Table of Content

02 March 2016, Volume 0 Issue 2

Previous Issue    Next Issue

Virtual Scene Real Time Navigation Based on Spherical Panorama

ZHAO Yi-li1,2, XU Dan1, ZHANG Yan2

2016, 0(2):  1-4,10.  doi:10.3969/j.issn.1006-2475.2016.02.001

Asbtract ( 371 )  

References | Related Articles | Metrics

A virtual scene real time navigation system is designed and implemented based on spherical panorama. Spherical panorama rendering can be achieved by re-projection method. Different perspective views of the virtual scene can be obtained by re-projecting visible portion of the spherical panorama onto the view plane. Camera rotation can be accomplished with the re-projection of a spherical panorama and image rotation, and camera zooming can be accomplished through changing the camera’s field of view. In order to achieve real-time rendering demand, the paper analyses the re-projection process carefully and uses look-up tables and incremental computing to accelerate spherical panorama rendering. The system can render virtual scene based on spherical panorama in real time with high frame per second.

 Detection and Correction of Traffic Signs Shape in Nature Scene

PAN Ming-xing, SUN Han

2016, 0(2):  5-10.  doi:10.3969/j.issn.1006-2475.2016.02.002

Asbtract ( 311 )  

References | Related Articles | Metrics

The automatic recognition of traffic signs can be applied to the automatic or semi-automatic auxiliary driving system which can provide the information of surrounding road conditions to the driver. The traffic signs always suffer distortion and deformation problem because of the complexity of road conditions, the changing illumination conditions and the vibrating of car. In this paper, a fast and efficient algorithm is proposed which can locate the position of traffic signs based on the color and shape information, and the traffic signs can be corrected using perspective transformation. Experiment results show that the algorithm is of a good robustness to illumination, and can overcome the distortion problem caused by the unsuitable camera angle.

 Multi-scale Multiple Kernel SVM Classification for Hyperspectral Imagery Data

CHAO Shuan-she1, CHU Heng1,2, WANG Xing1

2016, 0(2):  11-14,20.  doi: 10.3969/j.issn.1006-2475.2016.02.003

Asbtract ( 197 )  

References | Related Articles | Metrics

Aiming at support vector machine（SVM） using single kernel learning not handling the classification problem of hyperspectral imagery data that the sample distribution is irregular and complex, a hyperspectral imagery data classification method based on sampling strategy and multiple kernel support vector machine is proposed in this paper. Firstly the method does sampling referring to the minority support vectors (SVs) rather than the training data to provide a balanced distribution during multiple kernel support vector machine mode, and then uses the weighted sum approach to multiple kernel learning(MKL) and optimizes parameters by gradient descent algorithm. Finally, a series of two-class classifiers are used to achieve the multi-class classification. Experimental results show that overall classification accuracy increased by 4.07%, average classification accuracy increased by 9.62% compared with the traditional SVM.

A Face Expression Recognition Method Based on Fusion of Supervised #br#  Super-vector Encoding and Adaptive GMM Model

LI Rong, WANG Hua-jun, XU Yan-hua, MENG De-jian

2016, 0(2):  15-20.  doi: 10.3969/j.issn.1006-2475.2016.02.004

Asbtract ( 210 )  

References | Related Articles | Metrics

 Facial expression recognition under different lighting conditions and states is a challenging research. A fusion algorithm based on adaptive Gaussian Mixture Model (GMM) and supervised super-vector encoding is proposed. Firstly, the overlapping image blocks are extracted. Then, local descriptor from each block is extracted by the adaptive GMM so as to map images in low-dimensional space to high-dimensional space. Finally, supervised super-vector encoding is used to do classification training. Experimental results on the Multi-PIE and BU3D-FE multi-view facial expression databases show that the recognition accuracy of proposed algorithm can achieve 91.8% and 95.6% respectively on Multi-PIE and BU3D-FE. It takes only 0.142 seconds in identifying a sample on BU3D-FE. Proposed algorithm has higher recognition accuracy and less recognition time-consuming than several other excellent algorithms.

 Anisotropic Diffusion Image Denoising Based on Grey System Theory

YANG Xing-jiang1,3, LIAO Zhi-wu2, PU Yong-hua3

2016, 0(2):  21-23,27.  doi: 10.3969/j.issn.1006-2475.2016.02.005

Asbtract ( 241 )  

References | Related Articles | Metrics

In order to overcome the shortage of anisotropic diffusion image denoising algorithm for weak boundary and details, the author puts forward a kind of  anisotropic diffusion image denoising method of fusion of grey correlation degree. Through calculating the grey correlation degree of local neighborhood of image pixels, by which determining its neighborhood is edge of the area or the smooth area, the image noise is removed by using different diffusion coefficient. The experimental results show that the algorithm can effectively remove the image noise, preserve image details, and provide an effective way to study the gray correlation theory.

 Multi-target Real-time Tracking Algorithm Based on #br#  Color Edge Statistics and Color Histogram

LIU Yi-chen

2016, 0(2):  24-27.  doi:10.3969/j.issn.1006-2475.2016.02.006

Asbtract ( 396 )  

References | Related Articles | Metrics

Human tracking based on the color histogram is one of the most popular tracking methods. However, it is hard to accurately describe the target signature in bigger number of people scene if it solely depends on the color histogram. Therefore, we propose a multi-target real-time tracking algorithm based on the color edge statistics and the color histogram. Furthermore, this algorithm can deal with the tracking process, match the target by the color edge distribution and the color characteristics, and confirm the scope of target searching. The experimental results show that the algorithm can effectively improve the multi-target tracking accuracy and robustness.

 Scalable Multi-view Video Coding Based on Video Correction

YAN Tao1, ZHOU Xiao-xiong1, DENG An-yuan1, LUO Wen-ting2, ZHANG Pan-dong1

2016, 0(2):  28-31,37.  doi: 10.3969/j.issn.1006-2475.2016.02.007

Asbtract ( 217 )  

References | Related Articles | Metrics

Since scalable multi-view video coding (SMVC) has not been well studied, this paper proposes scalable multi-view video coding based on video correction. Firstly, the histogram matching is mainly used to eliminate color differences among views. Then compensating the DC coefficient is mainly used to eliminate the difference between the views of image brightness. Finally, based on correlation analysis, all views are divided into two groups: base views and enhancement views, which use SMVC. The proposed SMVC scheme is tested with several multi-view sequences to determine its view scalability, high coding efficiency.

Image Filtering System for Natural Reserves Based on Color Features

LIU Zi-jun1, ZHANG Jie-bin1, ZHOU Xin-xin2, GAO Qiang1

2016, 0(2):  32-37.  doi: 10.3969/j.issn.1006-2475.2016.02.008

Asbtract ( 234 )  

References | Related Articles | Metrics

To solve the problem of heavy workload and low efficiency when filtering images of natural reserves, on the basis of analyzing the characteristics of the collected images of natural reserves, an image filtering system based on color features is designed and implemented. Images are segmented, global color features and local color features are extracted by the system, and earth mover’s distance (EMD) is adopted to match pending images with filtered images. After judging the images according to the matching results, the images without animals are removed, thus reducing workload. By filtering the images collected at several collection points in a certain natural reserve, it is verified that the system can effectively reduce about one third of the workload of manual image filtering, and results in higher work efficiency.

A Recommendation Algorithm Based on Denoising Autoencoders

XIE Lin-quan1, LIANG Bo-qun2

2016, 0(2):  38-41.  doi:10.3969/j.issn.1006-2475.2016.02.009

Asbtract ( 433 )  

References | Related Articles | Metrics

 As the autoencoders neural network can extract the features of data from low-level to high-level, and find the potential correlation between the samples, in order to improve the accuracy of the recommendation system, this paper proposes a recommendation algorithm based on denoising autoencoders. Firstly, this paper pretreats score data by ZCA whitening, adds the random noise in the processed data, and builds the autoencoders neural network, then obtains the network weights through the model pre-training, 〖JP2〗fine-tuning. Finally according to the trained network weights the test sample score is reconstructed to forcast user rating  and calculate score error. Experimental results show that denoising autoencoders can effectively improve the recommendation accuracy.

 Transfer Learning Algorithm Based on Tensor Data

XIE Jiang-long, LIU Bo

2016, 0(2):  42-48.  doi: 10.3969/j.issn.1006-2475.2016.02.010

Asbtract ( 363 )  

References | Related Articles | Metrics

 Most of the existing transfer learning algorithms have been extersively studied based on the vector data.We may transfer a higher order tensor data into a vector-based data as input, and utilize the transfer learning algorithm for vector-based data in higher order tensor data analysis. However, this way will often result in the loss of tensor data’s space information and the curse of dimensionality problems, so it is necessary to study the transfer learning algorithm based on tensor data. In view of traditional vector-based transfer learning algorithm, this paper presents a transfer learning algorithm based on tensor data. This method makes us to directly use tensor data as an input. In this way we solve the above problems, but also improve the classification accuracy. Experiments show that our method can obtain higher classification accuracy compared with others.

 A Madaline Network Learning Algorithm Based on Improved MRII

ZHENG Xing, LI Zong-xiang

2016, 0(2):  49-51,57.  doi: 10.3969/j.issn.1006-2475.2016.02.011

Asbtract ( 193 )  

References | Related Articles | Metrics

In this paper, by analyzing the Madaline network work principle,  according to the network characteristic and the shortcomings of the existing algorithms including the adjust weigh formula used in MRII algorithm with many parameters, but most of which come from experience in practice without theoretical reason, it made networks fall into “local recycle” with turning over the neuron based on confidence principle. We present an improved learning algorithm based on MRII, by establishing neuron’s sensitivity as a tool for measuring the turn of each hidden neuron, which reduces the number of weight adjustment. Some experimental results verify the effectiveness of the proposed algorithm.

Differential Private Histogram Publication for Data Stream

XIA Xiao-ling, LIU Hui-yi

2016, 0(2):  52-57.  doi: 10.3969/j.issn.1006-2475.2016.02.012

Asbtract ( 199 )  

References | Related Articles | Metrics

Current research on differential private publication associcted with data stream mainly considers a binary data stream, which cannot efficiently deal with the general data stream’s private publication. An efficient differential private histogram publication algorithm called DDPA was proposed, which is oriented toward non-uniform distributed numerical stream. Basing on the sliding window model, the similarity on two adjacent timestamps of data distribution is applied to allocate the budget privacy dynamically, which makes each window’s total budge not exceed the privacy budget ε, and after that, the grouping and merging strategies are used to calculate the local optimal histogram quickly. According to comparing and analyzing the proposed algorithm with the other similar algorithms on the published data’s availability, the experimental results show that the proposed algorithm is effective and feasible.

Algorithm for Minimum Bounding Rectangle of Fast Extracting Leaves

XIANG Yuan-ping, HE Yan-ping, WEI Yu-lin, LIANG Huan, GUO Ben-chu

2016, 0(2):  58-61.  doi: 10.3969/j.issn.1006-2475.2016.02.013

Asbtract ( 281 )  

References | Related Articles | Metrics

Geometric feature such as rectangle degree, which is a crucial foundation for image classification and recognition. Extracting minimum bounding rectangle is of important value. A new algorithm of minimum bounding rectangle of fast extracting plant leaves was studied in this paper. First, image was scanned from horizontal direction and vertical direction to extract the initial bounding rectangle. Horizontal axis and vertical axis were defined through the center of initial bounding rectangle. Then, with the main axis rotated and moved, four boundaries and a new bounding rectangle were obtained. Compared with the area, the minimum bounding rectangle was extracted. The effectiveness of this algorithm is verified by using Flavia plant leaves database. The results show that minimum bounding rectangle can be quickly found.

Personalized Recommendation Algorithm Based on Tags and Collaborative Filtering

LIU Jian, ZHANG Kun, CHEN Xuan

2016, 0(2):  62-65,71.  doi: 10.3969/j.issn.1006-2475.2016.02.014

Asbtract ( 377 )  

References | Related Articles | Metrics

Existing collaborative filtering algorithm is primarily based on user’s ratings for resources, but usually user ratings matrix data are sparse, a small amount of data can not express very well the characteristics of users and resources. However, the tag can reflect the user’s interest and can describe the features of the resource. Therefore, by introducing tag, a personalized recommendation algorithm based on tags and collaborative filtering is proposed in this paper. The basic idea is to consider the tag as an intermediate link between the user and resource. Then the correlation degree between user and tag, tag and resource is calculated through splitting three-dimensional relationship among the users, tags and resources, and then build the user’s interest model. Finally, according to the user’s interest model to predict preference values for other new resources, and produce the Top-N recommendation set. Compared with the existing algorithm, the precision and the recall rate of the algorithm proposed are all improved, and reach a better recommended effect.

 An Efficient Attribute Based Encryption Data Sharing Scheme of #br#  Supporting Attribute Revocation

WEI Ming-lun, LYU Xin, XU Shu-fang, WANG Long-bao, LI Shui-yan, PING Ping, MA Hong-xu, LIU Xuan

2016, 0(2):  66-71.  doi:10.3969/j.issn.1006-2475.2016.02.015

Asbtract ( 344 )  

References | Related Articles | Metrics

With the popular of cloud computing, more and more sensitive data are stored on cloud and it can easily be shared over the Internet. At the same time, the problems of security come. To counter this problem, this paper proposes an effectual way for data shared based on CP-ABE. This encryption scheme is very applicable to fine-grained access controls of outsourced database, and can be used to solve some safety problems of cloud usage. In this paper, we use the access structure with only AND gates and make attributes as public key. Every attribute can express three different situations, including positive, negative, and irrelevant, in AND gate. In particular, It associates ciphertext with user‘s private key properties, supporting proxy re-encryption. Based on the above, this proposed scheme can also have high flexibility in representing access control policy and reduce the burden of authorized agency. In addition, it implemented instant revoking of user properties and efficient decryption.

Sequential-multi Digital Signature Application in Electronic Contract Signing

LEI Qiong

2016, 0(2):  72-74,81.  doi:10.3969/j.issn.1006-2475.2016.02.016

Asbtract ( 384 )  

References | Related Articles | Metrics

Electronic contract is an important protocol in today’s society for some users, which plays a vital role in the development and interest of business or personal, it becomes a main problem how to design a fair and efficient electronic contract signing protocol. In this paper, based on difficulty of discrete logarithm problem, using a special signature—sequential-multi digital signature， we design a fair and efficient electronic contract signing protocol. By increasing the time stamp, it solves the defect of signature delay, makes sure that signatures will be done in the certain order and time strictly. So it is more efficient. By proving and analyzing, the scheme is correct and feasible.

 Elimination of Memory Disclosure Attacks Using Data Isolation

CHEN Ke-xin, LIU Yu-tao

2016, 0(2):  75-81.  doi: 10.3969/j.issn.1006-2475.2016.02.017

Asbtract ( 309 )  

References | Related Articles | Metrics

As more and more memory-disclosure bugs been discovered, it’s important to protect memory safety. Nowadays mainstream of this field prefers enforcing memory isolation to protect memory safety. However, none of them can achieve a fine-grained protection while easy-to-deploy solution since most solutions requires manually modification to source code, or they enforce memory isolation in a coarse-grained way. We present a novel method, which can traces and collects all possible operations that access the sensitive data in runtime, and can perform automatically code transformation during compilation. Our solution do not require manually modification, and evaluation shows that we can efficiently eliminate memory disclosure attacks, such as Heartbleed, found in OpenSSL. Meanwhile, the overhead of compilation and memory cost are lower than 1%.

 Research on Safety Protection Strategy of Smart Substation Application Layer Data Encryption

YU Hao1, JIA Xue1, WANG Qiang2

2016, 0(2):  82-85.  doi: 10.3969/j.issn.1006-2475.2016.02.018

Asbtract ( 191 )  

References | Related Articles | Metrics

 Currently, protocol IEC61850 is widely used in smart substations. The feature of openness stay hidden defects in the information security. This paper transformed

the field in the application layer protocol which based on IEC61850 mapping to MMS, GOOSE, SV and designed an encryption and authentication mechanism using the Public Key

Infrastructure (PKI) theory in the application layer protocol transmission. This method ensures the confidentiality, integrity, non-repudiation of protocol IEC61850 in

transmission. The experiment proved the feasibility and effectiveness using message after transformation.

 Design of Linux Supporting Platform for Distance Education in Chinese, #br#  Uyghur and Kazakh Based on DVB-S

YANG Qin1,3, WANG Liang-liang2,RUI Xue1

2016, 0(2):  86-90.  doi:10.3969/j.issn.1006-2475.2016.02.019

Asbtract ( 274 )  

References | Related Articles | Metrics

In accordance with the demand of hardware update, system efficiency and ease of use during the engineering, which is based on the early research results in the

rural party members and cadres’s modern distance education engineering in the Xinjiang Uyghur autonomous region, this paper puts forward the solution that improves each

function of distance education by study on Linux operating system for the latest kernel. Firstly, the construction process and functional design of distance  education

supporting platform, which is based on the latest software and hardware architecture, should be researched. Then, intelligent analysis of DVB-S data and bilingual localization

technology in Chinese, Uyghur and Kazakh, is studied. Finally, the hardware compatibility is tested. The research indicates that the performance and compatibility of supporting

platform has been promoted. Moreover, the problem of automatic identification and analysis of satellite resources have been resolved, and the Linux operating system, which is

displayed with two languages in Chinese and Uyghur, another in Chinese and Kazakh, have been realized. This research is significant to promote the independent innovation ability

of information technology and the diversified development of information industry in Xinjiang area.

Bilingual Point Reading Learning System Based on 581 Chinese Characters

ABUDUKELIMU Yusufu1,2, WANG Liang-liang1, QIN Qi-yi1

2016, 0(2):  91-93,97.  doi: 10.3969/j.issn.1006-2475.2016.02.020

Asbtract ( 245 )  

References | Related Articles | Metrics

This paper mainly introduces the research and implementation of the Chinese characters reading learning system based on 581 Chinese characters most commonly

used. Firstly, the paper introduces the research and results of statistical analysis of coverage of 581 Chinese characters in basic Chinese textbooks. Secondly, the learning

resources are construted and the localization technology of the minority languages are studied, namely the technology solutions of Chinese, Uighur, Kazakh languages in the Web

page to show normal, are created. Finally, combined with the previous two basic researches, the paper introduces a cross platform reading learning system based on 581 Chinese

characters learning resource library.

TWDM-PON System Based on PolSK Orthogonal Modulation #br#  Technology to Realize Multicast Communication

LI Yuan-yuan1,2

2016, 0(2):  94-97.  doi:10.3969/j.issn.1006-2475.2016.02.021

Asbtract ( 258 )  

References | Related Articles | Metrics

A new method based on TWDM-PON system is proposed in this paper. Based on its original four communication wavelengths based on ASK modulation, a separate PolSK

modulation based communication channel is added. The channel can be used to transmit multicast data, enabling the ONU to achieve receiving unicast and multicast services in a

single wavelength. The transmission efficiency and performance of TWDM-PON system can be improved by using this kind of method.

Study on Profit Allocation Model of Collaborative Value Chain in Cloud Environment

CHEN Ying, WANG Ping, GE Shi-lun

2016, 0(2):  98-103.  doi:10.3969/j.issn.1006-2475.2016.02.022

Asbtract ( 306 )  

References | Related Articles | Metrics

Because of the lack of research on profit allocation of value chain in cloud environment, a profit allocation model based on gray relational analysis is

proposed. The collaborative value chain in cloud environment is characterized by dynamic building and running process, the customer-focused network, open and transparent

internal operation as well as agile and intelligent collaboration. This paper proposes the collaborative framework of value chain enterprises in cloud environment, and analyzes

the factors, index systems, and profit allocation model which influenced value chain profit allocation. Thus, a profit allocation model is built and then solves through GRA

method. Finally, a case is put forward to verify rationality and validity of the model. The results show that the proposed model is feasible to carry out the profit allocation

fairly and reasonably, and can adapt to the characteristics of collaborative value chain in cloud environment.

 An Improved Robust SLAM Algorithm

KE Jiang-sheng, NI Jian-jun, WU Liu-ying

2016, 0(2):  104-106,111.  doi:10.3969/j.issn.1006-2475.2016.02.023

Asbtract ( 391 )  

References | Related Articles | Metrics

When the robot works in an unknown environment, it is often affected by the external disturbance, which will make the localization failure based on the general

SLAM algorithm. To deal with this problem, an improved robust SLAM algorithm is proposed, to obtain a more accurate positioning result and reduce the effect of the external

disturbance, by adjusting the state prior estimate error co-variance and measurement noise co-variance of the system simultaneously. Finally, some simulation experiments are

conducted, the results show that the proposed algorithm can decrease the robot localization error more effectively under the environment with external disturbance, and the

performance of the proposed algorithm is superior to other algorithms.

 Research on Wireless Remote Control and Location of Locomotive in Coal Mine

XU Hai-fei, TIAN Hong-xian

2016, 0(2):  107-111.  doi: 10.3969/j.issn.1006-2475.2016.02.024

Asbtract ( 249 )  

References | Related Articles | Metrics

Many coal mines have entered the stage of deep mining, a scheme of remote control and wireless location of locomotive is proposed to avoid miners entering the

dangerous area, which can achieve the goal of remote mining. The target of wireless control of locomotive in the visual range can be achieved by wireless fidelity(WiFi)

technology and Mesh base station can increase the distance between the controller and the locomotive. Workers can get the location and the the surroundings of the locomotive in

the monitoring center by localization algorithm and video cameras.

Role-collaborative Resolution Strategies for Resource Conflicts Model in#br#  Teaching Video Reservation System

CHEN Qian-guo1, LIU Kui2, ZU Shen-cong1

2016, 0(2):  112-116,126.  doi: 10.3969/j.issn.1006-2475.2016.02.025

Asbtract ( 223 )  

References | Related Articles | Metrics

In order to improve the traditional manual teaching video reservation and administration model in university, according to the workflow and role-collaboration of

teaching video work process, using the shared resource conflict resolution strategy, a teaching video reservation system was researched and designed based on resolution

strategies for resource conflict and role-collaboration. The experiment results show that the system of the model implementation can effectively improve the traditional

appointment model bringing inconvenience, improve work efficiency, carry out the convenience of teachers and staff of the video, and enhance greatly the information level of

college teachers’ teaching.

 A Log Storage Method of Multiple Water Conservancy Systems Based on WebService

FENG Jun, ZHANG Li-xia, TANG Zhi-xian

2016, 0(2):  117-121,126.  doi:10.3969/j.issn.1006-2475.2016.02.026

Asbtract ( 303 )  

References | Related Articles | Metrics

Because of the lack of hydraulic log storage method and centralized storage system, multiple hydraulic systems’ logs are stored distributed, which is hard to

mine data from the logs. In order to store the massive hydraulic logs efficiently and to manage the hydraulic logs intensively, we propose a new log storage method, which is

based on WebService and on the strength of this method, we design and complish a hydraulic log storage system. HBase is used as a database in the system, which makes full use of

its mass heterogeneous data storage features; WebService is adopted because of its one-to-many service pattern in calling and its remote interoperability. WebService transforms

log storage work from hydraulic systems to server, which reduces the pressure of the hydraulic systems. Finally, we make a verfication of the system’s concurrent execution and

compare WebService with log4j about log storage efficiency.

 Predicting Student Grades Ranking Based on Improved TrAdaboost Algorithm

XIE Xing-yu1, ZHANG Ying-lu2

2016, 0(2):  122-126.  doi:10.3969/j.issn.1006-2475.2016.02.027

Asbtract ( 396 )  

References | Related Articles | Metrics

This paper provides an improved TrAdaboost algorithm, and uses the algorithm to  predict the rank of students who come from different departments. This algorithm

can overcome the influence of different data distribution bringing, the experiments confirm its flexibility and accuracy. It makes some significance to bring convenience to the

management of the university.