检索结果-内蒙古大学图书馆

电光系统 2015年第3期 11-14页

作者：秦臻中国电子科技集团公司第二十七研究所郑州450047

为解决遥感影像处理过程中存在的计算密集型以及数据密集型问题，深入研究GPU集群的体系结构，遥感平台类型和遥感并行处理方法等。采用了基于CUDA的GPU并行运算平台和基于Hadoop的分布式并行运算平台，重点分析了几种遥感数据处理并行... 详细信息

为解决遥感影像处理过程中存在的计算密集型以及数据密集型问题，深入研究GPU集群的体系结构，遥感平台类型和遥感并行处理方法等。采用了基于CUDA的GPU并行运算平台和基于Hadoop的分布式并行运算平台，重点分析了几种遥感数据处理并行方法。最终得出基于集群的遥感并行方法较传统方法有着明显优势，并对遥感影像处理系统并行计算前景和研究热点问题进行了展望。

关键词：遥感并行集群大数据

来源：评论

学校读者我要写书评

暂无评论

Research on Gridding algorithm of Remote Sensing image Segmentation by Region Growth 3

Research on Gridding algorithm of Remote Sensing Image Segme...

引用

3rd International Conference on Advanced Engineering Materials and Architecture Science (ICAEMAS)

作者： Zhu, Hongchun Liu, Haiying Jiang, Tao Shandong Univ Sci & Technol Geomat Coll Qingdao 266590 Peoples R China

ISBN: (纸本)9783038352679

Region growth is an important method based on regional information, and its gridding process also is an important problem. This research is satisfied with the needs of the remote sensing massive data real time and fast transmission and processing, and developed of the theory and methods research of remote sensing processing based on grid operation. This paper starts with basic principle and mathematical models of region growth. Then some new serial region growth algorithms proposed in recent years are classified and analyzed, based on which their related gridding strategies are emphatically discussed. Then this research analyze and put forward the algorithm, which should be feasible when designing gridding parallel segmentation, and give an evaluation of the parallel regional segment algorithm. Finally, the problems and challenges of future research were pointed out.

关键词： region growth image segmentation gridding parallelism algorithm

来源：评论

学校读者我要写书评

暂无评论

Proposal of Eye-gaze Recognize Method for Input Interface without Infra-red Ray Equipment 15

Proposal of Eye-gaze Recognize Method for Input Interface wi...

引用

15th IEEE/ACIS International Conference on Software Engineering, Artificial Intelligence, Networking and parallel/distributed Computing (SNPD)

作者： Fukushima, Kazuki Shirahama, Naruki Kitakyushu Natl Coll Technol Dept Control Engn Adv Engn Kitakyushu Fukuoka Japan Kitakyushu Natl Coll Technol Dept Elect & Control Kitakyushu Fukuoka Japan

ISBN: (纸本)9781479956043

The main purpose of this study is to develop the inexpensive eye-gaze input interface for disabled people. The eye-gaze inputting can suit many situations, and it has little load for user because it is non-contact input interface. Many projects about the eye-gaze have been studied recently. Most of the productions of eye-gaze inputting use infra-red rays (IR) to detect the iris in the eyes. However the harm of IR for human eyes has been pointed out. In addition, if the interface requires the camera which has IR, the users must purchase the specific devices. Therefore we adopted a PC and the camera doesn't have the IR function. We propose the system by using the motion-template that is one of the functions of OpenCV library for tracking motion. However this function can't recognize the point where the user watches on monitor. It can recognize only the motion. That's why we made calibrate function to relate the eye-gaze with the monitor. We propose two methods to recognize eye-gaze. Both methods require the iris binary image. This image shows the iris shape and we expect that user's visual point can be calculated from this iris image. One method uses gravity point to calculate the point. Other method uses the rectangular approximation to calculate it. We did experiments for some subjects by both methods and compared the results to validate which method is proper and how much the calibrate function is accurate. In the experiment, the function randomly spotted a blue target on the monitor. The target position changes on a regular basis. In this experiment, the user stares at the target and we check the accuracy. If this function or methods are proper, the function correctly recognizes the user stare at target or near. For more accuracy, we will consider about how to detect the iris correctly in the future.

关键词： assistive technology eye-gaze web-camera image processing welfare support-equipment OpenCV

来源：评论

学校读者我要写书评

暂无评论

distributed non-convex ADMM-inference in large-scale random fields 25

Distributed non-convex ADMM-inference in large-scale random ...

引用

25th British Machine Vision Conference, BMVC 2014

作者： Miksik, Ondrej Vineet, Vibhav Pérez, Patrick Torr, Philip H.S. Department of Engineering Science University of Oxford Oxford United Kingdom Technicolor Research and Innovation Cesson Sévigné France

We propose a parallel and distributed algorithm for solving discrete labeling problems in large scale random fields. Our approach is motivated by the following observations: i) very large scale image and video processing problems, such as labeling dozens of million pixels with thousands of labels, are routinely faced in many application domains;ii) the computational complexity of the current state-of-the-art inference algorithms makes them impractical to solve such large scale problems;iii) modern parallel and distributed systems provide high computation power at low cost. At the core of our algorithm is a tree-based decomposition of the original optimization problem which is solved using a non convex form of the method of alternating direction method of multipliers (ADMM). This allows efficient parallel solving of resulting sub-problems. We evaluate the efficiency and accuracy offered by our algorithm on several benchmark lowlevel vision problems, on both CPU and Nvidia GPU. We consistently achieve a factor of speed-up compared to dual decomposition (DD) approach and other ADMM-based approaches. © 2014. The copyright of this document resides with its authors.

关键词： Trees (mathematics)

来源：评论

学校读者我要写书评

暂无评论

A System for Large-Scale Analysis of distributed Cameras

A System for Large-Scale Analysis of Distributed Cameras

引用

IEEE Global Conference on Signal and Information processing (GlobalSIP)

作者： Kaseb, Ahmed S. Berry, Everett Koh, Youngsol Mohan, Anup Chen, Wenyi Li, He Lu, Yung-Hsiang Delp, Edward J. Purdue Univ Sch Elect & Comp Engn W Lafayette IN 47907 USA

ISBN: (纸本)9781479970889

Thousands of cameras are connected to the Internet providing streaming data (videos or periodic images). The images contain information that can be used to determine the scene contents such as traffic, weather, and the environment. Analyzing the data from these cameras presents many challenges, such as (i) retrieving data from geographically distributed and heterogeneous cameras, (ii) providing a software environment for users to simultaneously analyze large amounts of data from the cameras, (iii) allocating and managing computation and storage resources. This paper presents a system designed to address these challenges. The system enables users to execute image analysis and computer vision techniques on a large scale with only slight changes to the existing methods. It currently includes more than 65,000 cameras deployed worldwide. Users can select cameras for the types of analysis they can do. The system allocates Amazon EC2 and Windows Azure cloud instances for executing the analysis. Our experiments demonstrate that this system can be used for a variety of image analysis techniques (e.g. motion analysis and human detection) using 2.7 million images from 1274 cameras for three hours using 15 cloud instances to analyze 141 GB of images (at 107 Mbps).

关键词： Network Camera image Analysis Big Data Cloud Computing

来源：评论

学校读者我要写书评

暂无评论

Building High Performance System for processing a Daily Large Volume of Chinese Satellites imagery 4

Building High Performance System for Processing a Daily Larg...

引用

Conference on High-Performance Computing in Remote Sensing IV

作者： Deng, Huawu Huang, Shicun Wang, Qi Pan, Zhiqiang Xin, Yubin Natl Capital Off PCI Geomat Quebec City PQ J9J 0L6 Canada

ISBN: (纸本)9781628413106

The number of Earth observation satellites from China increases dramatically recently and those satellites are acquiring a large volume of imagery daily. As the main portal of image processing and distribution from those Chinese satellites, the China Centre for Resources Satellite Data and Application (CRESDA) has been working with PCI Geomatics during the last three years to solve two issues in this regard: processing the large volume of data (about 1,500 scenes or 1 TB per day) in a timely manner and generating geometrically accurate orthorectified products. After three-year research and development, a high performance system has been built and successfully delivered. The high performance system has a service oriented architecture and can be deployed to a cluster of computers that may be configured with high end computing power. The high performance is gained through, first, making image processing algorithms into parallel computing by using high performance graphic processing unit (GPU) cards and multiple cores from multiple CPUs, and, second, distributing processing tasks to a cluster of computing nodes. While achieving up to thirty (and even more) times faster in performance compared with the traditional practice, a particular methodology was developed to improve the geometric accuracy of images acquired from Chinese satellites (including HJ-1 A/B, ZY-1-02C, ZY-3, GF-1, etc.). The methodology consists of fully automatic collection of dense ground control points (GCP) from various resources and then application of those points to improve the photogrammetric model of the images. The delivered system is up running at CRESDA for pre-operational production and has been and is generating good return on investment by eliminating a great amount of manual labor and increasing more than ten times of data throughput daily with fewer operators. Future work, such as development of more performance-optimized algorithms, robust image matching methods and application workflows

关键词： High Performance Computing (HPC) parallel Computing distributed Computing GPU CPU System Satellite imagery Orthorectification Geometric Accuracy Ground Control Points (GCP)

来源：评论

学校读者我要写书评

暂无评论

大规模空间数据融合分析关键技术研究

引用

科技资讯 2016年第1期14卷 171-172页

作者：刘洪元毕淼张楠北京大学北京天耀宏图科技有限公司中标软件有限公司

首先,研究了基于HBASE存储技术的栅格数据,海量栅格数据存储拟基于分布式文件和分布式内存对象系统,采用纵向分层、横向分块的方式,将不同分辨率的栅格数据逐级等分,切分成256×256像素小影像数据,由顶向底构成数字影像分辨率逐渐... 详细信息

首先,研究了基于HBASE存储技术的栅格数据,海量栅格数据存储拟基于分布式文件和分布式内存对象系统,采用纵向分层、横向分块的方式,将不同分辨率的栅格数据逐级等分,切分成256×256像素小影像数据,由顶向底构成数字影像分辨率逐渐增高的金字塔模型。由于遥感影像文件通常数据量较大,需要较大的内存时通常会采用压缩编码来减少存储空间,在GIS中多采用信息无损编码,而在对原始遥感影像进行压缩编码时,有时也采取有损压缩,拟采用Run-Length codes和Pack Bits想结合的方法对影像数据进行压缩,以提高压缩效率。其次,在研究地址模糊匹配算法的基础上,根据工商行业地址数据库特定存储格式,选取适于工商行业地址的拆分方案,提出一种分级结构的地址库的模糊匹配策略,以一票否决和模糊匹配共存的方式进行企业地理要素匹配,提高地址匹配的效率和准确度。另外,位置服务作为一种流行的基于上下文的推荐系统能根据用户当前位置信息推荐兴趣点数据。但是,这种内在特征导致泄露用户的敏感位置信息给不可信的位置服务提供者。我们主要关注了查询预测中的位置隐私问题,提出了一种新的位置隐私保护方案,并用实验评估了此方案的性能表现并展示了它的实用性。最后,我们研究了基于ITSS理论与用户运维管理最佳实践相结合的资源管控产品,它可以对客户整个IT资源的运行的状态、事件、故障、性能等进行全面监控。从业务角度帮助客户有效管理和优化IT环境的运营效率,实现IT基础架构的全方位监控,优化IT服务流程,提升IT服务质量。

关键词：密码学 HBASE存储技术内存泄漏攻击冷启动栅格数据位置服务 ITSS 统一事件处理引擎

来源：评论

学校读者我要写书评

暂无评论

image Classification Optimization of High Resolution Tissue images

Image Classification Optimization of High Resolution Tissue ...

引用

9th International Conference on Large-Scale Scientific Computations (LSSC)

作者： Kozlovszky, M. Hegedus, K. Windisch, G. Kovacs, L. Pinter, G. Obuda Univ John von Neumann Fac Informat Budapest Hungary

ISBN: (纸本)9783662438800;9783662438794

Generic image classification methods are not performing well on tissue images. Such software solutions are producing high number of false negative and positive results, which prevents their clinical usage. We have created the MorphCeck high resolution tissue image processing framework, which enables us to collect morphological and morphometrical parameter values of the examined tissues. Size of such tissue images can easily reach the order of 100 MB-1 GB. Therefore, the image processing speed and effectiveness is an important factor. Our main goal is to accurately evaluate high resolution H-E (hematoxilin-eozin) stained colon tissue sample images, and based on the parameters classify the images into differentiated sets according to the structure and the surface manifestation of the tissues. We have interfaced our MorphCheck tissue image measurement software framework with the WND-CHARM general purpose image classifier and tried to classify high resolution tissue images with this combined software solution. The classification is by default initiated with a large training set and three main classes (healthy, adenoma, carcinoma), however the new image classification process' wall-clock time was intolerably high on single core PC. The processing time is depending on the size/resolution of the image and the size of the training set. Due to the tissue specific image parameters the classification effectiveness was promising. So we have started a development process to decrease the processing time and further increase the accuracy of the classification. We have developed a workflow based parallel version of the MorphCheck and WND-CHARM classifier software. In collaboration with the MTA SZTAKI Application Porting Centre the WND-CHARM has been ported to some distributed computing infrastructure (DCI). The paper introduces the steps that were taken to optimize WND-CHARM applications running faster using DCIs and some performance results of the tissue image classification proce

关键词： Application porting Medical image processing workflow HP-SEE gUSE Scalability MorphCheck WND-CHARM

来源：评论

学校读者我要写书评

暂无评论

Sequential Monte Carlo in Bayesian Assessment of Contaminant Source Localization Based on the Sensors Concentration Measurements 1

引用

10th International Conference on parallel processing and Applied Mathematics (PPAM)

作者： Wawrzynczak, Anna Kopka, Piotr Borysiewicz, Mieczyslaw Natl Ctr Nucl Res Otwock Poland

ISBN: (数字)9783642551956

ISBN: (纸本)9783642551956

Accidental atmospheric releases of hazardous material pose great risks to human health and the environment. In this context it is valuable to develop the emergency action support system, which can quickly identify probable location and characteristics of the release source based on the measurement of dangerous substance concentration by the sensors network. In this context Bayesian approach occurs as a powerful tool being able to combine observed data along with prior knowledge to gain a current understanding of unknown model parameters. We have applied the methodology combining Bayesian inference with Sequential Monte Carlo (SMC) to the problem of the atmospheric contaminant source localization. The algorithm input data are the on-line arriving concentrations of given substance registered by the distributed sensor's network. We have proposed the different version of the Hybrid SMC along with Markov Chain Monte Carlo (MCMC) algorithms and examined its effectiveness to estimate the probabilistic distributions of atmospheric release parameters. The proposed algorithms scan 5-dimensional parameters' space searching for the contaminant source coordinates, release strength and atmospheric transport dispersion coefficients.

关键词： Bayesian inference Stochastic reconstruction MCMC methods SMC methods

来源：评论

学校读者我要写书评

暂无评论

An Efficient Architecture for parallel Skyline Computation over Large distributed Datasets

引用

网际网路技术学刊 2014年第4期15卷 577-588页

作者： He Li Sumin Jang Jaesoo Yoo

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：