiaofang Liu Wukui Tang Xinyu Li Feng Wang Jianglan Shui Ronghai Yu School of Materials Science and Engineering Beihang UniversityNo.37 Xueyuan RoadBeijing 100191PR China School of Materials Science and Engineering&Guangxi Key Laboratory of Information Materials Guilin University of Electronic TechnologyGuilin 541004PR China

Developing efficient catalysts is of great significance in improving the sluggish kinetics and high desorption temperature of Mg H_(2)hydrogen storage ***,ultrathin Ni Ti-layered double hydroxide(Ni Ti-LDH)nanosheets are used as precursors to prepare Mg_(2)Ni/TiH_(1.5)composite catalysts to improve the hydrogen storage properties of MgH_(2).The variation of Ni/Ti ratio in LDH plays an important role in regulating the composition,morphology and distribution of Mg_(2)Ni/Ti H_(1.5)catalysts,which significantly affect their synergistic catalytic ***_(2)Ni/TiH_(1.5)composite catalyst exhibits significantly improved catalytic performance compared with conventional Ni-,Ti-and Ni/Ti-based *** optimal Mg H_(2)/Mg_(2)Ni/TiH_(1.5)system shows a significantly reduced desorption temperature of 212℃which is 133℃lower than that of pure MgH_(2)(345℃),and can release 5.97 wt%hydrogen within 300s at 300℃.Further mechanism analysis reveals that the unique flaky morphology and suitable composition of Ni/Ti LDH can significantly enhance the synergistic effect of Mg_(2)Ni and TiH_(1.5),which promotes the fracture of the H–H and Mg-H bonds.

关键词： Hydrogen storage materials MgH_(2) Layered double hydroxides Nanocatalysts

来源：评论

学校读者我要写书评

暂无评论

Copy-move detection method based on Decoupled Edge Supervision and multi-domain cross correlation modeling

引用

Multimedia Tools and Applications 2024年第18期84卷 1-23页

作者： Jing, Niantai Nie, Jie Wang, Jingyu Wang, Xiaodong Liang, Xinyue Gao, Xuesong Faculty of Information Science and Engineering Ocean University of China QingDao China Qingdao Hisense State Key Laboratory of Digital Multimedia Technology QingDao China

Copy move forgery detection is defined as the process of moving one region in the image (the source region) to another region in the image (the tampered region). Common classification methods include copy move forgery location (CMFL, which does not distinguish between source regions and tampered regions) while existing copy move source target detection (CMSTD) use limited information to distinguish the source and region. Since the edges of the tampered region able to act as the important clues such as blurring, bringing the challenge of detecting the copy-move forgery target with the complete structure. Therefore, we propose a Copy-move Detection Method based on Decoupled Edge Supervision and Multi-domain Cross Correlation (DM-Net), including Multi-scale Similar Region Detection module (MSD), Decoupled Edge Supervision module (DEM), and Multi-domain Correlation Modeling module (MCM), which can overcome the problem that the tamper trace is fuzzy caused by the post-processing operation. Specifically, the MSD module is proposed to extract coarse similar regions by multi-scale method. The DEM module is proposed to extract the tamper region by the method of decoupling edge supervision, which avoids information redundancy while using shallow edge features. The MCM module conducts cross-correlation modeling between the tampered, source and similar region, further optimizes detection targets of similar region by mining the correlation among multiple domains. By adding edge information, we can improve the efficiency of distinguishing source and target regions by 2%. We performed experiments on USC-ISI data set, and the accuracy was improved by 0.21% compared with CNN-T GAN method, and the F1-score index was improved by 0.87% compared with DOA-GAN. The accuracy of CASIA v2.0 data set is 2.89% higher than that of Busternet method, and the precision index is 3.98% higher than that of CMSD-STRD method on source. The accuracy of CoMoFoD data set is improved by 0.93% compared with

关键词： Classification (of information)

来源：评论

学校读者我要写书评

暂无评论

Hydrogel electrodes with conductive and substrate-adhesive layers for noninvasive long-term EEG acquisition

引用

Microsystems & Nanoengineering 2023年第3期9卷 357-370页

作者： Hailing Xue Dongyang Wang Mingyan Jin Hanbing Gao Xuhui Wang Long Xia Dong’ang Li Kai Sun Huanan Wang Xufeng Dong Chi Zhang Fengyu Cong Jiaqi Lin Key State Laboratory of Fine Chemicals School of BioengineeringDalian University of Technology116024 DalianChina School of Biomedical Engineering Faculty of Electronic Information and Electrical EngineeringDalian University of Technology116024 DalianChina Key Laboratory of Energy Materials and School of Materials Science and Engineering Dalian University of Technology116024 DalianChina

Noninvasive brain–computer interfaces (BCIs) show great potential in applications including sleep monitoring, fatigue alerts, neurofeedback training, etc. While noninvasive BCIs do not impose any procedural risk to users (as opposed to invasive BCIs), the acquisition of high-quality electroencephalograms (EEGs) in the long term has been challenging due to the limitations of current electrodes. Herein, we developed a semidry double-layer hydrogel electrode that not only records EEG signals at a resolution comparable to that of wet electrodes but is also able to withstand up to 12 h of continuous EEG acquisition. The electrode comprises dual hydrogel layers: a conductive layer that features high conductivity, low skin-contact impedance, and high robustness;and an adhesive layer that can bond to glass or plastic substrates to reduce motion artifacts in wearing conditions. Water retention in the hydrogel is stable, and the measured skin-contact impedance of the hydrogel electrode is comparable to that of wet electrodes (conductive paste) and drastically lower than that of dry electrodes (metal pin). Cytotoxicity and skin irritation tests show that the hydrogel electrode has excellent biocompatibility. Finally, the developed hydrogel electrode was evaluated in both N170 and P300 event-related potential (ERP) tests on human volunteers. The hydrogel electrode captured the expected ERP waveforms in both the N170 and P300 tests, showing similarities in the waveforms generated by wet electrodes. In contrast, dry electrodes fail to detect the triggered potential due to low signal quality. In addition, our hydrogel electrode can acquire EEG for up to 12 h and is ready for recycled use (7-day tests). Altogether, the results suggest that our semidry double-layer hydrogel electrodes are able to detect ERPs in the long term in an easy-to-use fashion, potentially opening up numerous applications in real-life scenarios for noninvasive BCI.

关键词： conductive electrode impedance

来源：评论

学校读者我要写书评

暂无评论

Near-Lossless Compression of Two Independent Binary Sources with One Encoder 16

Near-Lossless Compression of Two Independent Binary Sources ...

引用

16th International Conference on Wireless Communications and Signal Processing, WCSP 2024

作者： Zhu, Tingting Ma, Xiao Sun Yat-sen University School of Computer Science and Engineering MoE Key Laboratory of Information Technology Guangdong Key Laboratory of Information Security Technology Guangzhou China

ISBN: (纸本)9798350390643

In this paper, we consider the near-lossless compression of two distinct but independent binary random sources U and V, both of which are asymmetrical independent and identically distributed (i.i.d.) sources. In particular, V is assumed to be sparser than U. We show that, to efficiently compress U and V, explicit source coding (compression) for V (the sparser one) is not required. The basic idea is to encode U with an error-correction code and then superimpose V on the codeword. Hence, only one encoder-decoder pair (EDP) suffices to compress two sources. Theoretical analysis shows that the proposed coding scheme is entropy-approaching in terms of bit-error rate (BER) of both sources, and numerical results verify our analysis. Furthermore, we analyze the complexity of compression using the block Markov superposition transmission-repetition (BMST-R) codes to discuss the advantage of the proposed scheme. Simulation results demonstrate that the proposed scheme compresses the two sources simultaneously in a more compact way but causes no performance degradation. © 2024 IEEE.

关键词： Convolutional codes

来源：评论

学校读者我要写书评

暂无评论

Plasma sprayed thermal barrier coatings with outstanding ultrahigh temperature thermal shock resistances through gradient porosity level optimizations

引用

Ceramics International 2025年第13期51卷 17484-17491页

作者： Liu, Liqiang Xu, Wenhu Shi, Junmiao Zhang, Xian-Cheng Tu, Shan-Tung Key Laboratory of Pressure Systems and Safety Ministry of Education School of Mechanical and Power Engineering East China University of Science and Technology Shanghai200237 China School of Mechanical and Electrical Engineering Changchun University of Science and Technology Changchun130022 China

In this study, the YSZ thermal barrier coatings (TBCs) with gradient porosity was developed through design the spraying parameters. The effect of the gradient porosity on the bond strength and thermal shock life was investigated, and the failure mechanism of the TBCs was explored. This article prepared three coatings with different porosities, tested the bonding strength of different porosities using the tensile method, and tested the thermal shock resistance of the coatings using water quenching after thermal shock. By simulating the residual stress inside the coatings after spraying, the failure mechanism of the coatings after thermal shock was analyzed. With the decrease of the porosity from TBC I to TBC III, the residual stress concentration after the spraying raised. Subsequently, the bond strength of the TBCs decreased form 37 MPa–31 MPa owing to the residual stress evolution. All of the three TBCs underwent three thermal shock test without macroscopic spalling. During the thermal shock test, tensile horizontal stress formed and promoted the formation of vertical cracks. Tensile vertical stress concentrated at the top - coat/bond - coat (TC/BC) interface significantly, inducing horizontal crack adjacent to the interface. Additionally, the TC layer surface zone sintered under the oxygen-acetylene flame heating. Then tensile vertical stress concentration was formed in the sintering zone region, and horizontal crack was detected in the TC layer. The connection of the vertical cracks and horizontal cracks was responsible for the spalling and failure of the TC layer. © 2025

关键词： Spalling

来源：评论

学校读者我要写书评

暂无评论

Additive Noise Model Structure Learning Based on Spatial Coordinates

IEEE Transactions on Artificial Intelligence

引用

IEEE Transactions on Artificial Intelligence 2024年第8期5卷 3858-3871页

作者： Yang, Jing Lu, Ting Zhu, Youjie Hefei University of Technology Intelligent Interconnected Systems Laboratory of Anhui Province Hefei230009 China Hefei University of Technology School of Computer Science and Information Engineering Hefei230009 China

Discovering causal relationships from a large amount of observational data is an important research direction in data mining. To address the challenges of discovering and constructing causal networks on nonlinear and high-dimensional data, this article proposes a new structural learning algorithm called spatial coordinates based (SCB). The SCB algorithm effectively discovers causal networks from large scale nonlinear data. In this article, we make three main contributions. Firstly, based on the Hilbert-Schmidt independence criterion (HSIC), we propose new independence test coefficients: spatial coordinate (SC) coefficient and conditional spatial coordinate (CSC) coefficient. We also prove that the statistical distribution of the CSC coefficient follows a standard normal distribution. Secondly, using the statistical distribution of the CSC coefficient, we redefine the correlation of variables and combine it with local learning to propose the SCB algorithm. Finally, we demonstrate the effectiveness of the algorithm through experiments on data generated from various nonlinear function dependencies. Compared to existing algorithms, the causal network models constructed using the SCB algorithm exhibit significant improvements in accuracy and time performance. The effectiveness of the proposed method is further validated on real power plant data. © 2020 IEEE.

关键词： Additive noise

来源：评论

学校读者我要写书评

暂无评论

H_(2)O_(2) treatment boosts activity of NiFe layered double hydroxide for electro-catalytic oxidation of urea

引用

Journal of Environmental sciences 2023年第7期129卷 152-160页

作者： Jinshan Wei Jin Wang Xiaoming Sun College of Materials Science and Engineering Key Laboratory of Optoelectronic Devices and SystemsCollege of Physics and Optoelectronic EngineeringShenzhen UniversityShenzhen 518060China State Key Laboratory of Chemical Resource Engineering College of ChemistryBeijing University of Chemical TechnologyBeijing 100029China

Urea oxidation reaction(UOR)provides a method for hydrogen production besides wastewater treatment,but the current limited catalytic activity has prevented the ***,we develop a novel H_(2)O_(2) treatment strategy for tailoring the surface oxygen ligand of NiFe-layered double hydroxides(NiFe-LDH).The sample after H_(2)O_(2) treatment(NiFeOLDH)shows significant enhancement on UOR efficiency,with the potential of 1.37 V(RHE)to reach a current density of 10 mA/cm^(2).The boost is attributed to the richness adsorption O ligand on NiFeO-LDH as revealed by XPS and Raman *** calculation indicates formation of two possible types of oxygen ligands:adsorbed oxygen on the surface and exposed from hydroxyl group,lowered the desorption energy of CO_(2) product,which lead to the lowered onset *** strategy is further extended to NiFe-LDH nano sheet on Ni foam to reach a higher current density of 440 mA/cm^(2) of UOR at 1.8 V(RHE).The facile surface O ligand manipulation is also expected to give chance to many other electro-catalytic oxidations.

关键词： NiFe-LDH Urea oxidation reaction H_(2)O_(2)treatment Electro-catalysis Hydrogen production

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：