Skip to main content

Über dieses Buch

This volume presents the proceedings of the First Euro-China Conference on Intelligent Data Analysis and Applications (ECC 2014), which was hosted by Shenzhen Graduate School of Harbin Institute of Technology and was held in Shenzhen City on June 13-15, 2014. ECC 2014 was technically co-sponsored by Shenzhen Municipal People’s Government, IEEE Signal Processing Society, Machine Intelligence Research Labs, VSB-Technical University of Ostrava (Czech Republic), National Kaohsiung University of Applied Sciences (Taiwan), and Secure E-commerce Transactions (Shenzhen) Engineering Laboratory of Shenzhen Institute of Standards and Technology.



Innovative Computing Technology for Applications


High Availability and High Scalability to in-Cloud Enterprise Resource Planning System

In this paper the host system architecture with high availability and high scalability has introduced to an in-cloud Enterprise Resources Planning (in-cloud ERP) deployed in the virtual environment to tackle the crucial problem of unexpected down-time or the failure of system failover that causes data loss and system terminated. Access control authentication has been adopted in the cloud to prevent the service-oriented hosts form external fraud or intrusion. As a result, the experiments have shown that the number of access for in-cloud ERP is 5.2 times as many as in-house ERP. The total cost of in-cloud ERP has decreased significantly to 48.4% of total cost of in-house ERP. In terms of operational speed, the approach proposed in this paper outperforms significantly two well-known benchmark ERP systems, in-house ECC 6.0 and in-cloud ByDesign.

Bao Rong Chang, Hsiu-Fen Tsai, Ju-Chun Cheng, Yun-Che Tsai

Prostate Tumor Identification in Ultrasound Images

There are various medical imaging instruments used for diagnosing prostatic diseases. Ultrasound imaging is the most widely used tool in clinical diagnosis. Urologist outlines the prostate and diagnoses lesions based on his/her experiences. This diagnostic process is subjective and heuristic. Active contour model (ACM) has been successfully applied to outline the prostate contour. However, application of ACM in outlining the contour needs to give the initial contour points manually. In this paper, an automatic prostate tumor identification system is proposed. The sequential floating forward selection (SFFS) is applied to select significant features. A support vector machine (SVM) with radial basis kernel function is used for prostate tumor identification. Experimental results showed that the proposed method achieved higher accuracy than those of other methods.

Chuan-Yu Chang, Meng-Yu Tu, Yuh-Shyan Tsai

QoS of Triple Play Services in LTE Networks

This paper deals with studying the effects of performance LTE network throughput for data traffic. Utilisation rate of networks has a significant impact on the quality parameters of triple play services. LTE network was simulated by the software module additionally implemented in a development environment MATLAB. Throughput model has been obtained using this simulation that was used to test the QoS parameters for voice/video. Voice and video data streams using different codecs are transmitted for the obtained throughput. The measured qualitatively QoS parameters determine the resulting quality services from the perspective of end user perception.

Lukas Sevcik, Karel Tomala, Jaroslav Frnda, Miroslav Voznak

New Genetic Algorithm for the p-Median Problem



-median problem is a well-known combinatorial optimization problem with several possible formulations and many practical applications in areas such as operational research and planning. It has been also used as a testbed for heuristic and metaheuristic optimization algorithms. This work proposes a new genetic algorithm for the


-median problem and evaluates it in a series of computational experiments.

Pavel Krömer, Jan Platoš

Hybrid Bat Algorithm with Artificial Bee Colony

In this paper, a hybrid between Bat algorithm (BA) and Artificial Bee Colony (ABC) with a communication strategy is proposed for solving numerical optimization problems. The several worst individual of Bats in BA will be replaced with the better artificial agents in ABC algorithm after running every


iterations, and on the contrary, the poorer agents of ABC will be replacing with the better individual of BA. The proposed communication strategy provides the information flow for the bats to communicate in Bat algorithm with the agents in ABC algorithm. Four benchmark functions are used to test the behavior of convergence, the accuracy, and the speed of the proposed method. The results show that the proposed increases the convergence and accuracy more than original BA is up to 78% and original ABC is at 11% on finding the near best solution improvement.

Trong-The Nguyen, Jeng-Shyang Pan, Thi-Kien Dao, Mu-Yi Kuo, Mong-Fong Horng

Compact Bat Algorithm

Addressing to the computational requirements of the hardware devices with limited resources such as memory size or low price is critical issues. This paper, a novel algorithm, namely compact Bat Algorithm (cBA), for solving the numerical optimization problems is proposed based on the framework of the original Bat algorithm (oBA). A probabilistic representation random of the Bat’s behavior is inspired to employ for this proposed algorithm, in which the replaced population with the probability vector updated based on single competition. These lead to the entire algorithm functioning applying a modest memory usage. The simulations compare both algorithms in terms of solution quality, speed and saving memory. The results show that cBA can solve the optimization despite a modest memory usage as good performance as oBA displays with its complex population-based algorithm. It is used the same as what is needed for storing space with six solutions.

Thi-Kien Dao, Jeng-Shyang Pan, Trong-The Nguyen, Shu-Chuan Chu, Chin-Shiuh Shieh

The New Procurement System Based on MRP Algorithm

To a good number of medium and small enterprises, it is a serious problem how to decline production cost and improve the response to the market by normative management and sharing the information. As an advanced management method and an IT tool, ERP can solve all these problems well. And how to combine ERP with other advanced management methods (such as SCM) and make ERP fit to the mid-small enterprises becomes a hot question recently. Purchase stands on the upstream of SC and takes a strategic effect on cost and quality control. Advanced and efficient purchase management not only can supply continuous material to keep enterprise run perfectly but also can make enterprise more competitive in controlling the whole SC. MRP algorithm designed in this paper can be a good application of procurement management in the ERP system, and achieved good economic.

Lei Meng, Chuansheng Zhou

Context Awareness Services and Intelligent Computing Applications


Wireless Sensor and Mobile Application of an Agriculture Security System

This study uses wireless sensor and mobile APP technologies to plan and build an agriculture security system to prevent farmers’ losses. This security system consists of a variety of different sensor functions. The data collected will be analyzed and judged through the Arduino microprocessor and the information is sent to the control end by the ZigBee wireless transmission module. The control end will activate all buzzers and the searchlight if an invasion has been confirmed. Also, the installation of the CCD camera will activate when suspicious activity occurs and will thus take clear and definite photos as evidence and send an email to notify the user. Using the mobile APP program developed by ourselves, when the smart phone APP detects new photos being stored in the cloud database, it sends an alert message to the user to remind him/her to check the new photo.

Chun-Chieh Fan, Rong-Hou Wu, Liang-Lin Jau, Yu-Ming Li

Knowledge Integration for Diabetes Drugs Ontology

The rising and developing of information technologies has made information overflow. Even in the same topics, a lot of different aspects of knowledge were setup. So, the knowledge integration is one of the important research topics. In this paper, the diabetes drug prescriptionsare usedas examples to do the knowledge integration which according to American Diabetes Association (ADA), American Journal of Clinical Endocrinology Society (AACE), the Republic of China Diabetes Association and the British National Health Service Bureau (NHS). The system will integrate the four medication diabetes associations to establish knowledge ontologies. The system includes three parts. First, the ontologies pre-processing will calculate the similarity between the ontologies and then find out the correlation between ontologies, Next, the system transfers the ontologies format into Joseki, and finally the user through a graphical user interface to obtain information.

Rung-Ching Chen, Yu-Wen Lo, Bo-Ying Liao, Cho-Tscan Bau

Increasing Customer Loyalty in Internet Marketing

In recent years, with rapid development of social networking websites, more and more travelers make their travel and accommodation decisions by referring to online comments (electronic word of mouth). It’s especially true for the customers who live in bed and breakfast. But, due to the limited marketing budget, a bed and breakfast (B&B) enterprise needs an effective and cheap to promote their products and services through internet marketing. Social media marketing could one of cheap and powerful internet advertising channels. However, most of bed and breakfast enterprises lack sufficient human resource and time to interact to the online users of social networking websites. Moreover, there are lots of social media marketing techniques, but we don’t know which one is crucial for a bed and breakfast enterprise. Therefore, this study aims to define the key factors of social media marketing, and then use decision tree to identify the important factors for increasing customers’ loyalty. A survey of social media marketing in Facebook will be provided to demonstrate the effectiveness of our utilized methods.

Long-Sheng Chen, Tzung-Yu Kevin Yang

A Novel Approach for Sustainable Supplier Selection Using Differential Evolution: A Case on Pulp and Paper Industry

Diverse sustainable supplier selection (SSS) methodologies have been suggested by the practitioners in earlier, to find a solution to the SSS problem. A SSS problem fundamentally is a multi-criteria practice. It is a judgment of tactical significance to enterprises. The nature of this decision usually is difficult and unstructured. Optimization practices might be useful tools for these types of decision-making difficulties. During last few years, Differential Evolution has arisen as a dominating tool used for solving a variety of problems arising in numerous fields. In the current study, we present an approach to find a solution to the SSS problem using Differential Evolution in pulp and paper industry. Hence this paper presents a novel approach is to practice Differential Evolution to select the efficient sustainable suppliers providing the maximum fulfillment for the sustainable criteria determined. Finally, an illustrative example on pulp and paper industry validates the application of the present approach.

Sunil Kumar Jauhar, Millie Pant, Ajith Abraham

A New Clustering Algorithm Based on Probability

Clustering is a hot topic of data mining. After studying the existing classical algorithm of clustering, this paper proposes a new clustering algorithm based on probability, and makes a new definition for clustering and outlier. According to the distribution characteristics of sample data, this algorithm determines the initial clustering center automatically. It also implements eliminating outliers in the process of clustering. The experiment results on IRIS show that this algorithm can clustering effectively.

Zhang Yue, Zhou Chuansheng

The Comparison between IABC with EGARCH in Foreign Exchange Rate Forecasting

Foreign exchange rate forecasting catches many researchers interests in recent years. Problems of the foreign exchange rate forecasting model selection and the improvement on forecasting accuracy are not easy to be solved. In this paper, the forecasting results obtained by conventional time-series models and by the Inter-active Artificial Bee Colony (IABC), which is a young artificial intelligent meth-od, are compared with each other with 4 years historical data. The sliding win-dow strategy is used in the experiment for both the training and the testing phases. In our experiments, we use continuous previous three days data as the training set, and use the training result to forecast the foreign exchange rate on the fourth day. In addition, we evaluate the forecasting accuracy with three criteria, namely, Mean Square Error (MSE), Mean Absolute Error (MAE), and Root Mean Square Error (RMSE). The experimental results indicate that feeding macroeco-nomic factors to IABC as the input data is capable to produce higher accurate data in the foreign exchange rate than the conventional time-series models such as EGARCH.

Jui-Fang Chang, Pei-Wei Tsai, Jung-Fang Chen, Chun-Tsung Hsiao

Equivalence Proof of Traditional and Random Grid-Based (2, 2) Visual Secret Sharing

Visual secret sharing (VSS) has attracted considerable attention to scientists and engineers as another branch alongside conventional cryptography to protect the sensitive visual information from several rapacious behaviors. In the literature, there are a number of several techniques used to protect the visual information, among which traditional VSS and random grid (RG)-based VSS are the primary branches. In this letter, we show, by examples, the two means are equal. In addition, the color representation of traditional VSS and RG-based VSS found it different from digital applications like images. Based on the given examples, it is demonstrated that the color representation of the two means can be the same and confirm with digital processing applications.

Shen Wang, Xuehu Yan, Jianzhi Sang, Xiamu Niu

Smart Living Technology


Assistive Listening System Using a Human-Like Auditory Processing Algorithm

Enhancing the quality of hearing perception in noisy environments plays a significant role to improve life quality of elderly persons and hearing impaired people. Accordingly, this study presents a human-like auditory processing (HAP) algorithm to enhance the speech signal in low signal-to-noise ratio (SNR) and non-stationary noise environments. The proposed algorithm comprises two modules, namely Cochlear Wavelet Transform (CWT) and AM-FM Demodulation (AFD); mimicking the human peripheral auditory processing system and the human cortical auditory processing system, respectively. The performance of the proposed HAP algorithm is evaluated in accordance with the ITU Perceptual Evaluation of Speech Quality (PESQ) standard. The results show that the proposed algorithm improves the speech quality by 16.9 % on average. In the other words, the algorithm has significant potential for assistive listening in noisy environments.

Po-Hsun Sung, Jhing-Fa Wang, Hsien-Shun Kuo

An Adaptive Harmony Search Algorithm with Zipf Distribution

Harmony search (HS) can be applied to various optimization problems and easy to implement. In this paper, we try to improve HS by change the reference probability distribution of harmony memory. Zipf distribution is used to balance the intensification and diversification. In addition, we propose the adaptive mechanism to avoid setting the new parameter. Experimental results show that the improvement is effective on the high dimensional numerical function optimization problem.

Shih-Pang Tseng, Jaw-Shyang Wu

LED Lighting Applications for Digital Life

LED lighting based on solid state IC technology works seamlessly with computer and/or computer network. The integration of LED lighting with microprocessor (MCU or PSoC) at system levels proliferate many new applica-tions toward a digital life style. Meanwhile, due to its long life span, compact-ness and efficient energy consumption LED lightings are environmental friend-ly. In this article we introduce four new applications and show how LED light-ing may advance in future “green” digital life.

Lih Wen Hwang

The Exclusive Challenge in Pervasive Learning Technology – Points from the Tiger

Universal computing is a global goal worthy of the most serious consideration of engineering, e-commerce, and educational leaders. For the benefits of the information age to effect general humanity, it is essential that computing become more widely embraced across the diverse groups of society, to confront the increasing problem of the digital divide. Currently, disenfranchisement, disinterest, and overt contempt for technology still abounds (Macionis, 2001). Innovators in engineering and ecommerce need to devote some of their creative energies to systematically making the “universal computing society” a practical reality rather than a myth. The markets for e-commerce, let alone the benefits of telecommunication for humanity, require continuing energy to educate the majority of society about tools and visions for change.

Tzong-Song Wang, Yun-Chung Lin

Human Fetus Health Classification on Cardiotocographic Data Using Random Forests

Pregnancy and fetus development is an extremely complex biological process that, while generally successful and without complications, can go wrong. One of the methods to determine if the fetus is developing according to expectations is cardiotocography. This diagnostic technique’s purpose is to measure the heartbeat of the fetus and uterine contractions of its mother, usually during the third trimester of pregnancy when the fetus’ heart is fully functional. Outputs of a cardiotocogram are usually interpreted as belonging to one of three states: physiological, suspicious and pathological. Automatic classification of these states based on cardiotocographic data is the goal of this paper. In this research, the Random Forest method is show to perform very well, capable of classifying the data with 94.69% accuracy. A comparison with the Classification and Regression Tree and Self-organizing Map methods is also provided.

Tomáš Peterek, Petr Gajdoš, Pavel Dohnálek, Jana Krohová

Signal Recognition and Image Processing


Evolutionary Weighted Ensemble for EEG Signal Recognition

Recognition of an EEG signal is a very complex but very important problem. In this paper we focus on a simplified classification problem which consists of detection finger movement based on an analysis of seven EEG sensors. The signals gathered by each sensor are subsequently classified by the respective classification algorithm, which is based on data compression and so called LZ-Complexity. To improve overall accuracy of the system, the Evolutionary Weighted Ensemble (EWE) system is proposed. The parameters of the EWE are set in a learning procedure which uses an evolutionary algorithm tailored for that purpose. To take full advantage of information returned by sensor classifiers, setting negative weights are permitted, which significantly raises overall accuracy. Evaluation of EWE and its comparison against selected traditional ensemble algorithm is carried out using empirical data consisting of almost 5 hundred samples. The results show that the EWE algorithm exploits the knowledge represented by the sensor classifiers very effectively, and greatly improves classification accuracy.

Konrad Jackowski, Jan Platos, Michal Prilepok

Hierarchical Ensemble of Global and Local Classifier for Texture Classification

In this paper, we propose a novel hierarchical ensemble classifier for texture classification by combining global Fourier features and local Gabor features. Specifically, in our method, global features are extracted from images firstly by 2D Discrete Fourier Transform. Then, real and imaginary components of low frequency band are concatenated to form a single feature set for further processing. Gabor wavelet transform is exploited for local feature extraction. Firstly, Gabor wavelets are used to extract local features from the whole image. Then, these features are spatially partitioned into a number of feature sets, each corresponding to a local patch of the image. After the above processes, an image can be represented by one Global Fourier Feature Set (GFFS) and multiple Local Gabor Feature Sets (LGFSes). These feature sets contain different discriminative information: GGS contains global discriminative information and each LGFS contains different local discriminative information. In order to make full use of all these diverse discriminative information, we propose multiple component classifiers by applying Fisher Discriminant Analysis (FDA) on GFFS and each LGFS, respectively. At last, we combine them into one ensemble by weighted sum rule.

Ming Chen

Pedestrian Detection Using HOG Dimension Reducing in Video Surveillance

Pedestrian detection draws a mount of attention in these years. However, most of the classify-based pedestrian detection methods are facing huge training samples and high computation complexity. In this paper, it proposed a manifold learning based pedestrian detection method. First, modeling the video surveillance scene via mixed gaussian background model and collecting negative samples from the background images; Second, extract the positive and negative samples histogram of oriented gradients(HOG) features, using the local preserving projection(LPP) for dimensionality reduction; Finally, detecting the pedestrian from the input image under the framework of AdaBoost. Experiments show that the algorithm achieved good results both in speed and accuracy of pedestrian detection.

Kebin Huang, Feng Wang, Xiaoshuang Xu, Yun Cheng

Reversible Watermarking Based on Position Determination and Three-Pixel Block Difference

Reversible watermarking based on position determination and three-pixel block difference is proposed in this paper. In the proposed method, for a three-pixel block, no modification is allowed to its center pixel (CP). This unchanged pixel along with all the neighbors surrounding this block constitute a set used for evaluating the intra-block correlation. The incorporation of CP in this set helps to largely enhance the estimation accuracy. According to the strength of correlation, we determine this block into a smooth or complex region. When the desired embedding rate is low, we only modify those blocks located in smooth regions while keeping the others unchanged. Therefore, the PSNR (peak signal to noise ratio) value is largely increased. Experimental result also demonstrate that the proposed method is effective.

Shaowei Weng, Jeng-Shyang Pan, Tien-Szu Pan

A Novel Encryption Algorithm for Quantum Images Based on Quantum Wavelet Transform and Diffusion

In this paper, a novel quantum encryption scheme for quantum images based on quantum wavelet transform (QWT) and double diffusions is proposed. Firstly, diffusion operation applied on the input quantum image, and then QWT worked on the new quantum image to transform this image to the frequency domain. and following the diffusion operation is implemented on the QWT transformed quantum image. finally ,inverse QWT are used.The encryption keys are generated by a sensitive chaotic logistic map, which guarantee the security of the scheme. at the same time,we designed the corresponding quantum circuits to demonstrates that the reasonable of the proposed scheme.

Shen Wang, Xianhua Song, Xiamu Niu

Interleaving and Sparse Random Coded Aperture for Lens-Free Visible Imaging

Coded aperture has been applied to short wavelength imaging (e.g., gamma-ray), and it suffers from diffraction and interference for taking longer wavelength images. This paper investigates an interleaving and sparse random (ISR) coded aperture to reduce the impact of diffraction and interference for visible imaging. The interleaving technique treats coded aperture as a combination of many small replicas to reduce the diffraction effects and to increase the angular resolution. The sparse random coded aperture reduces the interference effects by increasing the separations between adjacent open elements. These techniques facilitate the analysis of the imaging model based only on geometric optics. Compressed sensing is applied to recover the coded image by coded aperture, and a physical prototype is developed to examine the proposed techniques.

Zhenglin Wang, Ivan Lee

Computational Intelligence Approaches for Digital Media Analysis and Description

This paper provides an overview of recent research efforts for digital media analysis and description. It focuses on the specific problem of human centered video analysis for activity and identity recognition in unconstrained environments. For this problem, some of the state-of-the-art approaches for video representation and classification are described. The presented approaches are generic and can be easily adapted for the description and analysis of other semantic concepts, especially those that involve human presence in digital media content.

Alexandros Iosifidis, Anastasios Tefas, Ioannis Pitas

Computational Systems and Its Applications


Register Allocation Based on Boolean Satisfiability

Graph Coloring is an effective method which is used to solve the register allocation problem, it is also an NP-complete problem, heuristic algorithms and various evolutionary algorithms have been proposed in order to improve the performance of register allocation, in this paper, we propose to solve this problem by converting the graph coloring problem into Boolean Satisfiability problem (SAT), the experiments show that our algorithm can use fewer number of registers, which can improve the execution efficiency of the generated codes.

Yang Mengmeng, Liu Jie

Robust Medical Image Watermarking Scheme with Rotation Correction

In order to protect the image contents, many reversible medical image watermarking schemes have been proposed, although reversibility guaranteed the lossless of the cover image, but it also has some shortcomings such as it is vulnerable to geometrical attacks. So a novel robust watermarking scheme for medical image based on the combination Redundancy Discrete Wavelet Transform (RDWT) and Singular Value Decomposition (SVD) is proposed in this letter. Different from the reversibility, which guarantees the perceptional lossless, the proposed scheme achieves satisfied visual quality by exploiting the visually masking property of RDWT, in the meantime, Speeded-Up Robust Features (SURF) and Random sample consensus (RANSAC) based rotation correction scheme is put forward, which can be used to restore the attacked image to the original state. The experimental results show that the proposed scheme has the large amounts of embedding capacity; it is also robust against rotation attacks; and the perceptional quality of watermarked image meets the need of usage in medical images


Lin Gao, Tiegang Gao, Guorui Sheng, Shun Zhang

Design of Data Encryption Transmission System Based on FPGA

In this paper, the hardware system of data encryption and security transmission is studied. A kind of encryption nuclear is designed using chaotic sequences and stream cipher. In addition, a scheme of data encryption system based on FPGA is proposed, thus a new type of encryption system is obtained. Meanwhile, the simulation of the key generation circuit is accomplished. The experiment shows that this hardware encryption system can achieve the encryption and decryption. It ensures the secure transmission of data further.

Yan Yu, Bingbing Song, Xiaozhen Liu, Qun Ding, Ziheng Yang

An Efficient Image Encryption Scheme Based on ZUC Stream Cipher and Chaotic Logistic Map

Digital color image encryption is different from text encryption because of some inherent features of image such as huge data capacity and high correlation among the neighboring pixels. Because of the desirable cryptographic properties of the chaotic maps such as sensitivity to initial conditions and random-like behave, more and more researches use these properties for encryption. This paper proposed an efficient image encryption scheme. Logistic chaos-based stream cipher is utilized to permute the color image. The MD5 hash function and the ZUC stream cipher algorithm are combined to diffusion the color image. Theoretical and experimental analyses both confirm the security and the validity of the proposed algorithm.

Hai Cheng, Chunguang Huang, Qun Ding, Shu-Chuan Chu

The Complexity Analysis of Chaotic Systems

The complexity of the sequence is an important index of quantify the performance of chaotic sequence. In order to select a higher complexity of chaotic sequence and apply it in hardware encryption system, this paper analyzes chaotic complexity quantitative analysis methods and presents the approximate entropy and permutation entropy as criterion of measuring the complexity of the chaotic sequences. Set tent, logistic and henon three kinds of chaotic systems as examples, and we analysis and comparison their complexity. It is proved that the two kinds algorithms are effective, and can distinguish different complex chaos and chaotic sequences. Researches show that the complexity of the Logistic map is greater than that of other chaotic systems. The results of the study provide the theoretical and experimental basis for the application of chaotic sequence in hardware encryption system and the information security communication.

Wei Xu, Bingbing Song, Chunlei Fan, Qun Ding, Shu-Chuan Chu

Implementation of Audio Data Packet Encryption Synchronization Circuit

The paper chooses the cyclone the second generation of the FPGA chip as the gate array (FPGA) used by AES encryption algorithm. The paper also uses speech codec WM8731 chip to realize eight voice and data coding. Then the paper refers to the PCM frame structure TS0 times lot function design code word synchronization. At last, the paper also uses AES128 grouping encryption algorithm to encrypt digital signal, and voice encryption is realized on the FPGA hardware.

Hongbin Ma, Yingli Wang, Gaoling Li

Discriminative Image Representation for Classification

The Bag-of-visual Words (BoW) image representation is a classical method applied for various problems in the fields of multimedia and computer vision. During the process of BoW image representation, one of the core problems is to generate discriminative and descriptive visual words. In this paper, in order to represent the image completely, we propose a visual word filtering algorithm, which filters the lower discriminative and descriptive visual words. Based on the traditional method of generating visual words, the filtering algorithm includes two steps: 1) calculate the probability distribution of the various visual words, and then, delete the words with gentle probability distribution; 2) delete the visual words with less instances. In this way, the generated visual features become more discriminative and descriptive, furthermore, multiple cues fusion, such as shape, color, texture, is also taken into account, we compare our approach with traditional Bag-of-visual Words method applied for image classification on three benchmark datasets, and the performances of the classification all get improvements to some extent.

Zhize Wu, Shouhong Wan, Lihua Yue, Ruoxin Sang

Database and Visualization Technologies


Design of VIP Customer Database and Implementation of Data Management Platform

Accurate and personalized analysis management for VIP Customer data are an important means for enterprises to make business decisions and win benefits. In this paper, take telecom industry as an example, SQL Server 2008 is applied to design the database according to the problem of traditional data analysis and management methods of VIP customers. Combined with Browser/Server structure, an intelligent VIP customer data management platform is designed and implemented, which achieves rapid extraction, effective analysis, early warning and graphic display for the data. In addition, it effectively guides customer managers to make business decisions, and provides managers with first-hand data on the customer managers’ performance evaluation. The actual operation in a large domestic telecom company proves that the platform is stable, efficient and safe.

Rui Li, Miao-Jie Sang, Ke-Bin Jia

Design and Implementation of Operation Energy Management System Based on AJAX-SSH2

With the development of communication network construction, the operation consumption of communications industry grows significantly in recent years, and the high cost of electricity is hard to be ignored. Therefore, it is necessary to develop an operation energy management system to ameliorate the cost control of electricity. By integrating AJAX with SSH2 framework, AJAX-SSH2 lightens the burden of server and optimize users’ interactive experience with the intrinsic performance of low-cost system maintenance and function extension.This paper designs and implements the Operation Energy Management System based on AJAX-SSH2 framework, achieving the functions include data entry, data query, report export, budget analysis, energy-saving evaluation, tendency prediction, and user management.This system provides a platform for centralized management and analysis in order to make the process of energy-saving and consumption controllable, which finally achieves the power optimization of communication network.

Ye Yuan, Ke-Bin Jia, Qi-Te Wang

Network Operation and Maintenance System of Meticulous Management Based on Data Analysis

Due to rapid development of the domestic communication industry, energy consumption on network operation and maintenance has become one of the main energy consumptions in China. With the aim of obtaining meticulous managementon enterprise power, raising electricity availability and evaluating the effect of energy-saving measures, it’s crucial for us to develop an intelligent system for data analysis. We first introduces a management system based on B/S architecture and MVC framework with multi-functions of information inquiry, budget analysis, energy management and etc.; and then focus on keytechnologies such as database modeling, database index, stored procedure and trigger and least square method. This intelligent system has been successfully employed in a communication enterprise and has been proved accurate, stable and efficient.

Qi-Te Wang, Ke-Bin Jia, Ye Yuan, Yu-Xin Song

Image Annotation and Refinement with Markov Chain Model of Visual Keywords and the Semantics

This paper presents a discriminative stochastic method for image annotation and refinement. We first segmented the images into regions and then cluster them into visual blobs with a small number than the whole training image regions. Each visual blob is regarded as a key visual word. Given the training image set with annotations, we find that annotation process is conditioned by the selection sequence of both the semantic word and the key visual word. The process could be described in a Markov Chain with the transition process both between the candidate annotations and the visual words set. Experiments show the performance of this annotation method outperforms the state of art methods.

Zhong-Hua Sun, Ke-Bin Jia

The Power Spectrum Estimation of Signal Based on Neural Networks

In this paper, in order to extract the needed information from Wireless Signal Detector, some different frequently–used methods and the related parameter estimation algorithms are compared. Besides, a new method is proposed based on signal spectrum estimation of neutral net which was used to compute signal power spectrum estimation. By simulating the practical data on MATLAB, the factor of neural network can be trained to extract the signal spectrum. The velocity of spectrum estimation is far better than some other algorithm of AR model spectrum estimation method when using proposed method, and spectral estimation quality consistent.

Guang-Min Sun, Wei Liu, Hai-Tao Yan, Fan Zhang, Hao-Cong Ma

Algorithm of Laser Spots Recognition Based on “Cat Eye Effect”

The paper designs an efficient recognition algorithm to detect laser spots with high Real-time. An improved Otsu is proposed to identify the target and exclude the noise on the condition of atmospheric turbulence. The problem that spot center is not accurate caused by the phenomenon of "supersaturated" is efficiently solved as well. Firstly, the difference the foreground image and the background image is calculated, the morphological filtering is done to segment the target. The elliptic characteristic of the target is used for preliminary identification. Many steps are designed in order to remove False-alarms and improve recognition accuracy. The algorithm is applied in the hardware TMS320C6455 system. Extensive experiments show that the algorithm not only ensures the matching accuracy but also improves the time response.

Qiang Wu, Li-Xiao Yao, Xu-Wen Li

Computer Networks and Mobile Computing


A Synchronization-Free Cooperative Transmission Protocol to Avoid Energy-Hole Problem for Underwater Acoustic Networks

Underwater acoustic sensor network with terrestrial wireless sensor network characteristics distinct, underwater acoustic sensor network uses sound waves to be transmitted, resulting in underwater network with high propagation delay characteristics, if applied directly to land-based routing mechanism water network, will give rise to a number of problems. Therefore, this paper designed for underwater network environment routing protocols, thereby reducing the underwater acoustic sensor network to send packets to the end-to-end delay and for collaborative work on underwater acoustic transmission sensor network, encountered the challenges of high propagation delay characteristics discussed. This paper presents the work last sleep period, coordinated and collaborative collection of neighbor information transfer protocol. Through simulation, this paper found that the proposed routing protocol SCTP, the delay in the point to point, network transmission performance and power consumption of a significant improvement in performance.

Ying-Hong Wang, Yao-Te Huang, Kai-Ti Chang

The Study of Using Game Theory for Live Migration Prediction over Cloud Computing

Cloud computing was a technology in recent years which had been concerned. More and more network applications provided client a more convenient experience for use on the cloud computing service. Cloud computing is using virtualization technology. It can not only improve the performance on the server, but including a characteristic dynamic data assignment. Additionally, any server with fault, over loading or maintenance…etc. which need to be stopped, the user is not aware that the service has interrupted, that is because the technology of live migration will quickly backup the remaining data from original server to another server. The study [1] used Gilbert-Elliot model has a capability to predict the probability on dirty page until performing 10 times iteration. From this study, using Game Theory model of reducing predicted number effectively can early determine whether to go the stop-and-copy phase. That saves the time on live migration.

Yen-Liang Chen, Yao-Chiang Yang, Wei-Tsong Lee

A Greedy-Based Supply Partner Selection for Side-by-Side 3D Streaming on P2P Environment

In the world of today, the mobile device technologies are advancing with each passing day, the humans use them to watch videos and deal business anywhere. However, to support the huge stream so that the bandwidth will face enormous challenges. Because such technology is easy to lead the network congestion and longer waiting time. In this paper, we use the front camera of a mobile device to track user’s viewing angle then to calculate the current the 3D stream of current needs then find the most suitable peer for source supplying. This way is consider to the maximum available upload bandwidth to chosen supply partners to relieve the network additional burden.

Yu-Jhih Wang, Hsin-Hung Cho, Wei-Chung Liu, Han-Chieh Chao, Timothy K. Shih

A SIP/IMS Platform for Internet of Things in WLAN-3GPP Integration Networks

With the growth of internet access technologies, sensors/machines performing environment sensing and control can connect to Internet anytime and anywhere. However, there is no framework for users to integrate the sensors/machines and the server. The

IP Multimedia Subsystem

(IMS) based on

Session Initiation Protocol

(SIP) and all-IP architecture has been proposed as a common platform for

Next Generation Network

(NGN). This paper proposes a SIP/IMS platform for

Internet of Things

(IoT) in WLAN-3GPP integrated networks. This IMS platform includes

Instant Message and Presence Service


application servers

(ASs). The

Call Session Control Functions

(CSCFs) perform the session setup and termination for a long-term communications. The IMPS ASs handles message exchange for short-term communications. Finally, this paper analyzes the efficiency of difference query/report methods and gives a conclusion.

Whai-En Chen, Shih-Yuan Cheng

Globally Optimized Cooperative Game for Interference Avoidance in WBAN

Wireless Sensor Networks (WSNs) have been developed for collecting and monitoring environmental data over the years. Recently, based on the idea of WSN, researchers have begun to monitor the human or animal body by placing sensor nodes on the skin or inside the body. The wearable relay nodes then collect biosignals from the sensor nodes and send the collected data to a sink node for data storage. However, the coverage is a typical problem in WSNs, which may not only generate interference but also affect system reliability. While dealing with very important bioinformation, supposing the system reliability is decreased due to high delay or packet loss, important bioinformation might be lost and could be life-threatening when, for example, a patient’s heart stops beating but medical personnel are not warned. Therefore, this paper proposes a non-zero-sum cooperative game model to control the transmission power of the system for reducing the interference level between simultaneous transmissions and solving contention between different messages.

Wen-Kai Liu, Tin-Yu Wu

A Study of Random Neural Network Performance for Supervised Learning Tasks in CUDA


Graphics Processing Unit (GPU)

have been used for accelerating graphic calculations as well as for developing more general devices. One of the most used parallel platform is

Compute Unified Device Architecture (CUDA)

. This one allows to implement in parallel multiple GPU obtaining a high computational performance. Over the last years, CUDA has been used for the implementation of several parallel distributed systems. At the end of the 80s, it was introduced a stochastic neural network named

Random Neural Networks (RNN)

. The method have been successfully used in the Machine Learning community for solving many learning tasks. In this paper we present the gradient descent algorithm for the RNN model in CUDA. We evaluate the performance of the algorithm on two real benchmark problems about energy sources, and we compare it with the obtained using a classic implementation in



Sebastián Basterrech, Jan Janoušek, Vaclav Snášel

Multimedia Signal Processing and Classification


Feature Line-Based Local Discriminant Analysis for Image Feature Extraction

In this paper, a novel image feature extraction algorithm, entitled Feature Line-based Local Discriminant Analysis (FLLDA), is proposed. FLLDA is a subspace learning algorithm based on Feature Line (FL) metric. FL metric is used for the evaluation of the local within-class scatter and local between class scatter in the proposed FLLDA approach. The Experimental results on COIL20 image database confirm the effectiveness of the proposed algorithm.

Jeng-Shyang Pan, Shu-Chuan Chu, Lijun Yan

Research and Design of Campus Network VOD System

VOD (Video on Demand) is referred to as video-on-demand technology, also known as interactive television on-demand system, meaning that the corresponding video playback program according to the user’s needs. If the broadband network were highway, then VOD is the most eye-catching car on the road. VOD originally appeared is to better meet user’s demand for autonomy to watch video programs, but with the continuous progress of VOD technology, the widely use have produced a strong impact on mass culture and business models, and it has got currently being widespread attention and application of the education sector. In order to meet students’ needs and aspirations of real-time learning, in order to better carry out two-way multimedia school teaching, in order to achieve the network management of educational resources, it is imperative to build a campus network-based video-on-demand system.

Kuiliang Xia, Xiaoming Song, Xiangfeng Suo

Prediction of Chaotic Time Series of RBF Neural Network Based on Particle Swarm Optimization

Radial basis function (RBF) neural network has very good performance on prediction of chaotic time series, but the precision of prediction is great affected by embedding dimension and delay time of phase-space reconstruction in the process of predicting. Based on hereinbefore problems, we comprehensive optimize embedding dimension and delay time by particle swarm optimization, to get the optimal values of embedding dimension and delay time in RBF single-step and multi-step prediction models. In addition, we made single step and multi-step prediction to the Lorenz system by this method, the results show that the prediction accuracy of optimized prediction model is obvious improved.

Baoxiang Du, Wei Xu, Bingbing Song, Qun Ding, Shu-Chuan Chu

Segmentation and Description of Human Facial Features Region

Although there are many search engines based on the text content, that do not meet our extensive search for multimedia content, interests and effectiveness requirements.MPEG-7 is an important criterion for the multimedia content descriptions, so it has a great application value for us to have an in-depth study. Combined with the face recognition algorithm of face regions, feature relationships describing face images content in this paper, according to the study of MPEG-7 standard construction of face related structure models.

Yingli Wang, Yao Wang, Song Li

Design and Implementation in Image Compression Encryption of Digital Chaos Based on MATLAB

A chaotic encryption algorithm used digital image compression and encoding technology based on discrete cosine transform and discrete wavelet transform is proposed in this paper. By taking the redundancy of images and the shortcomings of human visual into consideration, this passage conducts the original image compression firstly, and then uses the discrete Logistic chaotic sequence to achieve image encryption and transmission. From the theoretical analysis and experimental results: this compression encryption algorithm is feasible and it decreases the redundant information of the image, furthermore, it reduces the storage space and improves the efficiency of data transmission, computer simulations proved the security of the image processing method.

Zhiqiang Li, Xiaoxin Sun, Qun Ding

Depth Map Coding Based on Arbitrarily Shaped Region

In 3D video processing systems, efficient depth map coding is vital since the quality of the synthesized virtual views highly depends on the accuracy of the depth map. In this paper, we propose an efficient depth map coding method based on arbitrarily shaped region. A depth map is first divided into some regions along the detected edges, and then the edges are coded without loss using a directional 8-connected chain code and an arithmetic codec while the pixels inside the arbitrarily shaped regions are coded by a Differential Pulse Code Modulation (DPCM) with uniform quantization. Experimental results show that our method can obtain better quality of rendered views than JPEG at the high bitrate.

Ruizhen Liu, Anhong Wang

Emotional Impact on Neurological Characteristics and Human Speech

This article discusses impact of human emotions on physiological characteristics and their changes. Many fields require applications that provide information about the emotional state of a human. Today’s research is mainly concerned with increasing the accuracy of the methodology for obtaining this information. Studied subjects were psychologically stimulated to change their neutral calm state to stress. Subjects were measured physiological characteristics and the change of speech also. Blood samples, ECG and EEG form part of the neurophysiological data that were collected during the neutral state and during stress. Voice activity was recorded from reading text that read, patients in both emotional state. Features extraction was focused on the Mel-frequency Cepstral coefficients and their dynamic and accelerated derivations. Change in emotional state from neutral to stress was recognized by using a GMM classifier that has been trained and tested by mentioned speech features. Psychological stimulus was induced using professional psychological methods. The measurement was performed in a special EMC interference protected chamber to prevent undesirable electrical influences from the external environment especially on sensitive EEG measurement.

Pavol Partila, Jaromir Tovarek, Jaroslav Frnda, Miroslav Voznak, Marek Penhaker, Tomáš Peterek

Intelligent Data Analysis and System Reliability Modeling


An Effective Approach Based on Partial Duplication for Reducing Soft Error Rate in SRAM-Based FPGA

In this paper, we present an effective approach for reducing soft error rate (SER) in SRAM-based FPGA. First, the entire system is divided into several modules according to its function. Then the soft error rate of each module is calculated by an analytical estimation method. Finally, rather than performing mitigation for all the modules in the system to achieve high reliability, the modules with highest soft error rate have a priority to be mitigated, i.e. we perform mitigation for the entire system based on partial duplication. Experimental results verify our proposed method.

Baolong Guo, Guochang Zhou, Jinfu Wu, Xiang Gao, Yunyi Yan

Research on the System Reliability Modeling Based on Markov Process and Reliability Block Diagram

The on board computer systems have a high requirement on the reliability because of single event upset. So this paper analyzes the system reliability model, and presents a method to model the reliability of on board computer system. This paper defines the basic module of the system reliability using Markov model and proposes a system reliability prediction algorithm using reliability block diagram. By simplifying the reliability block diagram using the graph theory, to predict the reliability of a system.

Guochang Zhou, Baolong Guo, Xiang Gao, Dan Zhao, Yunyi Yan

Precision Mosaicking of Multi-images Based on Conic-Epipolar Constraint

In this paper a robust mosaic method based on conic-epipolar constraint is proposed. The main characteristics of the proposed method include: (1) Several new methods are presented to realize fast and accurate interest points extraction under various different scenes, including SURF based feature points detection, interest points selection based on uniform distribution. (2) the transformation parameters are estimated using the invariant of conic-epipolar constraint and the most “useful” matching points are used to register the images. Experiment results illustrate that the proposed algorithm carries out real-time image registration and is robust to large image translation, scaling and rotation.

Meng Yi, Yan Chu, Yunyi Yan

A Fast Image Enhancement Algorithm Using Bright Channel

After summarizing the poor-illumination image enhancement methods and analyzing the shortcomings of the currently well-performed multi-scale Retinex algorithm, this paper proposed a new image speedy algorithm with detailed illumination component information. It combined illumination imaging model with target reflection features on RGB color channel, raised a new bright channel concept, and obtained computation method of illumination components by analysis. Then, illumination components were gained precisely through image bright channel gray-scale close computation and fast joint bilateral filtering. Consequently, target reflection components on RGB channel could be solved by illumination/reflection imaging model. The proposed algorithm can get excellent edge details through simple and quick computation. After being removed from the illuminative effects, the images gained are natural-colored, highly visible, and with no halo artifacts. This paper also resolved color casting problem. Compared with NASA method based on multi-scale Retinex, the proposed algorithm improved computation speed, received vivid colors and natural enhancement result.

Long Chen, Wei Sun, Jiaxing Feng

Software Analysis for Transient Faults: A Review of Recent Methods

Transient faults in a computer system could influence the behavior of the software and pose a big threat to the dependability of the system. So a variety of software analysis methods are invented to characterize the property of the program in the face of transient faults and yield a lot of instructive information that could be utilized to improve the dependability of the software. In this paper, we first summarize some typical software analysis methods that are related to transient faults and introduce each method briefly. Then we make a comment on them and recommend some prospective methods.

Guochang Zhou, Baolong Guo, Xiang Gao, Weikang Ning, Yunyi Yan

A Novel LCD Driver Chip for Brightness Adjustment System

Based on the liquid crystal display, in order to obtain the best image display effect and how to prolong the life of the liquid crystal the two problems, basing on BiCOMS technology of Samsung company, we design and implement a simple structure, economical and practical LCD brightness control circuit and a polarity switching circuit. The relationship between the gray voltage and gray scale brightness using in the circuit, so we can design economical and practical LCD brightness control circuit; at the same time in order to extend the life of liquid crystal, we design a polarity switching circuit, the value of liquid crystal voltage providing the data driving circuit become near zero in the average time and reduce the DC component, thus it can extend the life of liquid crystal.

Hui Wang, Song-lin Wang, Ling-xia Zhang, Yun-yi Yan

Arrhythmias Classification Using Singular Value Decomposition and Support Vector Machine

The main aim of this work is to recognize arrhythmias in ECG records. Many algorithms for this task have been proposed in the past, but in our solution we try to reduce redundancy of information in the signals by Singular Value Decomposition. The reduced dataset is classified by Support Vector Machine. Our approach gives very satisfactory results which can be used in medical practice. This expert system should offer automated recognition between physiological beat and one of the three basic pathological beats: Premature ventricular contractions, Right bundle branch block and Left bundle branch block.

Tomáš Peterek, Lukáš Zaorálek, Pavel Dohnálek, Petr Gajdoš


Weitere Informationen

Premium Partner

BranchenIndex Online

Die B2B-Firmensuche für Industrie und Wirtschaft: Kostenfrei in Firmenprofilen nach Lieferanten, Herstellern, Dienstleistern und Händlern recherchieren.



Best Practices für die Mitarbeiter-Partizipation in der Produktentwicklung

Unternehmen haben das Innovationspotenzial der eigenen Mitarbeiter auch außerhalb der F&E-Abteilung erkannt. Viele Initiativen zur Partizipation scheitern in der Praxis jedoch häufig. Lesen Sie hier  - basierend auf einer qualitativ-explorativen Expertenstudie - mehr über die wesentlichen Problemfelder der mitarbeiterzentrierten Produktentwicklung und profitieren Sie von konkreten Handlungsempfehlungen aus der Praxis.
Jetzt gratis downloaden!