Loading...

Table of Content

    29 March 2017, Volume 0 Issue 3
    Research and Implementation of Remote Control System Based on Mesh Network
    ZHAO Kai, TIAN Hongxian
    2017, 0(3):  1.  doi:10.3969/j.issn.1006-2475.2017.03.001
    Asbtract ( 207 )  
    References | Related Articles | Metrics
    A remote control system based on the Mesh network was accomplished. In the control center where tens of kilometers from the device, people can still control the device correctly with smoothly playing monitoring image and videos. From the principle of every single system, it was finally accomplished by designing the Mesh network, the communication protocol, the hardware and the software. After that, the whole system was analyzed in detail and its firsthand data is collected after the communication test. The packet loss rate, delay, jitter were calculated to guarantee the QoS of the system. The remote control system is installed to the model of scraper, but it can be used to any device needed to be controlled far away.
    Implementation of User Identity and Location Awareness in Attendance#br# Management System Based on WeChat
    MA Shangyin1, GAO Guanxin1, LIU Jiaji1, TIAN Wei2
    2017, 0(3):  8.  doi:10.3969/j.issn.1006-2475.2017.03.002
    Asbtract ( 257 )  
    References | Related Articles | Metrics
    The paper introduces a way for user identity and location awareness based on a WeChat embedded browser. WeChats OpenID and Cookies Token are used to confirm user identity, and AP positioning information and QR code are used to locate user in which solved the lack of cell phone location information in indoor location. It has the advantage of accurate identification, easy to use, crossplatform compatibility, less expensive than other techniques such as biometric technology, and easy to deploy.
    An Essay Context Recognition Model Based on Syntax Decision Tree and SVM Algorithm
    WANG Zheng1, LIU Shipei2, PENG Yanbing1
    2017, 0(3):  13.  doi:10.3969/j.issn.1006-2475.2017.03.003
    Asbtract ( 204 )  
    References | Related Articles | Metrics
    With the increasing maturity of the networked social life, many fields such as research and commerce have encountered the problems of processing Chinese texts. Parsing
    the texts from all aspects is necessary for the increasingly deepening research of text classification. On the one hand, semantic parsing is essential to text mining. On the
    other hand, context recognition can be widely applied in numerous text mining problems, such as sentiment analysis, public feeling analysis and so on. In this paper, a context
    classification model is proposed based on syntactic decision trees, Ngram feature extraction and SVM classifiers to recognize the different meanings of the same words under
    different short contexts. The results show that the proposed model, which can batch process data, has favorable generalization ability, indicating that the model can solve the
    problems of context recognition efficiently.
    Android WebRTC Technology Application in Collaborative Design
    WANG Huaiting, LIANG Zhenghe, ZHUANG Tianlong
    2017, 0(3):  18.  doi:10.3969/j.issn.1006-2475.2017.03.004
    Asbtract ( 203 )  
    References | Related Articles | Metrics
    Recently, WebRTC is a popular technology which makes point to point video transmission and data transmission realizable in Web applications. When combining with mobile
    technologies and applied to collaborative design, designers in different areas could communicate through network in real time. The key point of WebRTC is to establish a peer
    channel. However, when poor network or multiple communicating exists, the course may fail. Therefore, a delayed resend mode was introduced to solve the first problem.
    Specifically, after supplicant sending SDP descriptors and waiting for a period of time, if the recipient could receive the corresponding message, the channel is been connected,
    else the supplicant should resend the message. If this process failed five times, the trail would be terminated. Moreover, most endtoend file transfer relies on the
    server, which means that high performance servers should be supplied as transferring large files. As for this problem, the solution is to split the file into pieces and to
    transfer massage among terminals through WebRTC peer channel directly, which makes it no longer depends on the server, and takes full advantage of the upstream broadband of the
    client. It shows that delay and resend mode can not only guarantee a successful connection of channels, but also be able to transfer files efficiently and quickly between
    peers.
    A Realtime Traffic Information Collection and Display System#br# Based on Mobile Terminal
    XU Penghui, GUO Ling, SHI Pan
    2017, 0(3):  22.  doi:10.3969/j.issn.1006-2475.2017.03.005
    Asbtract ( 247 )  
    References | Related Articles | Metrics
    In order to solve the problems of high cost and low accuracy of traffic information collection, a realtime traffic information collection and display system based on
    mobile terminal is designed and implemented. In this system, traffic flow is firstly converted from the moving speed and direction of the mobile terminal, and then is used to
    grade traffic congestion. The server is designed by Servlet technology, and MySQL is used as database in server. The client is designed as Web App using jQuery Mobile+Html5
    technology to adapt to different mobile devices. Baidu Map is used to display the traffic flow information in this system, and AJAX technology is adopted to update the
    statistics. Testing results show that the system has advantages of lowcost and highaccuracy in getting realtime traffic information based on popular use of 4G network and
    mobile intelligent terminal, which is helpful to relieve traffic congestion pressure and facilitate people to travel.
    Research on Customer Scoring Application Based on MLP Neural Network Model
    WANG Bing1, HAN Junyu2
    2017, 0(3):  27.  doi:10.3969/j.issn.1006-2475.2017.03.006
    Asbtract ( 204 )  
    References | Related Articles | Metrics
    The judgment of the possibility that customers purchase products is the focus of the enterprise marketers attention. The paper addresses the crossselling for the
    insurance product customers and other financial product customers, thereby highly accurately quantifies their potential purchasing power. According to the summary of the applied
    theory of the personal insurance business, this paper presents a customer purchasing behavior scoring model. Through the use of the simplified insurance customer data of China
    Post Financiers, this paper carries out an empirical analysis of the model effectiveness. The results of the study show that the model, with a good forecasting performance,
    provides more efficient predictions and specific evaluation criteria, thus being able to help enterprises timely identify the most effective marketing tool for customer
    acquisition and retention, ultimately improving marketing success to a greater extent.
    Design and Implementation of Protection for Ethernet Linear One by One Based on PTN
    CHENG Xiangkun1, HE Yan2
    2017, 0(3):  32.  doi:10.3969/j.issn.1006-2475.2017.03.007
    Asbtract ( 253 )  
    References | Related Articles | Metrics
    Ethernet linear protection is a significant characteristic of the PTN network to meet the needs of the operator level. In order to guarantee the transmission of
    service in the PTN network equipment with reliable transmission channel and selfhealing ability to recover the faulty link, this paper discussed a linear 1∶1 protection
    switching mechanism based on G.8031 protocol, realized the fast automatic protection switching function of the main and standby link within 50 ms of the telecommunication level,
    and improved the reliability of PTN network.
    Study on Grade Categorization Model of Question Based on Online Test Data
    FAN Yuling, XU Tao, WANG Qin
    2017, 0(3):  36.  doi:10.3969/j.issn.1006-2475.2017.03.008
    Asbtract ( 142 )  
    References | Related Articles | Metrics
    To tackle with the blindness of random questions choosing for exercise and test of online learning system, this paper clusters questions by exploiting various
    feature subsets and parameters based on online judge data and Kmeans. For the test data of ACM online judge system, the features of temporal fluctuations, mean of time
    consumption and repeat submission rate are used to build a classification model which will be optimized based on experimental analysis of number of features and clusters. The
    experimental results show that the proposed method is simple but effective, the classification results of the model are consistent with the empirical results.
    Measurement Method and Evaluation Model of Voice Quality on Electric IMS Network
    WANG Pan1, LIU Shidong2
    2017, 0(3):  40.  doi: 10.3969/j.issn.1006-2475.2017.03.009
    Asbtract ( 113 )  
    References | Related Articles | Metrics
    An electric IMS voice quality measuring method based on passive probe message analysis is proposed, which obtains the basic performance indicators such as time delay
    of voice session through parsing RTP, RTCP, SIP packets. A voice quality evaluation model is put forward to help the network operation and maintenance department to monitor and
    evaluate the quality of a large number of concurrent voice conversations on the backbone network. The developed prototype system verifies the feasibility of the scheme.
    Research on Hot News Topic Detection of Incremental Clustering
    WANG Liying1,2, GE Lina1,2, ZHANG Yipeng 1,2, WANG Hong1,2
    2017, 0(3):  46.  doi:10.3969/j.issn.1006-2475.2017.03.010
    Asbtract ( 218 )  
    References | Related Articles | Metrics
    News text clustering is applied to achieve ranking and finding hot news topics. To obtain hot topic news ranking by the formula of heat, the reporting rate is studied.
    To overcome the shortcomings of huge cost of calculation and inaccuracy of selecting eigenvalues when SinglePass is used to cluster, its proposed enhanced methods to solve
    the problems. The improved algorithm is verified and analyzed through the experiment using fetched news corpus. The result of experiments is compared to the SinglePass, and a
    more desirable hot news clustering effect is presented, the feasibility of modified algorithm is demonstrated.
    Factorial of Large Number and Super High Precision Solution of Natural Logarithm
    MA Xu, WANG Dayong
    2017, 0(3):  51.  doi:10.3969/j.issn.1006-2475.2017.03.011
    Asbtract ( 161 )  
    References | Related Articles | Metrics
    In view of the program design requirements of the large data factorial and the super high precision solution of the natural logarithm in engineering practice, the
    paper proposes the practical algorithm and the program flow diagram description of super high precision multiplication and division operation, and the time complexity degree
    analysis of correlational algorithm is given also. In addition, we issue the complete C language programs of the factorial of large data and the super high precision solution of
    the natural logarithm in which provides the necessary experimental data and algorithm to effectively solve the problem of super high precision programming in engineering
    practice.
    Search Rank Model Based on Random Forests and LambdaMART
    LEI Wu1,2, LIAO Wenjian2, PENG Yanbing2
    2017, 0(3):  54.  doi: 10.3969/j.issn.1006-2475.2017.03.012
    Asbtract ( 158 )  
    References | Related Articles | Metrics
    Recent studies have shown that Boosting provides excellent predictive performance across a wide variety of tasks. In learningtorank, boosted models such as
    RankBoost and LambdaMART have been shown to be among the best performing learning methods based on evaluations on public data sets. In this paper, we investigate Random Forests
    (RF) and LambdaMART. Then we combine the two algorithms by first learning a ranking function with RF and using it as initialization for LambdaMART to create a new rank model.
    We report our results on the public learningtorank data sets using two metrics ERR and NDCG. The new rank model performs better than two original algorithms models.
    Research on Differential Privaty for Decision Tree Release Technology
    CHEN Yang, YU Shoujian
    2017, 0(3):  59.  doi:10.3969/j.issn.1006-2475.2017.03.013
    Asbtract ( 218 )  
    References | Related Articles | Metrics
    Privacy disclosure issue is becoming more and more serious due to big data. We proposed a differential private generalization data publishing algorithm for decision
    tree. The algorithm is based on the noninteractive model, in the process of attribute segmentation by combining similar branch to reduce the overall noise, and keeps more of
    original information, so it can improve the accuracy of classification. According to the need of exponential mechanism, adaptive allocation privacy budget, compared with the
    previous algorithms, under the condition of the same privacy budget it can make the data more differentiated, and decision tree classification accuracy is higher. The
    experimental results also prove the validity and superiority of this algorithm.
    Research on Security Situation Awareness and Intrusion Intention#br# Recognition Based on Threat Intelligence Sharing
    LI Weijian, JIN Qianqian, GUO Liang
    2017, 0(3):  65.  doi:10.3969/j.issn.1006-2475.2017.03.014
    Asbtract ( 176 )  
    References | Related Articles | Metrics
    With large scale introduction of “Internet +” technology represented by cloud computing, Internet of things, and big data, information security exposure surface of
    power grid is rapidly increasing. In order to cope with the severe challenges of security protection of power grid, one of the important means is using the threat intelligence
    as the starting point, through sharing important thirdparty security intelligence data on the security situation assessment of power grid to timely detect abnormal behavior,
    and finally identify intrusion intention. This paper presents a method for sharing threat intelligence, realizes the design of security threat intelligence sharing system, and
    puts forward the security situation awareness and intrusion intention identification method based on threat intelligence sharing.
    Security Access System of Substation Based on IEC61850 Standard
    LI Muye1, JIN Qianqian1, SONG Shuting2
    2017, 0(3):  71.  doi:10.3969/j.issn.1006-2475.2017.03.015
    Asbtract ( 153 )  
    References | Related Articles | Metrics
    At present, network and computer equipment in most of the domestic substation system are lack of function for security access surveillance. There are not normative
    control means for operating and debugging computers accessed to every layer of network and U disk devices on each computer. In order to ensure the safe running of substation
    network and service system, a substation security access system is developed based on the analysis of smart substation layered structure according to IEC61850, which uploads
    realtime monitoring information of network equipment and U disks insertionextraction to master station client in control layer though the MMS protocol. After a spot test
    in China Southern Power Grid Company, the system runs stable as expected, and which ensures the safety of substation system from the source.
    New Infrared Image Nonlinear Enhancement Algorithm #br# Based on Nonsubsampled Contourlet Transform
    GUO Min, JIANG Aimin, CAO Mei
    2017, 0(3):  76.  doi:10.3969/j.issn.1006-2475.2017.03.016
    Asbtract ( 213 )  
    References | Related Articles | Metrics
    Considering the problems that the low contrast and large noise are two main characteristics of infrared image, the infrared image enhancement method is presented in
    this paper, which is based on normal distribution characteristics and NSCT transform. First of all, images high frequency coefficients and lowfrequency coefficients are got
    by the actual acquisition of nighttime infrared image nonsubsampled contourlet transform, for high frequency coefficients, the adaptive threshold method is used to determine the
    high frequency threshold. By judging the edge information, the weak edge and strong edge are made different processing to enhance the edge of the image information. The low
    frequency coefficients are made nonlinear correction by using the improved Gamma transform to improve the contrast of the image. Finally, the enhanced image is obtained by
    inverse transform. The experimental results show that the proposed algorithm can effectively suppress the noise of infrared image, improve the image contrast information, and
    the visual effect is good.
    Performance Evaluation Software of Seekers Image#br# Processing Algorithm Based on OpenCV
    LIU Qingsong, XIE Xiaofang, CAO Jian, ZHANG Longjie
    2017, 0(3):  80.  doi:10.3969/j.issn.1006-2475.2017.03.017
    Asbtract ( 167 )  
    References | Related Articles | Metrics
    Aiming at the evaluation of the image processing algorithm of the seeker, this paper designs an image processing algorithm evaluation software platform. The algorithm
    test software is developed by using the plugin method, which separates the tested method from the evaluation software. The design of the index class is carried out by using
    the plant model which makes the index design scalable. The common algorithm interface and data structure are designed, which can evaluate various kinds of preprocessing,
    detection and tracking algorithms. The visual simulation module is used to generate test images under various conditions, which can cover all kinds of test conditions. By using
    OpenCVs diverse function library and standard image format, the efficiency of software development is improved. Finally, the feasibility and generality of the system are
    verified by the simulation experiment.
    Hand Location and Hand Gesture Recognition in Fisheye Image
    ZHANG Yan1, TIAN Xiang2
    2017, 0(3):  86.  doi:10.3969/j.issn.1006-2475.2017.03.018
    Asbtract ( 220 )  
    References | Related Articles | Metrics
    In view of the demand for hand gesture interaction in Augmented Reality (AR), this paper proposes a method of hand location and hand gesture recognition in fisheye
    image. First, skin color region is detected in YCbCr color space, then HSV space color histogram feature and shape feature is used to detect mark in the skin region, and to
    divide hand area. Second, this method uses the centroid of mark as hand positioning center, according to imaging principle of fisheye lens, to calculate the 3D position of the
    point relative to fisheye lens. Last, this paper extracts hand gesture shape features, to complete the procedure of recognizing hand gestures. Through the experiment of the 85
    fisheye images with different background and light, the correct rate for recognizing gestures is 96%, recognizing single fisheye image takes about 0.03 s. The experimental
    results show that the proposed method can realize hand location and hand gesture recognition quickly and effectively in complex background.
    Face Recognition Method by Weight Fusion of Original and Mirror Samples
    CHEN Xiang, WANG Ling
    2017, 0(3):  91.  doi:10.3969/j.issn.1006-2475.2017.03.019
    Asbtract ( 152 )  
    References | Related Articles | Metrics
     Focused on the small sample size problem, a method for face recognition based on local binary pattern features of weight fusing the original and mirror image was
    proposed. Firstly the proposed scheme generated the mirror images of the original training samples through the geometric transformation. Then each sample was divided into
    several blocks, and the concatenated LBP histogram calculated over each block was used as the face feature of the sample. Next, the original and their mirror images face
    feature was merged into new samples by weighted fusion scheme. Finally, the test sample was classified into the correct subject by using the histogram intersection method.
    Experimental results showed that the algorithm has achieved a higher recognition rate, with the fact the recognition rate on FERET face database increased by 25.17%, 15.8% and
    25.17%, respectively, compared with the contrast algorithm when the number of training sample per subject was 1,2 and 3 respectively. 
    Design and Implementation of Big Emergency Medical Supplies Scheduling System
    ZHOU Dongyang
    2017, 0(3):  96.  doi:10.3969/j.issn.1006-2475.2017.03.020
    Asbtract ( 212 )  
    References | Related Articles | Metrics
    In view of that the existing emergency medical security system is not perfect, the medical emergency material dispatching ability of hospital departments is insufficient, the paper puts forward construction of large medical supplies for emergency dispatch system based on grey correlation degree. In the system design, according to the needs of practical application demand and expansion by material management control layer, emergency communication access layer, terminal equipment application layer of three layers of structure design, the paper has established a medical emergency material command scheduling system platform which sharing of rich access module and all kinds of emergency communication resource. Design and build of medical emergency material demand prediction model and scheduling model, can provide accurate medical emergency supplies scheduling program for the system, and through the simulation experiments, the rationalization of the generated medical emergency supplies scheduling scheme is validated, so enhance the ability of relevant government departments to treat emergencies.
    Research on Organization Mode of Satellite Quality#br# Evaluation Module Based on Multidimensional Data Model
    SHEN Xiajiong1, CHEN Wei1,2, HUANG Xiangzhi2, ZANG Wenqian2
    2017, 0(3):  101.  doi:10.3969/j.issn.1006-2475.2017.03.021
    Asbtract ( 163 )  
    References | Related Articles | Metrics
    Research on organization mode of satellite quality evaluation module is the key of mining and using satellite data resource to help users to find the remote sensing information effectively. In order to solve the problem of single organizational structure of satellite quality evaluation module and retrieval efficiency, we designed a multidimensional data model of satellite quality evaluation module information based on “star” pattern. The model can be used for quickly displaying dynamic module information of satellite quality evaluation from multiple perspectives and improving the retrieval speed. The model was applied to design the organizational structure of ZGZHLPJ system, and the system organizational structure became more orderly and stereo, by which verified the practicability of the model.
    An Acetone Gas Monitoring System Based on ZigBee Wireless Sensor Network
    WANG Liuqing, NI Wei, WANG Wei, ZHONG Qun
    2017, 0(3):  106.  doi:10.3969/j.issn.1006-2475.2017.03.022
    Asbtract ( 167 )  
    References | Related Articles | Metrics
    For the trouble brought by acetone solvent leakage, and for the problems existing in industrial gas monitoring system, such as wiring difficulties, poor accuracy, low reliability and high cost, this paper puts forward an acetone gas monitoring system based on ZigBee wireless sensor networks, which can effectively monitor the acetone solvent leakage, and take measures timely. The system can enhance the security of industrial production.
    Design of Unified Commercial Database Management
    LIU Qun1,2, SONG Qianyi2, CHANG Yong1,2, YANG Jianxu1,2
    2017, 0(3):  109.  doi:10.3969/j.issn.1006-2475.2017.03.023
    Asbtract ( 190 )  
    References | Related Articles | Metrics
     In computer software development, when the commercial database is used by multiple related products, due to the different database driven, different configuration and different operating grammar of the databases, it  may be different from the database selection to the application layer interface. To solve this problem, a unified commercial database is designed and developed. Different implementation is given for different platform when a uniform access interface is provided. By implementing the management of the business table, not involving the specific business, the use for across various systems is realized. And in order to avoid collapse to affect the access efficiency when a large number of databases are connected, the database connection pool is used to implement for connecting and accessing the database connection repository, and it makes the users can focus on business.
    Data Integration Architecture Design for Serviceoriented #br# Ordnance Equipment Support Service
    ZHOU Gang, GUO Fuliang
    2017, 0(3):  112.  doi:10.3969/j.issn.1006-2475.2017.03.024
    Asbtract ( 214 )  
    References | Related Articles | Metrics
    Ordnance equipment safeguard business data integration is of great significance to solve the problem of security information, improve the security of business capabilities based on information system. Based on the basic concept of data integration and serviceoriented, a serviceoriented system integration method is put forward. Through norming general packet specification of security business and establishing a method of service business logic, the application integration of requirements is completed, and a multisource heterogeneous data integration of serviceoriented architecture is designed.
    Emotion Analysis of Hotel Customers Reviews Based on SVM
    SHI Qiangqiang1, ZHAO Yingding2,3, YANG Hongyun2,3
    2017, 0(3):  117.  doi:10.3969/j.issn.1006-2475.2017.03.025
    Asbtract ( 183 )  
    References | Related Articles | Metrics
    This paper improves the accuracy of word segmentation and emotion analysis of network vocabulary and expressions by increasing the variety of emotion dictionary. On the other hand, customer reviews of a hotel are used as the original data. After extracting the amount of text features, such as positive and negative words, negative words, the degree of adverbs and the amount of special symbols, we make different feature combinations, and hope to find the optimal combination of parameters SVM including C and g through the kfold Cross Validation and grid search algorithm. Training and testing different feature combinations by SVM and analyzing the correct rate of each combination, we find out the most suitable combination of text feature and feature analysis which are used for study of user reviews of emotion. The results show that under the premise of satisfying the optimal combination of parameters C and g, the correct rate of the feature combination using positive and negative emotional words, negative words, emotion score and degree adverbs is the highest and reaches 93.4%.
    Research on Notes Recognition Based on Weight Distribution
    BIAN Yuwei, GUO Ling
    2017, 0(3):  122.  doi:10.3969/j.issn.1006-2475.2017.03.026
    Asbtract ( 153 )  
    References | Related Articles | Metrics
    Pitch is one of the most important parameters of music signal in the pitchbased music recognition. In this paper, a new pitch detection algorithm based on weight distribution is proposed by taking piano music as an example, in which autocorrelationbased pitch detection is carried out first, then each sample frame of the music signal is analyzed, and finally an improved weight distribution method is used to detect the fundamental frequency of the music. Experiments show its effectiveness in improving the recognition accuracy.