Table of Content

24 May 2016, Volume 0 Issue 5

Previous Issue    Next Issue

Remote Sensing Image Classification Based on Fusion of #br# Multiple Features with Block Feature Point Density Analysis

JIANG Yaping， LI Shijin

2016, 0(5):  1-6.  doi:doi: 10.3969/j.issn.1006-2475.2016.05.001

Asbtract ( 199 )  

References | Related Articles | Metrics

With the development of remote sensing and the related techniques, the resolution of these images is largely improved. Compared with moderate or low resolution images, highresolution images can provide more detailed ground information. However, a variety of terrain has complex spatial distribution. The different objectives of highresolution images have a variety of features. The effectiveness of these features is not the same, and some of them are complementary. Considering the above characteristics, a new method is proposed to classify remote sensing images based on the hierarchical fusion of multifeature. Firstly, these images are preclassified into two categories in terms of whether feature points are uniform or nonuniform distributed. Then, the color histogram and Gabor texture feature are extracted from the uniform distributed categories, and the ScSPM（linear spatial pyramid matching using sparse coding）feature is obtained from the nonuniform distributed categories. Finally, the classification is performed by the two different support vector machine classifiers. The experimental results on a large remote sensing image database with 2100 images show that the overall classification accuracy is boosted by 10% in comparison with the highest accuracy of single feature. Compared with other methods of multiple features fusion, the proposed method has achieved the highest classification accuracy which has reached 90.1%, and the time complexity of the algorithm is also greatly reduced.



Research on Visionbased Monitoring System of Vehicles for Violation of Traffic Line

LIU Changde， XU Guili， CHEN Xi， CHENG Yuehua

2016, 0(5):  7-12.  doi:10.3969/j.issn.1006-2475.2016.05.002

Asbtract ( 178 )  

References | Related Articles | Metrics

At present, the effective range of the monitoring system for violation of traffic line is small, and the detection accuracy of violation is still to be improved. In order to expand the scope of monitoring and improve accuracy, this paper proposes a new method of vehicles for violation of traffic line based on high speed ball type camera. First of all, according to the actual demand, we design the hardware system and software system, and by highspeed ball machine mobility, a method is proposed based on static detection and dynamic capture, to expand the scope of supervision. In addition, aiming at the problem that the accuracy rate is not high, the two step rolling line detection algorithm is proposed, That is, according to the prospect of the smallest external rectangular box to determine the rolling line, and then further through the region near the rolling line statistical prospects to accurately judge, thereby improving the accuracy of the rolling line detection and antiinterference ability. Experimental results show that, compared with the traditional rolling line detection system, the system expands monitoring range 1 times, and the detection accuracy is improved by 21.1%, realizes the effective supervision of the vehicle in a large range.



An Image Denoising Algorithm of RiemannLiouville Fractional Integral

GOU Rong

2016, 0(5):  13-17.  doi:10.3969/j.issn.1006-2475.2016.05.003

Asbtract ( 233 )  

References | Related Articles | Metrics

Traditional image denoising algorithms usually lose the image edge and texture information. This can make the image blurred and bring difficulty for the subsequent image processing. To avoid these shortcomings, a mask operator which based on the RiemannLiouville definition of the fractional integral was proposed and experimented with the test image. Meanwhile, the definition of the PSNR and the GLCM was introduced to analyze the experiment result. The experiment results show that, different from the traditional image denoising algorithms, the image denoising algorithm of the fractional integral can denoise image and keep the image edge and texture information effectively.

A Defogging Method Based on Improved Dark Original Prior Image

QI Na， HUI Hongmei， MU Chang

2016, 0(5):  18-21.  doi:10.3969/j.issn.1006-2475.2016.05.004

Asbtract ( 199 )  

References | Related Articles | Metrics

In order to reduce the influence of fog weather on the outdoor imaging system, this paper proposes an improved dark original prior image defogging method. First, the analysis of the imaging mechanism of haze weather images is made. Then we make a simple introduction and objective analysis on the traditional dark original prior defogging principle and algorithm, and make some optimization and improvement. Finally, simulation experiments are took for fog images using various algorithms. The results show that, the method can significantly improve the definition and contrast ratio of the original image.


TopicEye: An Interactive Approach of Exploring Topic Information from #br# Multiple Sources Based on Visual Analysis

ZHAO Fangyu

2016, 0(5):  22-32.  doi:10.3969/j.issn.1006-2475.2016.05.005

Asbtract ( 250 )  

References | Related Articles | Metrics

This paper presents a visual analysis approach to develop a full picture of relevant topics discussed in multiple textual sources. The key idea behind our approach is to jointly cluster the topics extracted from each source in order to interactively and effectively analyze common and distinctive topics. Firstly, This approach extracts topics from textual corpora by using a correlated topic model method. Different sources of textual corpora are matched together with some common topics. Next， this paper develops a visualization tool consisting of three visualization views for better understanding and analyzing matched topics， as well as the relatively independent ones. The major feature of these three visualizations is that they clearly present what each topic is and the relations among them from multiple perspectives， which allows users to analyze deeper relationships and features， from multiple textual corpora and over different periods of time. To meet different users’ needs and help better understanding the structure of different textual corpora， these visualizations provide interactive analysis methods at multiple scales. To verify the usability of this method， this paper has applied it in a variety of data sets including IEEE Transactions on Knowledge and Data Engineering (TKDE)， ACM Transactions on Graphics (TOG) and IEEE Transactions on Visualization and Computer Graphics (TVCG). Qualitative evaluation and several realworld case studies with domain experts demonstrate the promise of our approach， especially in support of analyzing a visualization based full picture of topics at different levels of detail.


Research on Guarantee Method of Integrity and Availability of Synthetic Vision System

ZHOU Yanqing， WANG Lisong

2016, 0(5):  33-38.  doi: 10.3969/j.issn.1006-2475.2016.05.006

Asbtract ( 157 )  

References | Related Articles | Metrics

The integrity of SVS system is an important part of SVS, which can increase the credibility of SVS, and the availability of SVS is also an important property of SVS. SVS also ensures that the integrity and availability of the system is a key system and not just a system for use. The realtime monitoring of SVS can effectively guarantee the integrity of the system, but it will reduce the availability of the system. Therefore, this paper is to discuss the availability and integrity of SVS in two kinds of properties, and then puts forward the guarantee method. Then, the paper discusses and puts forward the alarm algorithm based on the architecture level. The method presented in this paper can effectively ensure the integrity and availability of SVS, but it has some requirements on its hardware and architecture, and it has the redundancy of data resources. 

Efficient Compressed Sensing Magnetic Resonance Imaging #br# Based on Compound Regularization

SHEN Huijun

2016, 0(5):  39-45.  doi:10.3969/j.issn.1006-2475.2016.05.007

Asbtract ( 225 )  

References | Related Articles | Metrics

Compressed sensing(CS) is well utilized to accelerate magnetic resonance imaging(MRI) without degrading images quality. The present compound regularization imaging

models seldom select different regularization tools according to the different structural features of images. The imaging methods thus require high sampling rates to obtain

images with diagnostic value. In this paper, a novel compound regularization CS MRI model integrating two regularization tools: uniform discrete curvelet transform and total

variation, is introduced. It includes spatial image and lowpass subband coefficients total variation regularization, highpass subbands coefficients l1

regularization and kspace data fidelity constraint. Then this CS MRI model formulation is solved via variable splitting and alternating direction method of multipliers.

Simulated results on in vivo data are evaluated by objective indices and visual perception, which indicates that the proposed method outperforms the existing regularization

models established on the total variation and wavelets under the same sampling rate.

Clustering Algorithm Research on Mixed Received Signal Strength Sequences #br# of Multiple Nodes on Wireless Sensor Network

CHEN Shu, LU Ying

2016, 0(5):  46-50.  doi:10.3969/j.issn.1006-2475.2016.05.008

Asbtract ( 174 )  

References | Related Articles | Metrics

It confirms number of unknown nodes without identification by dealing with mixed received signal strength sequences achieved by mobile anchor node, the clustering

results of the mixed received signal strength directly determines the positioning accuracy of unknown nodes on the wireless sensor network. EM algorithm is sensitive to the

initial values. To overcome the drawback above, this paper proposes a three stage iterative processing method, adding front and rear channel to EM algorithm. Kmeans algorithm

is used as front processing channel to improve the selection of initial values on EM algorithm. Bayesian information criterion is employed as the rear processing channel to

improve the accuracy of the clustering. Finally, the simulation result shows that the proposed algorithm can successfully estimate the number of unknown nodes, and achieve

precise received signal strength.

Research on Mechanism of Multiplesource Multicast of GDOI

WU Tao， ZHI Yang， BAI Lizhen

2016, 0(5):  51-54.  doi:10.3969/j.issn.1006-2475.2016.05.009

Asbtract ( 198 )  

References | Related Articles | Metrics

Through analyzing the mechanism of GDOI, some improved advices are proposed aiming to the limitation that it can only be applied to singlesource multicast when GDOI

is used to IPSec. Then, the work procedure of the improved GDOI protocol is described and a part of improved scheme is simulated in host. The tests show that the improved GDOI

protocol can support the multiplesource multicast based on IPSec more perfectly and be feasible.

Study on Indices of DCN Structural Robustness

ZHANG Chan， FENG Guojun， XIAO Yunbo

2016, 0(5):  55-60.  doi:10.3969/j.issn.1006-2475.2016.05.010

Asbtract ( 251 )  

References | Related Articles | Metrics

In recent years, with the rapid development of cloud computing and dataintensive super computing(DISC), data center network(DCN) is playing an increasingly important

roles as the underlying infrastructure. The DCN needs to be robust to failures and uncertainties to deliver the required qualityofservice (QoS) level and satisfy service

level agreement (SLA). In the paper we study the classical robustness metrics and perform a comparative analysis on considering various failure scenarios, present the inadequacy

of the classical network robustness metrics to appropriately evaluate the DCN robustness, and propose some new metrics to quantify the DCN robustness. For the detailed study is

unavailable for centering the DCN robustness, our study will lay a firm foundation for the future DCN robustness research.

Research on Network Group Emergency Early Warning Method Based on Cloud Model

SUN Lingfang1,2， LIN Weijian2

2016, 0(5):  61-66.  doi:10.3969/j.issn.1006-2475.2016.05.011

Asbtract ( 179 )  

References | Related Articles | Metrics

With the arrival of information age, the interaction of network public opinion and sudden mass incidents has formed new challenge to network public opinion analysis

and early warning. First a network public opinion crisis early warning index system is built, including the risk of public opinion, public opinion diffusion degree and public

opinion favorite degree. On this basis, the mathematical model of the cloud model is applied to the network public opinion crisis warning, based on cloud model, a network

community emergency early warning method is put forward, and finally the goal of the network public opinion crisis warning of real time, intelligence and availability, is

realized.

Research on Key Technologies of Blind Classification#br# in MIMOSTBC Communication Systems

LING Qing, ZHANG Limin, YAN Wenjun, DENG Xiangyang

2016, 0(5):  67-72.  doi:10.3969/j.issn.1006-2475.2016.05.012

Asbtract ( 170 )  

References | Related Articles | Metrics

To improve the reliability and validity of multiinput multioutput (MIMO) communication system, SpaceTime Block Codes (STBC) has been widely used. This behavior

has brought new challenges for MIMO communication. The blind classification of signals in MIMOSTBC communication systems, including blind estimation of the number of the

transmit antenna, blind classification of the modulation type and blind classification of the STBCs, is the attractive research topic of signal processing in the last decade or

more. Firstly, the research status of MIMOSTBC communication systems is introduced. Secondly, the main idea and method of MIMOSTBC communication is proposed. Finally, the

development of MIMOSTBC is concluded. Simulations show that the proposed methods perform well and are suitable for practical application.

Grid Management Through Intelligent Agents Cooperation

ZHU Haoyue

2016, 0(5):  73-76.  doi:10.3969/j.issn.1006-2475.2016.05.013

Asbtract ( 164 )  

References | Related Articles | Metrics

This paper introduces some details towards the use of collaborating intelligent agents in the environments of grid management, and it shows how rules can be used at

different levels of the hierarchy to facilitate the cooperation among intelligent agents. It also shows how the communication messages can be inferred automatically from rules

and get generated. The paper introduces and analyzes three schemes according to the different cooperation levels in detail, and gives the time and percentage in a “violated”

state for the three schemes, the results show that there is definitely an advantage when IAs cooperate together.

Social Tagging Recommendation Based on Content Analysis and Tag Expansion

YU Chongwei, WU Gang

2016, 0(5):  77-83.  doi:10.3969/j.issn.1006-2475.2016.05.014

Asbtract ( 186 )  

References | Related Articles | Metrics

With the rapid development of social tagging system such as Douban and CiteULike, the social tagging recommendation technology has become a hot research topic

recently. However, compared to traditional recommendation system, social tagging system has a new evaluation criterion and tag, which leads to the problems of low tag quality,

data sparsity, cold start and unified model applicability and so on. Hence, the traditional recommendation method can’t achieve satisfactory results. This paper proposed TECA

(Tag Expansion and Content Analysis), a method based on tag expansion and content analysis, and used TECA to implement tag and user recommendation. TECA can use classification

to refine the resource model, use semantic topic to recommend tags and use tag expansion to alleviate data sparsity. The experiment on CiteULike’s dataset shows TECA achieves

better tag and user recommendation performance than traditional collaborative filtering recommendation method.

A Management System of Import and Export Food Safety Information

GUO Ping1,3， LIU Haiyan1， AI Shirong2

2016, 0(5):  84-89.  doi:10.3969/j.issn.1006-2475.2016.05.015

Asbtract ( 293 )  

References | Related Articles | Metrics

The establishment of scientific food safety information platform for China’s import and export food safety alerts work is very important. The design and

implementation of a system about existing data association analysis was introduced. Based on JavaEE platform, the system which was developed according the MVC model had the

crossplatform characteristics and good expansibility. It realized the function of inquiring and analyzing the massive data including alerts, standard methods, hazard limit,

laboratory information and laws. The actual using shows the system can greatly improve working efficiency.

An Optimal Algorithm for Bus Trip Supporting Walking Transfer

YANG Yongping， CHEN Hongshun， TANG Jian

2016, 0(5):  90-94.  doi:10.3969/j.issn.1006-2475.2016.05.016

Asbtract ( 189 )  

References | Related Articles | Metrics

Less transfer and quick arrival is the goal of bus travel. In this paper, we use the existing bus station and bus lines, and consider the characteristics of public transportation network stability, establish the basic transfer table on base of time consuming and transfer times on which supports realtime walking transfer query. A pair of source and target sites can provide K transfer schemes with multiple attributes, including estimated time, pass nodes list, travel distance, cost, walking distance, etc. Practice shows that the results of the algorithm meet the user expectations.

Design and Application of Marine Power Generation Device #br# Test Data Monitoring and Management System

MA Changli1, HU Naijun2, GUO Zhizhuo3

2016, 0(5):  95-99.  doi:10.3969/j.issn.1006-2475.2016.05.017

Asbtract ( 190 )  

References | Related Articles | Metrics

With continuous decrease of global fossil energy, coastal countries attach importance to the ocean energy which is a kind of green renewable energy. Power generation is one of the main purposes of ocean energy development. To build the platform which is used to test marine power generation devices is very important. It can provide efficient ocean energy development with data reference and evaluation method. This paper designs a data integration and management system by using the technology of data interface standardization and Web Service. It can monitor the device test data and platform status realtimely and integrate with other subsystem. The longterm complete data support from this data integration and management system is helpful for the development and evaluation of the marine power generation devices.


Research on Dynamic Slicing Technique of JavaScript Programs

YE Jiabin, YU Haibo

2016, 0(5):  100-105.  doi:10.3969/j.issn.10062475.2016.05.018

Asbtract ( 278 )  

References | Related Articles | Metrics

Program slicing technique can effectively facilitate program debugging, but traditional slicing technique is difficult to be directly utilized on JavaScript programs because of the dynamic feature introduced by JavaScript. In this paper, we propose a new dynamic program slicing technique. Combining with the features of JavaScript, this paper extends the definition of system dependence graph for JavaScript programs and includes the design and implementation of constructing algorithm. Based on this system dependence graph, we can perform program slicing work. The experimental result shows this technique can be effectively used in slicing JavaScript program and the slicing result occupies a small proportion of the original program.

Research on Relational Database Data Block Replication Model

YUE Junsong， LIU Sai， NIE Qingjie， ZHANG Lei， HU Nan， XU Xuefei

2016, 0(5):  106-110.  doi:10.3969/j.issn.1006-2475.2016.05.019

Asbtract ( 213 )  

References | Related Articles | Metrics

 In view of the problems existing in the current logic level database replication technology, this paper presents a model of database data replication based on physical level. The model uses a database replication group to establish its reference relationship. By analyzing the Oracle log data in the file system format, the raw device and the redo log based on the database, the data fragments are synchronized to the end of the data segment. By end of log fragment reorganization, the log head sign is modified a synthetic log file and source side validation are registered to disaster libraries, and synthetic log is automatically or manually written to disaster libraries, so as to realize the physical database replication. By comparing and analyzing the test data, the model can not only ensure the data consistency, but also reduce the time delay of data replication.

A New Generation Intelligent Cloud Computing Data Center of #br# China’s Civil Aviation

WANG Li1, LIU Xiang2, LIN Enai2, DENG Gang2

2016, 0(5):  111-115.  doi:10.3969/j.issn.1006-2475.2016.05.020

Asbtract ( 194 )  

References | Related Articles | Metrics

The goal of construction of the China’s civil aviation new generation intelligent cloud computing data center is to build a cloud platform for new generation passenger service system of China’s civil aviation, which provides data visualization, resource supply automation, and operation standardization. This paper analyzes the background of this cloud computing platform, the development of cloud computing home and abroad, the business architecture and technical architecture of the new generation intelligent data center which is based on cloud computing, and, finally, the significance of the achievement of this platform.

Intrusion Detection System Based on Integration of #br# Various Detection Technologies

ZHENG Shengjun1, XIA Yechao2，3, LI Jianhua2，3, WANG Li1, WANG Hongkai4

2016, 0(5):  116-121.  doi:10.3969/j.issn.1006-2475.2016.05.021

Asbtract ( 169 )  

References | Related Articles | Metrics

With the continuous development of network technology, inhouse network security is also increasingly subject to various known and unknown malware threats. In order

to detect these malicious programs, we designed and implemented an intrusion detection system based on integration of the depth detection technology, anomaly detection

technology, misuse detection technology. This intrusion detection system uses depth protocol analysis, behavioral analysis, feature matching, intelligent protocol

identification, protocol anomaly attack detection, traffic anomaly detection and so on. And this intrusion detection system realizes malicious programs detection through data

acquisition module, data reorganization module, data analysis module, console module and features for system management module. The system overcomes the shortcomings of

traditional single detection schemes, and can detect new generation of threats such as 0day attacks, polymorphic attacks, distortion attacks effectively.

Efficient Secure Data Aggregation Protocol on Wireless Sensor Networks

LIANG Chuan， TENG Cui

2016, 0(5):  122-126.  doi:10.3969/j.issn.1006-2475.2016.05.022

Asbtract ( 186 )  

References | Related Articles | Metrics

In some special applications, such as battle battlefield surveillance, confidential data will be disclosed to the adversaries if data security is not guaranteed. For

wireless sensor networks, this paper proposes an efficient secure data aggregation protocol ECIPAP based on present research works. This encryption algorithm can avoid network

delay and high energy consumption caused by complicated computation. ECIPAP can reduce the energy consumption while preserve data confidentiality and data integrity through

security analysis and theoretical performance analysis. After that, we implement this protocol by program on physical sensor nodes. We analyze the situations of the sensor nodes

deployed in the monitoring area and the temperature data sensed by them. The results show that ECIPAP has ideal security, efficiency and availability.