检索结果-内蒙古大学图书馆

Efficient linear fusion of partial estimators

DIGITAL SIGNAL processing 2018年 78卷 265-283页

作者： Luengo, David Martino, Luca Elvira, Victor Bugallo, Monica Univ Politecn Dept Signal Theory & Commun Madrid Spain Univ Valencia Image Proc Lab Valencia Spain IMT Lille Douai Villeneuve Dascq France CNRS CRIStAL UMR 9189 Villeneuve Dascq France SUNY Stony Brook Dept Elect & Comp Engn Stony Brook NY 11794 USA

Many signal processing applications require performing statistical inference on large datasets, where computational and/or memory restrictions become an issue. In this big data setting, computing an exact global centralized estimator is often either unfeasible or impractical. Hence, several authors have considered distributed inference approaches, where the data are divided among multiple workers (cores, machines or a combination of both). The computations are then performed in parallel and the resulting partial estimators are finally combined to approximate the intractable global estimator. In this paper, we focus on the scenario where no communication exists among the workers, deriving efficient linear fusion rules for the combination of the distributed estimators. Both a constrained optimization perspective and a Bayesian approach (based on the Bernstein-von Mises theorem and the asymptotic normality of the estimators) are provided for the derivation of the proposed linear fusion rules. We concentrate on finding the minimum mean squared error (MMSE) global estimator, but the developed framework is very general and can be used to combine any type of unbiased partial estimators (not necessarily MMSE partial estimators). Numerical results show the good performance of the algorithms developed, both in problems where analytical expressions can be obtained for the partial estimators, and in a wireless sensor network localization problem where Monte Carlo methods are used to approximate the partial estimators. (C) 2018 Elsevier Inc. All rights reserved.

关键词： distributed estimation Linear fusion Constrained minimization Big data Bayesian estimation Bernstein-von Mises theorem

来源：评论

学校读者我要写书评

暂无评论

A Video Semantic Segmentation Method Based on FCN and Data Argumentation

A Video Semantic Segmentation Method Based on FCN and Data A...

引用

IEEE International Conference on Big Data and Cloud Computing (BdCloud)

作者： Yuan Huang Qian Huang Qinglong Chen Yanping Li Xiaoqing Sun School of Computer and Information Hohai University Nanjing China College of Computer and Information Science Fujian Agriculture and Forestry University Fuzhou China

ISBN: (数字)9781728143286

ISBN: (纸本)9781728143293

Video semantic segmentation is an important and fundamental problem in computer vision. It has broad application prospects in the fields of mobile robot, drone, intelligent driving and monitoring. With the development of neural networks, the models commonly adopted are all based on full convolutional network (FCN). However, current methods are limited by a small training set, which makes it difficult to improve the segmentation accuracy. In this paper, we propose a robust method that uses different data argumentation methods to increase the data set according to different characteristics of the scene. On the basis of analyzing different video features, targeted data argumentation techniques are selected to increase training samples. Experimental results show that data argumentation techniques can significantly improve the accuracy of video semantic segmentation compared with traditional training methods that ignore video features.

关键词： image segmentation Semantics Training Convolution Deep learning image color analysis Neural networks

来源：评论

学校读者我要写书评

暂无评论

基于Hadoop的源相机识别方法研究

引用

现代计算机 2020年第18期26卷 88-92页

作者：朱新龙王佳斌工业智能化与系统福建省高校工程研究中心泉州362021

探索使用分布式方法解决数字图像取证中的关键问题——源相机识别问题的可能性。提出的解决方案是基于Lukas等人的算法,然后使用Hadoop框架中的MapReduce编程模型进行并行处理。使用几个优化来提高Lukas等人算法在Hadoop平台上的性能。... 详细信息

探索使用分布式方法解决数字图像取证中的关键问题——源相机识别问题的可能性。提出的解决方案是基于Lukas等人的算法,然后使用Hadoop框架中的MapReduce编程模型进行并行处理。使用几个优化来提高Lukas等人算法在Hadoop平台上的性能。初步试验结果显示,经过优化后的算法在Hadoop平台上的运行时间与单台计算机运行时间相比,分布式处理能够有效缩短任务的处理时间。

关键词：源摄像机识别分布式计算 Hadoop MapReduce

来源：评论

学校读者我要写书评

暂无评论

MARGIN-EMBEDDING CANONICAL CORRELATION ANALYSIS WITH FEATURE SELECTION FOR PERSON RE-IDENTIFICATION 25

MARGIN-EMBEDDING CANONICAL CORRELATION ANALYSIS WITH FEATURE...

引用

25th IEEE International Conference on image processing (ICIP)

作者： Ma, Linfei Zhang, Xiang Lan, Long Huang, Xuhui Luo, Zhigang NUDT Sci & Technol Parallel & Distributed Lab Changsha 410073 Hunan Peoples R China NUDT Coll Comp Changsha 410073 Hunan Peoples R China NUDT State Key Lab High Performance Comp Changsha 410073 Hunan Peoples R China NUDT Dept Comp Sci & Technol Changsha 410073 Hunan Peoples R China

ISBN: (纸本)9781479970612

Canonical correlation analysis (CCA) is a classical subspace learning method of capturing the common semantic information underlying multi-view data. It has been used in person re-identification (re-ID) task by treating the task of matching identical individuals across non-overlapping multi-cameras as a multi-view learning problem. However, CCA-based re-ID methods still achieve unsatisfactory results because few jointly consider discriminative margin information and selecting importantly relevant features. To address this issue, we propose a novel l(2,1)-norm regularized margin-embedding CCA (l(2,1)-MCCA), which learns a generalized discriminative subspace by employing more discriminative margin information. Moreover, the new method enforces the l(2,1)-norm regularization term over the learned subspace to identify the relevant features. Both lightweight and effective schemes can benefit from each other and endeavor to enlarge the interclass variations whilst reducing the intra-class variations. Experiments on three popular datasets show the efficacy of l(2,1)-MCCA as compared with recently representative re-ID methods.

关键词： Canonical correlation analysis l(2,1)-norm regularization margin embedding person re-identification

来源：评论

学校读者我要写书评

暂无评论

GAI: A Centralized Tree-Based Scheduler for Machine Learning Workload in Large Shared Clusters 18th

GAI: A Centralized Tree-Based Scheduler for Machine Learning...

引用

18th International Conference on Algorithms and Architectures for parallel processing (ICA3PP)

作者： Gao, Ce Ren, Rui Cai, Hongming Shanghai Jiao Tong Univ Sch Software Shanghai Peoples R China

ISBN: (纸本)9783030050542;9783030050535

With widespread applications in image recognition, language translation, computer vision and other areas, deep learning (DL) have been proliferating over the past decade. Practitioners from different business groups in industries train DL models on a shared cloud computing infrastructure for these applications with different priorities. During the model training process, one of the key challenges is to minimize the life-cycle of high priority model training jobs. This paper analyzes the distributed training of machine learning (ML) models and identifies short board effect in the training process: GPU training requires higher network bandwidth compared to CPU training. The key insight motivates the design of GAI, a centralized scheduler for ML workload. It relies on two techniques: (1) tree-based structure. The structure stores the cluster information hierarchically to apply multi-layer scheduling. (2) well-extended priority algorithm. We consider priorities from multiple dimensions for model training jobs comprehensively to support resource degradation and preemption. The prototype of GAI is implemented on top of Kubernetes, Kubeflow, and TensorFlow. It is evaluated using a simulator and a real cloud-based cluster. Evaluations show 28% increase in scheduling throughput and 21% training convergence speedup on DL models.

关键词： Resource management distributed machine learning Centralized scheduling Resource utilization

来源：评论

学校读者我要写书评

暂无评论

SMIM: Superpixel Mutual Information Measurement for image Quality Assessment 18th

SMIM: Superpixel Mutual Information Measurement for Image Qu...

引用

18th International Conference on Algorithms and Architectures for parallel processing (ICA3PP)

作者： Wang, Jiaming Lu, Tao Zhang, Yanduo Wuhan Inst Technol Sch Comp Sci & Engn Hubei Key Lab Intelligent Robot Wuhan 430205 Peoples R China

ISBN: (纸本)9783030050542;9783030050535

The image quality assessment (IQA) is a fundamental problem in signal processing that aims to measure the objective quality of an image by designing a mathematical model. Most full-reference (FR) IQA methods use fixed sliding windows to obtain structure information but ignore the variable spatial configuration information. In this paper, we propose a novel full-reference IQA method, named "superpixel normalized mutual information (SMIM)" based on the perspective of variable receptive field and information entropy. First, we find that consistence relationship exists between the information fidelity and human visual of individuals. Thus, we reproduce the human visual system (HVS) to semantically divide the image into multiple patches via superpixel segmentation. Then the weights of each image patches are adaptively calculated via its information volume. We verified the effectiveness of SMIM by applying it to data from the TID2008 database and data generated using some real application scenarios. Experiments show that SMIM outperforms some state-of-the-art FR IQA algorithms, including visual information fidelity (VIF).

关键词： image quality assessment Mutual information Superpixel segmentation

来源：评论

学校读者我要写书评

暂无评论

Robust Feature Matching via Two Constraints Interaction

Robust Feature Matching via Two Constraints Interaction

引用

IEEE International Conference on Big Data and Cloud Computing (BdCloud)

作者： Yizhang Liu Yanping Li Zheling Xu Xiaoliang Lu Xin Liu Luanyuan Dai Xiong Pan Changcai Yang College of Computer and Information Science Fujian Agriculture and Forestry University Fuzhou China College of Computer and Information Hohai University Nanjing China International Business Department Fuzhou Melbourne Polytechnic Fuzhou China MingByte Technology (Qingdao) Co. Ltd

ISBN: (数字)9781728143286

ISBN: (纸本)9781728143293

Feature matching is a critical step in a variety of vision-based applications. However, it is difficult to customize a general method due to complicated image transformations. In this paper, we formulate feature matching into a mathematical problem, which is simple and does not rely on any specific model. Specifically, we construct an objective function based on two constraints (i.e., geometric constraint and consistency constraint). Different from the previous work, we exploit the interaction of these two constraints to reject outliers. Besides, we adopt an iterative strategy to boost the number of inliers. To accelerate the matching process, we use the parallel computing to obtain the information of K nearest neighbors for each feature point. Experimental results on ten image pairs (undergo different image transformations) compared with other state-of-the-arts demonstrate our superiority both in accuracy and efficiency.

关键词： Linear programming Information science Agriculture Forestry Feature extraction Iterative methods Feature detection

来源：评论

学校读者我要写书评

暂无评论

Sequential Detection of Flying Small Target in Infrared images Based on Generalized Low-Rank Background Estimation

Sequential Detection of Flying Small Target in Infrared Imag...

引用

IEEE International Symposium on parallel and distributed processing with Applications and IEEE International Conference on Ubiquitous Computing and Communications (ISPA/IUCC)

作者： Chuanyun Wang Keyi Si Tian Wang Zhaokui Li Na Lin Ershen Wang Shenyang Aerospace University Shenyang China Beihang University Beijing China

ISBN: (纸本)9781728152103

Addressing the problems of flying small target detection in infrared image sequences, a new approach is proposed based on generalized low-rank background estimation. Firstly, the generalized low rank approximations is introduced to model infrared background image over sequential infrared images. Subsequently, the foreground target image is achieved by background subtraction with the generalized low-rank background estimation. Finally, flying small target detection is achieved over separated target image by threshold segmentation. The experiment results on two infrared image sequences of flying plane demonstrate that the proposed method have effective detection performance and outperform the baseline methods in precision and recall evaluation.

关键词： approximation theory estimation theory image segmentation image sequences infrared imaging object detection Infrared imaging Object detection estimation theory Approximation Theory image sequences sequential detection image segmentation Flying target image baseline method Infrared radiation

来源：评论

学校读者我要写书评

暂无评论

Blind Video Quality Assessment Based on Human Visual Characteristic

Blind Video Quality Assessment Based on Human Visual Charact...

引用

IEEE International Symposium on parallel and distributed processing with Applications and IEEE International Conference on Ubiquitous Computing and Communications (ISPA/IUCC)

作者： Shiyu Zhou Xiqing Liu Yuanbing Tian Bin Fu Xiuyan Xia Jin Ge China Unicom Network Technology Research Institute Beijing China Zhiyou Open Source Communication Research Institute (Beijing) Co. Ltd Beijing China Beijing University of Posts and Telecommunications Beijing China

ISBN: (纸本)9781728152103

Based human visual speed perception characteristic, a video quanlity assessment(VQA) algorithm is introduced in the paper. Natural video statistics features extraction and weighting factors are incorporated in the scheme. In the VQA, considering the impact both video content itself and HVS's characteristic on human subjective perception, we propose weighting factors to scale the effect of those features, and it contains two parts: motion information and perception noise. And natural video statistics features relate to the spatial and temporal domain are extracted. The weighting factors would be used to combine both the temporal and spatial features, then generate the quality of each frame. Finally, the video quality score can be obtained by pooling scheme. LIVE database, EPFL-PoliMI database and some other generated test videos were used in our experiments, and the results indicate our model has outstanding performance.

关键词： feature extraction image denoising image motion analysis statistical analysis video signal processing visual databases image motion analysis visual databases Video signal processing image denoising Feature extraction Videotapes Quality assessment Statistical methods Statistical analysis WEIGHTING FACTORS

来源：评论

学校读者我要写书评

暂无评论

Loop Closure Detection Based on Improved Hybrid Deep Learning Architecture

Loop Closure Detection Based on Improved Hybrid Deep Learnin...

引用

IEEE International Symposium on parallel and distributed processing with Applications and IEEE International Conference on Ubiquitous Computing and Communications (ISPA/IUCC)

作者： Yuan Liu Rong Xiang Qieshi Zhang Ziliang Ren Jun Cheng Guilin University of Electronic Technology Guilin China Key Laboratory of Human-Machine Intelligence-Synergy Shenzhen China

ISBN: (纸本)9781728152103

Loop closure detection plays a vital role in the visual Simultaneous Localization and Mapping (SLAM) system. Traditional Bag-of-Visual-Word (BoVW) methods for loop closure detection has been successfully applied to various SLAM systems. The Convolutional Neural Networks (ConvNets) based methods can automatically learn feature representation from original image, and it is more robust to illumination changes. However, these two methods all have some limitations. In order to overcome the shortcomings of the traditional methods extract insufficient features, and the ConvNets based methods unable to adapt complex environment changes, in this paper, we propose an improved hybrid deep learning architecture (HDLA) to generate high-level semantic image features for loop closure detection specifically. This network is based on hybrid ConvNet modified by us to especially cope with robust and real-time feature extraction. We present extensive experiments on three real world datasets to evaluate each of the specific challenges in loop closure detection. The results demonstrate that the proposed method achieves superior performances and provides a feasible solution for loop closure detection.

关键词： convolutional neural nets feature extraction image classification image representation learning (artificial intelligence) object detection robot vision SLAM (robots) robot vision Feature extraction image representation Object detection image classification surgical closure Loops Contraceptive Coil Feasible Solution original image simultaneous localization and mapping

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：