检索结果-南通市图书馆

Video-based Person Re-identification Based on Distributed Cloud Computing

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Journal of Artificial Intelligence and Technology 2021年第2期1卷 110-120页

作者： Chengyan Zhong Xiaoyu Jiang guanqiu qi College of Automation Chongqing University of Posts and TelecommunicationsChongqing 400065China College of Intelligence Science and Technology Chongqing University of Posts and TelecommunicationsChongqing 400065China Computer Information Systems Department State University of New York at Buffalo StateBuffaloNY 14222USA

Person re-identification has been a hot research issues in the field of computer *** recent years,with the maturity of the theory,a large number of excellent methods have been ***,large-scale data sets and huge networks make training a time-consuming *** the same time,the parameters and their values generated during the training process also take up a lot of computer ***,we apply distributed cloud computing method to perform person re-identification *** distributed data storage method,pedestrian data sets and parameters are stored in cloud *** speed up operational efficiency and increase fault tolerance,we add data redundancy mechanism to copy and store data blocks to different nodes,and we propose a hash loop optimization algorithm to optimize the data distribution ***,we assign different layers of the re-identification network to different nodes to complete the training in the way of model *** comparing and analyzing the accuracy and operation speed of the distributed model on the video-based dataset MARS,the results show that our distributed model has a faster training speed.

关键词： person re-identification distributed cloud computing data redundancy mechanism

Multi-focus image fusion via morphological similarity-based dictionary construction and sparse representation

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

CAAI Transactions on Intelligence Technology 2018年第2期3卷 83-94页

作者： guanqiu qi qiong Zhang Fancheng Zeng Jinchuan Wang Zhiqin Zhu College of Automation Chongqing University of Posts and Telecommunications Chongqing 400065 People's Republic of China School of Computing Informatics and Decision Systems Engineering Arizona State University Tempe AZ 85287 USA

Sparse representation has been widely applied to multi-focus image fusion in recent years. As a key step, the construction of an informative dictionary directly decides the performance of sparsity-based image fusion. To obtain sufficient bases for dictionary learning, different geometric information of source images is extracted and analysed. The classified image bases are used to build corresponding subdictionaries by principle component analysis. All built subdictionaries are merged into one informative dictionary. Based on constructed dictionary, compressive sampling matched pursuit algorithm is used to extract corresponding sparse coefficients for the representation of source images. The obtained sparse coefficients are fused by Max-L1 fusion rule first, and then inverted to form the final fused image. Multiple comparative experiments demonstrate that the proposed method is competitive with other the state-of-the-art fusion methods.

关键词：图像融合图像分析发展现状智能技术

Convolutional neural network based detection and judgement of environmental obstacle in vehicle operation

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

CAAI Transactions on Intelligence Technology 2019年第2期4卷 80-91页

作者： guanqiu qi Huan Wang Matthew Haner Chenjie Weng Sixin Chen Zhiqin Zhu Department of Mathematics & Computer and Information Science Mansfield University of Pennsylvania Mansfield PA 16933 USA College of Automation Chongqing University of Posts and Telecommunications Chongqing 400065 People’s Republic of China

Precise real-time obstacle recognition is both vital to vehicle automation and extremely resource intensive. Current deep-learning based recognition techniques generally reach high recognition accuracy, but require extensive processing power. This study proposes a region of interest extraction method based on the maximum difference method and morphology, and a target recognition solution created with a deep convolutional neural network. In the proposed solution, the central processing unit and graphics processing unit work collaboratively. Compared with traditional deep learning solutions, the proposed solution decreases the complexity of algorithm, and improves both calculation efficiency and recognition accuracy. Overall it achieves a good balance between accuracy and computation.

关键词： recognition study computation

Associative learning mechanism for drug‐target interaction prediction

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

CAAI Transactions on Intelligence Technology 2023年第4期8卷 1558-1577页

作者： Zhiqin Zhu Zheng Yao guanqiu qi Neal Mazur Pan Yang Baisen Cong College of Automation Chongqing University of Posts and TelecommunicationsChongqingChina Computer Information Systems Department State University of New York at Buffalo StateBuffaloNew YorkUSA Department of Cardiovascular Surgery Chongqing General HospitalUniversity of Chinese Academy of SciencesChongqingChina Emergency Department The Second Affiliated Hospital of Chongqing Medical UniversityChongqingChina Data Scientist Diagnostics Digital DH(Shanghai)Diagnostics Co.Ltd.Danaher CompanyShanghaiChina

As a necessary process of modern drug development,finding a drug compound that can selectively bind to a specific protein is highly challenging and *** drug‐target interaction strength in terms of drug‐target affinity(DTA)is an emerging and effective research approach for drug ***,it is challenging to model drug‐target interactions in a deep learning manner,and few studies provide interpretable analysis of *** paper proposes a DTA prediction method(mutual transformer‐drug target affinity[MT‐DTA])with interactive learning and an autoencoder *** proposed MT‐DTA builds a variational autoencoders system with a cascade structure of the attention model and convolutional neural *** not only enhances the ability to capture the characteristic information of a single molecular sequence but also establishes the characteristic expression relationship for each substructure in a single molecular *** this basis,a molecular information interaction module is constructed,which adds information interaction paths between molecular sequence pairs and complements the expression of correlations between molecular *** performance of the proposed model was verified on two public benchmark datasets,KIBA and Davis,and the results confirm that the proposed model structure is effective in predicting ***,attention transformer models with different configurations can improve the feature expression of drug/protein *** model performs better in correctly predicting interaction strengths compared with state‐of‐the‐art *** addition,the diversity of drug/protein molecules can be better expressed than existing methods such as SeqGAN and Co‐VAE to generate more effective new *** DTA value prediction module fuses the drug‐target pair interaction information to output the predicted value of ***,this paper theoretically proves that the proposed method maximises evidence lower bound

关键词： deep learning medical applications

A Fully Convolutional Neural Network-based Regression Approach for Effective Chemical Composition Analysis Using Near-infrared Spectroscopy in Cloud

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Journal of Artificial Intelligence and Technology 2021年第1期1卷 74-82页

作者： Daiyu Jiang Gang Hu guanqiu qi Neal Mazur College of Automation Chongqing University of Posts and TelecommunicationsChongqing 400065China Computer Information Systems Department State University of New York at Buffalo StateBuffaloNY 14222 USA

As one chemical composition,nicotine content has an important influence on the quality of tobacco *** and nondestructive quantitative analysis of nicotine is an important task in the tobacco ***-infrared(NIR)spectroscopy as an effective chemical composition analysis technique has been widely *** this paper,we propose a one-dimensional fully convolutional network(1D-FCN)model to quantitatively analyze the nicotine composition of tobacco leaves using NIR spectroscopy data in a cloud *** 1D-FCN model uses one-dimensional convolution layers to directly extract the complex features from sequential spectroscopy *** consists of five convolutional layers and two full connection layers with the max-pooling layer replaced by a convolutional layer to avoid information *** computing techniques are used to solve the increasing requests of large-size data analysis and implement data sharing and *** results show that the proposed 1D-FCN model can effectively extract the complex characteristics inside the spectrum and more accurately predict the nicotine volumes in tobacco leaves than other *** research provides a deep learning foundation for quantitative analysis of NIR spectral data in the tobacco industry.

关键词： nicotine tobacco leaves near-infrared spectroscopy fully convolutional network cloud computing

A Novel Person Re-ID Method based on Multi-Scale Feature Fusion

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

A Novel Person Re-ID Method based on Multi-Scale Feature Fus...

Blockchain and Artificial Intelligence Applications

第三十九届中国控制会议

作者： Yuanyuan Li Xiaofei Wang Zhiqin Zhu Xinghua Huang Penghua Li guanqiu qi Yujun Rong Chongqing University of Posts and Telecommunications Key Laboratory of Complex System Safety and Control Ministry of EducationChongqing University Computer Information Systems Department Buffalo State College China Mobile (Hangzhou) Information Technology Co. Ltd.

Person re-identification（Re-ID） is a challenging but significant research topic. Previous methods based on global features are not pay much attention to the fine-grained feature information. While the methods based on local features strictly rely on the accurate pedestrian detection bounding-boxes to align the image pairs. In this paper, the Multi-Scale Feature Fusion Network is proposed which can extract global and local features simultaneously and to learn jointly. The network structure consists of two branches. One is a global branch for global feature learning, in this branch, we extract the whole body feature information of human body to express the pedestrian image as a whole. The other is a local branch for local feature learning. In the local branch, according to the characteristics of pedestrian’s human structure, we segment the image into 6 horizontal strips to bring the local feature representation of the pedestrian image. In addition, we use a novel integration of multiple loss functions to further improve the recognition accuracy of the network. Experiments on Market-1501 and DukeMTMC-reID datasets show that our proposed method can achieve state-of-the-art results.

关键词： Person re-identification Multi-scale feature fusion Feature learning Multiple loss functions

来源： cnki会议评论

在线全文

cnki会议

学校读者我要写书评

暂无评论

Journal of Artificial Intelligence and Technology 2021年第2期1卷 83-83页

作者： guanqiu qi Zhiqin Zhu Computer Information Systems Department State University of New York at Buffalo StateBuffaloNY 14222USA College of Automation Chongqing University of Posts and TelecommunicationsChongqing 400065China

The emergence and popularity of blockchain,distributed ledger technology distributed computing,and network security and trust techniques are significantly changing the operation and management of computing and communication systems,as these techniques have the potential to disrupt any domain involving coordination among autonomous resources without trusted third *** techniques and their applications include finance and payments(e.g.,Facebook Libra),but also networks(e.g.,power grids or telecom networks),computing(e.g.,brokering of edge resources),Internet of Things(e.g.,supply chain or industry 4.0),and service platforms(e.g.,identity management).The market capitalization,investor appetite,and institutional coverage for cryptocurrency(as well as bitcoin and blockchain)have all jumped *** total market capitalization of the cryptocurrency market has significantly increased in the past three *** applications of blockchain exhibit a variety of complicated problems and new requirements,which brings more open issues and challenges for artificial intelligence(AI)and related research.

关键词： artificial finance autonomous

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

图像分解与色彩先验下的多曝光图像融合

中国图象图形学报 2021年第12期26卷 2800-2812页

作者：李嫄源王琴朱智勤齐观秋重庆邮电大学计算机科学与技术学院重庆400065 重庆邮电大学自动化学院重庆400065 纽约州立大学布法罗分校计算机信息系统学院纽约14222

目的多曝光图像融合(multi-exposure fusion,MEF)是利用一组不同曝光度的低动态范围(low dynamic range,LDR)图像进行合成,得到类似高动态范围(high dynamic range,HDR)图像视觉效果图像的过程。传统多曝光图像融合在一定程度上存在图... 详细信息

目的多曝光图像融合(multi-exposure fusion,MEF)是利用一组不同曝光度的低动态范围(low dynamic range,LDR)图像进行合成,得到类似高动态范围(high dynamic range,HDR)图像视觉效果图像的过程。传统多曝光图像融合在一定程度上存在图像细节信息受损、边界不清晰以及部分色彩失真等问题。为了充分综合待融合图像的有效信息,提出了一种基于图像分解和色彩先验的双尺度多曝光图像融合方法。方法使用快速导向滤波进行图像分解,分离出细节层对其进行增强处理,保留更多的细节信息,同时减少融合图像的光晕伪影;根据色彩先验,利用亮度和饱和度之差判断图像曝光程度,并联合亮度与饱和度之差以及图像对比度计算多曝光图像融合权重,同时保障融合图像的亮度和对比度;利用导向滤波对权重图进行优化,抑制噪声,增加像素之间的相关性,提升融合图像的视觉效果。结果在24组多曝光图像序列上进行实验,从主观评价角度来看,该融合方法能够提升图像整体对比度及色彩饱和度,并兼顾过曝光区域和欠曝光区域的细节提升。从客观评价标准分析,采用两种不同的多曝光图像序列融合结果的质量评估算法,评价结果显示融合性能均有所提高,对应的指标均值分别为0.982和0.970。与其他对比算法的数据结果比较,在两种不同的结构相似性指标上均有所提升,平均提升分别为1.2%和1.1%。结论通过主观和客观评价,证实了所提方法在图像对比度、色彩饱和度以及细节信息保留的处理效果十分显著,具有良好的融合性能。

关键词：多曝光图像融合(MEF) 高动态范围成像导向滤波快速导向滤波色彩先验

基于联合分布偏移差异的跨域滚动轴承故障诊断方法

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

电子测量与仪器学报 2022年第10期36卷 146-156页

作者：雷杨博朱智勤柴毅齐观秋安翼尧徐鹏重庆邮电大学自动化学院重庆400065 重庆大学自动化学院重庆400030 布法罗州立学院计算机信息系统系美国纽约14222

现有的无监督域自适应故障诊断方法大多只基于单一域信号实现,提取的故障信息不够全面。只注重实现源域和目标域特征的边缘分布对齐,忽略了样本的条件分布差异,限制了诊断精度的提升。为克服以上问题,提出一种基于联合分布偏移差异(join... 详细信息

现有的无监督域自适应故障诊断方法大多只基于单一域信号实现,提取的故障信息不够全面。只注重实现源域和目标域特征的边缘分布对齐,忽略了样本的条件分布差异,限制了诊断精度的提升。为克服以上问题,提出一种基于联合分布偏移差异(joint distribution offset difference,JDOD)的跨域滚动轴承故障诊断方法。使用两个结构一致的卷积神经网络(CNN)分别提取信号的时域与频域特征,获取更完整的故障信息。提出联合分布偏移差异,实现不同域特征的边缘分布对齐和条件分布对齐。在两个多工况轴承数据集上与多种先进方法展开对比实验,取得了99%以上的平均诊断精度。实验结果表明联合分布偏移差异有效提升了跨域故障精度。

关键词：跨域故障诊断域自适应分布对齐卷积神经网络