Abstract:Plot plan is a vital part of 3D mobile phone animation automatic generation system, and plays an important guiding role for the presentation of the animation. In this paper, we apply the Semantic Web technologies to build event ontology. Based on the key information extraction, the events which are hard to be expressed can be represented through a series of specific actions. It is possible to express a complete animation content, so that not only makes the animation with certain scenarios, but also creates a richer animation effects.

Abstract:Cloud computing provides a pretty new and efficient way to deploy scalable web application. In this way, it can allocate their computing resource on business requirements for enterprise applications. Micro-service architecture is used to split large applications into a series of small modules that can be developed, tested, deployed, operated and upgraded independently. Micro-service architecture provides a more efficient way for a large number of Internet companies to expand their applications in the cloud, reduce complexity of application development and gain agility. In this paper we analyze and test the micro-service architecture model in a specific case-we develop and deploy the enterprise application system in the cloud to implement performance tests of two architectures (single-mode architecture and micro-service architecture mode), and we can acquire the evaluation result. These results have some guiding significance for solving the problems that may be encountered in the micro-service application of enterprise application.

Abstract:In big data services, duplicate removal of public opinion information is inevitable, and it lacks theoretical guidance. There is a research on the classical duplicate removal algorithm such as SimHash, MinHash, Jaccard, Cosine Similarty, as well as common segmentation algorithm and feature selection algorithm in order to seek excellent performance of the algorithm. The Jaccard based on short article and the SimHash algorithm based on Cosine Distance are proposed to improve the traditional algorithms. Aiming at the problem of the low efficiency of experiment on many research subjects, the strategy is adopted that filters out algorithm of obvious advantages by vertical comparison firstly, and gets the most appropriate algorithm collocation by horizontal comparison secondly, at last, makes a comprehensive comparison. The experiment of 3000 public opinion samples shows that improved SimHash has better effect than traditional SimHash; improved Jaccard increases the recall rate by 17% and improves the efficiency by 50% compared with traditional Jaccard. Under the condition that the accuracy is higher than 96%, MinHash+Jieba full pattern word segmentation and Jaccard+IKAnalyzer intelligent word segmentation has more than 75% recall rate and good stability. MinHash is a bit weak than Jaccard in the aspect of removal effect, yet has the best comprehensive performance and shorter feature comparison time.

Abstract:3DStudio（3DS） is a file structure of 3dsMax used to store model information, which is one of the industry's standard format in 3D software. In order to ensure the models of different system can be called freely, to solve the problem that the different model format is not compatible in visual system, and to reduce the storage space of 3DS file, this paper proposes to use the Backus-Naur Form (BNF) specification to describe the 3DS file. This paper describes the characteristics of the BNF and the method of describing and analyzing the 3DS file structure, and takes the 3DS file of Tethys model as an example, using BNF to formally definite it, thus forms a unified format of standardization management, which facilitates retrieval and application of search engine.

Abstract:In front of multi-threaded concurrent test requirements, Junit and TestNG these two simple unit test framework has been used to compare, but users generally prefer to the test framework that configuration is simple, flexible, and easy to operate and satisfied the test requirements. This thesis has made an experimental comparison of Junit and TestNG on multi-threaded concurrent test methods, to verify the two frameworks implementation for multithreaded concurrent test, and provides the basis for the user to select the tool in the actual test scenario for users.

Abstract:To solve the problem of state space explosion caused by formal modeling methods, a simulation model and performance evaluation method for high performance cluster based on Petri net are proposed. Firstly, the system structure of high performance cluster is analyzed and a general model is constructed. Then the sub-models for task generation and scheduling are built respectively for cluster system. Finally, according to the characteristics of different kinds of tasks, the corresponding task processing models of different tasks are designed by coloring the Petri net. The simulation results show that the proposed model could effectively evaluate the effects of key parameters on cluster performance.

Abstract:A database construction method which is based on the integration of relational database and distributed file system is researched for a large sample data base for interpretation of remote sensing images. It analyzes the database's construction process, and discusses the key technical problems and solution method. Based on the sample database, it studies the application method of sample data, analyzes some kinds of using cases and explores the application mode of sample data. Individual region's massive sample data are selected for verifying the method and its efficiency. At the same time, it takes 10 first-level classes which are defined in the land cover classification system for example, to analyze the spatial distribution and density characteristics of all kinds of sample data. The results show that the method of database construction and management which is based on the integration of relational database and distributed file system is very effective and applicative for sample data's searching, analyzing and promoted application.

Abstract:Due to the environment and human activities, the growth and distribution of vegetation has a great change in the middle reaches of the Tarim River. We can monitoring the vegetation growth and distribution in the middle reaches of the Tarim River to provide a scientific and reliable basis for the enactment of the ecological conservation policy and vegetation protection in the middle reaches of the Tarim River. This paper adopts the object-oriented Geodatabase model, using geographic information system (GIS) to analysis spatial data and attribute data of the middle reaches of the Tarim River, selecting the normalized difference vegetation index (NDVI) in 2000, 2006, 2010, 2015 as the research data which is calculated through the remote sensing image processing platform ENVI, using ArcGIS for the research of the spatial distribution characteristics of NDVI. This paper designs a vegetation index spatial database based on Geodatabase for middle reaches of Tarim River to accomplish the integration of storage and management of spatial data and attribute data. This spatial database can reflect the changes of vegetation growth and distribution directly in the study area.

Abstract:With the growing demand for massive file storage and high number of concurrent accesses to enhance performance of the system. In this paper, we propose an optimization strategy of web applications, which can expand the storage capacity, availability and reliability with MongoDB GridFS, and also can support load balancing and hot standby with Nginx and Keepalived, optimizing concurrent performance. We test the concurrent performances of the web application and the experimental results show that the average response time could be reduce by 9% when the number of concurrent accesses increases to more than 80. When files are uploaded concurrently through Nginx Gridfs, the performance is stable.The file-download speeds of larger files are faster than those through local file system EXT4.

Abstract:In the real time video surveillance network system, if the initial video data received by the monitoring client is not complete, the screen will show blurred screen. For H.264 coding standard, this paper designs and implements real time video stream dispatch plugin based on the open source multimedia GStreamer framework. The plugin uses a randomly requested source pad to dispatch video data. It also caches current IDR frames group to ensure the integrity of initial IDR frame to be sent. The plugin can be used in streaming media server to solve the problem of the blurred screen of initial screen effect in video surveillance. Experiment shows that the streaming media server with this plugin can effectively distribute the real time video data and ensure the integrity of the initial playback screen, which has a significant effect on improving the real-time monitoring effect.

Abstract:In order to meet the quality management of quality and cost control in the Asphalt mixing plant admixture feeding process, we design remote monitoring system of intelligent batch feeder. Through the existing GRPS/GSM of China Mobile, the system monitors the feeder remotely with the comprehensive functions of data acquisition and transmission, data monitoring, analysis, statistics, management, and alarm handling, achieving real-time monitoring, user management intelligent feeding machine, user management alarm, and data statistics and analysis. The system is uses in the Ji River Highway, Shanxi province, to solve the bad environment of intelligent feeding machine, difficult issues such as monitoring, achieving the goal of remote monitoring and control of intelligent feeding machine, which can effectively guarantee the quality of the production of asphalt concrete.

Abstract:The rapid development of modern information technology makes the amount of every day data increase at the speed of TB, how to effectively use and manage the big data with explosive growth? It is a problem need to be solved urgently. Big data has penetrated into various fields including agriculture, with the agricultural informatization construction and the application of Internet technology in agricultural production, resulting in a large amount of agricultural data to be stored, managed and processed. Based on the background of the construction of agricultural informatization in Pengzhou Gexian mountain agricultural demonstration base of Chengdu agricultural science and technology vocational college, according to the demand of informatization construction of agriculture, we build high performance agricultural big data management platform based on Hadoop, realizing the agricultural big data safe and reliable storage, intelligent management and application. And ultimately we achieve the purpose of early intelligent warning of agricultural production, intelligent decision-making and intelligent analysis, providing professional guidance to farmers. This lays the foundation for China to enter the intelligent agricultural era based on fine planting, precise control, visual management and intelligent decision-making.

Abstract:In order to make the data format of NBI (Neutral Beam Injection) system compatible with EAST data and to make international exchange of experimental results effectively, an NBI data acquisition system based on MDSplus (Model Drive System plus) is developed. Double buffer technology and multi-thread mechanism are adopted in data acquisition module to achieve continuous and long-pulse data acquisition. Data storage module selects MDSplus and jScope which can access the experimental data and is packaged in MDSplus's program developing tools to resolve the problem of unifying NBI data format and sharing NBI experimental data. The system has met the requirements of the experiment in the test and lays a good foundation for the next round EAST-NBI long-pulse injection experiments.

Abstract:The course of design patterns is an important core course of software engineering. The course is characterized by both theory and applicability. But the abstract feature brings obstacles to students. According to the features of the course and the students' cognition, we design and implement a teaching platform based on Spring Framework-online education management system. The system adopts B/S architecture, and integrates multiple design patterns. It shows not only the features of individual design patterns, but also their composite application in a practical project. Thus it helps the students to have a deeper understanding of the essence theory of design patterns, enhance their programming ability and lay the foundation for them to learn more advanced knowledge in software engineering.

Abstract:The Airspace Reject Cognizance System is designed to solve the problem existing in the management of unqualified products in a certain aerospace industry in daily work. The system is highly expandable and also it has uniform management platform which can carry out overall managements such as: the unqualified product trial, the unqualified product prevention, the zero return request and the basic establishment. Using WebService to integrate external systems, the result of feedback data from Airspace Reject Cognizance System can be analyzed. The test and operation show that this system is feasible and reliable.

Abstract:Multiple MapReduce tasks are needed for most of current distributed parallel reasoning algorithm for RDF data; moreover, the reasoning of instances of triple antecedents under OWL rules can't be performed expeditiously by some of these algorithms during the processing of massive RDF data, and so the overall efficiency can't be fulfilled in reasoning process. In order to solve the problems mentioned above, a method named distributed parallel reasoning algorithm based on Spark with TREAT for RDF data is proposed to perform reasoning on distributed systems. First step, alpha registers of schema triples and models for rule markup with the ontology of RDF data are built; then alpha stage of TREAT algorithm is implemented with MapReduce at the phase of OWL reasoning; at last, reasoning results are dereplicated and a whole reasoning procedure within all the OWL rules is executed. Experimental results show that through this algorithm, the results of parallel reasoning for large-scale data can be achieved efficiently and correctly.

Abstract:Squarified layout algorithm is widely used in the Treemap Visualization, but classic Squarified algorithm cannot achieve the best average aspect ratio. By analyzing implementation details of Squarified Treemap layout algorithm, especially each step of the rectangular block position selection process, the paper demonstrates the drawback of classic Squarified algorithm caused by using greedy algorithm. Combining with depth-first search technique, it also proposes Squarified Treemap layout algorithm based on depth-first search (DSS algorithm). Based on elaborating implementation process of DSS algorithm, combining empirical research, the advantage of the DSS algorithm in the aspect ratio, the improvement direction and the essential characteristics of the time performance are discussed.

Abstract:In order to ensure that our tracking algorithm can real-time capture the fast-moving target and record its three-dimensional coordinates, the system uses a high-speed tracking algorithm based on Kernelized Correlation Filters (KCF). First, use KCF tracking algorithm to track the target. Second, use ORB feature point detection algorithm to calculate the target feature point. Then find out the corresponding point in Multi-Camera. After finding the corresponding points, use three-dimensional reconstruction theory of Multi-Camera to calculate the three-dimensional coordinates of the target object in each frame. Finally, using polynomial to fit the discrete points of each frame and then get the final trajectory. The experimental results show that this algorithm can track target efficiently and the whole system can meet actual requirements.

Abstract:The covert channel based on network protocol has been widely used for network attack. Mastering the mechanism of covert channel is important to the formulating of corresponding network defense strategy. Due to the widely use of network time protocol, a kind of covert channel technology based on NTP protocol is proposed in this paper. This paper analyzes features of query/response mechanisms in the NTP protocol, utilizes the message field as hidden payload and then designs separated downstream and upstream NTP covert channels, in which secret information is disguised and transmitted as normal NTP messages. The popularity and irreplaceability of NTP message make NTP covert channel endowed with the advantages of great ability of penetration and high quality of concealment. Test results show that the technology could carry relatively considerable information and easily penetrate the network monitoring device. The future work will focus on authentication, encryption and other security mechanisms in NTP covert channel.

Abstract:Based on the perspective of return threshold, this paper discusses the auction problem of grid resources through establishing the utility function model and combining with the dynamic game theory. With the analysis of the dynamic game process between grid resource providers and users, it finds that the optimal policy choice of the provider depends on its preference for monetary and non-monetary benefits, and the highest quotation of the users. Under the conditions of given return threshold, when the grid resource users have lower production profits or pay a higher cost, they will choose the lower price bidding strategy. The research shows that the introduction of return threshold mechanism can maximize the utility of both parties in a certain degree.

Abstract:The change point detection of time series is widely applied in various fields. In some applications, a minimum period is required before a state change. Motivated by such applications, a constrained Hidden Markov Model, which combines with the shortest state continuous length constraint, is proposed in this study. Moreover, a constrained Baum-Welch training algorithm and a constrained Viterbi state extraction algorithm are also given. And experimental results based on the simulation data and GNP data sets indicate that the constrained HMM has higher performance than the general HMM.

Abstract:The large oil depot map is different from city and mountainous region whose maps are complex. The map of large oil depot is very regular, oil tanks and other buildings arranged neatly and the roads of the oil tank area are straight. On the basis, the classic A^* algorithm is improved in this paper. On the one hand, according to the characteristics that the map of oil depot is simple and the number of search nodes is relatively small, the data structure of the Open table in the A^* algorithm is improved, to accelerate the search speed and the ranking algorithm is used to improve the search efficiency. On the other hand, because roads in the oil depot are straight, so we divide roads into two parts: roads with obstacles and barrier free roads, to improve the search efficiency. Experimental results show that, with the combination of the two improved methods, the time of searching roads is declined definitely by 6.86%.

Abstract:LATCH (Learned Arrangements of Three Patch Codes) improves the accuracy of local binary descriptor by comparing three pixel blocks rather than tow pixel blocks. However, the improvement of accuracy brings a larger time consuming. Based on the study of LATCH and other local binary descriptors, using integral graph theory in improved LATCH descriptor, it reduces repeated calculation of each pixel blocks in LATCH descriptors. According to the experiment results, the computation time of the improved algorithm is reduced by 30%-40% compared with the original algorithms, while the accuracy of the improved algorithm is similar to that of the original algorithm.

Abstract:In order to carry out 3D visualization reconstruction of underground tunnel, this paper presents a new method based on sparse 3D scattered points. According to the shape of the underground tunnel, a three-dimension mathematic model of tunnels with trapezium section is built up. Dividing the tunnel into segments, according to the sparse 3D scattered points to get the central position, length data and the vertexes of the tunnel's each segment, we can draw every surface of each segment, so as to reconstruct the whole tunnel. As is proved by experiments, this method needs only a small amount of three-dimension points, and has noise immunity and stability.

Abstract:In the process of oil exploration, development, refinement, transportation and storage, due to accidents or operational errors, the leakage of crude oil or oil products from the reservoir will flow to the ground, water, and the beach or sea. As the different thickness of oil, the leakage forms a film, the phenomenon is called oil spill. Large-scale oil spills often has a serious impact on national ecology, society and economy. Based on the satellite remote sensing data, this paper introduces the change detection into the problem of marine oil spilling, comparing several change detection algorithms with each other, and accomplishes an oil spill detection system. Results show that, the change detection can effectively reduce the impact of geographical location and lighting conditions, and it has a stronger generalization ability than the classification methods trained on single image. The research also shows the new method can effectively separate the sea water region with light oil or heavy oil region.

Abstract:We propose a fault diagnosis method based on the modified cuckoo search algorithm (WCS), steepest descent method and support vector machine (SVM) to improve the accuracy of transformer fault diagnosis. A new inertia weight is also proposed and applied to solve the problem that the convergence rate of cuckoo search algorithm decreases in final iterations. SVM parameters are optimized by the algorithm which is combined with improved cuckoo search algorithm and steepest descent method, overcoming the defects that SVM model is easy to fall into local optimum. Support vector machine is trained on the MATLAB platform using LIBSVM toolbox, and the well-trained SVM will be adopted to diagnose the #1 transformer fault for 110kV Gantang substation. Study of practical cases indicate that, with this method, transformer faults can be diagnosed effectively and accurately, and the accuracy is higher than that using particle swarm optimization(PSO)、genetic algorithm(GA) and grid search(GS).

Abstract:So far, in the field of data mining, people have achieved a variety of algorithms of clustering. And the most widely used is K-means clustering algorithm. But the main problem of K-means is the initial center selection problem. In this paper, a method is proposed to determine the initial K centers of the K-means algorithm through the relationship matrix and the Degree Centrality. Compared with the traditional algorithm, the proposed algorithm can get the better clustering result. Experimental results have proved the validity and feasibility of this algorithm.

Abstract:In view of the diversity of SQL query program, an accurate scoring model is presented. First, based on the common standard SQL syntax specification identifier, command clause, the synonym chain and the abstract syntax tree, SQL program is converted into a kind of unified intermediate form, fully eliminating the SQL program syntax and semantic differences. Then, referring to the artificial grading thought, the standardized code is transformed into the token sequence according to grading points, and the improved algorithm for Longest Common Subsequence(LCS) is used to grade the program similarity. The scores are calculated according to the weight of the scoring points, and the error location is given as well. Finally, samples are tested and anglicized to illustrate the effectiveness of the grading model.

Abstract:In fact, there widely exists a kind of time series data that is non-stationary but can be transformed into several local stationary time series data, the identification problem of the non-stationary time series data is studied, and then this paper proposes a piecewise recursive identification algorithm with mechanism. Based on the definition of stationary time series data, the precipitation algorithm which has the mean variance point, the mutation point and the mutation point of the autocorrelation function, is constructed based on the statistical characteristics such as the mean and variance and autocorrelation function. On this basis, a series of locally stationary sub sequences are identified from the identified non-stationary sequences, and then, the Burg algorithm is applied to the recursive identification of local stationary subsequences. The experimental results show that the new algorithm can divide the boundary points of the local stationary sub sequences with smaller position deviation. At the same time, the calculation efficiency is improved significantly under the condition of high accuracy.

Abstract:Online reading of electronic documents has been very common. Documents could be downloaded or their full content could be read online by those users who have paid for them, while limited content could be previewed by the common users. By this way, documents' owners can realize the control of the use of electronic documents to a certain extent. But there are imperfections in the way. If one user has pay for the documents and downloaded them, he or she could send them to others without any restriction. Therefore, in order to achieve better copyright protection, it is worthy of study and exploring access control method on how to still make some restrictions on documents when they are downloaded. This paper focuses on the method for making some access restrictions on the electronic documents both online and offline. For online documents, access restrictions would be done on registered users, while DES encryption and decryption is used for controlling the downloaded documents. Experimentations show that the proposed method based on DES encryption and decryption technology could control the access to electronic documents more efficiently and give a stronger force on copyright protection.

Abstract:Concerned the problem that the existing RFID security protocols focus on the privacy and anonymity of RFID system, but ignore the scalability and customizability, a scalable and customizable RFID bidirectional authentication protocol is proposed. Firstly, the tags and client readers are authenticated by target authentication module, the efficiency of the authentication is improved by using the tag grouping based on liner search method and a mapping table; secondly, the malicious users are detected by a simple ID matching detection; lastly, the mutual authentication between tags and servers is realized to enhance the security further more. The analysis results show that the proposed algorithm has a good computational efficiency and security, at the same time it realizes the customizability and scalability.

Abstract:In view of the fact that grid topology errors and bad telemetry information affect the safe operation of power grids seriously, this paper proposes a topology identification method based on improved local outlier factor algorithm. The method uses statistical theory to evaluate the state information of breakers and disconnectors and telemetry information of the power network. While taking into account the different influence of telemetry and remote communication on topology error identification, we weight its data by relative entropy. In the process of abnormal topology state detection, non-anomalous objects are shielded to enhance efficiency of the algorithm. Experimental results show that the algorithm can quickly identify the grid topology errors and find bad telemetry information.

Abstract:In view of the problem that service from provider cannot meet the needs of the users in the network environment, this paper introduces the Quality of Experience (QoE) and Quality of Service (QoS) parameters. Through simulation network, the corresponding data are obtained. And two evaluation models (QoE with bit rate and QoE with packet loss rate) are established by using MATLAB tools. Analyzing the data of evaluation models with statistical analysis method, the model is verified by comparing with Stevens' Power Law model and Weber-Fechner Law model. The result shows that the evaluation models can precisely express the relationship between QoE and QoS, and it can provide the direction of service and an important reference for the service providers.

Abstract:A growing number of cloud services have emerged on the Internet. In the face of a wide variety of cloud services, how to recommend high quality and low price service meeting the users' requirements to the user accurately has become a focus in cloud service recommendation field. Currently, many services recommendation methods are often just based on the current service status without taking into account the performance trend of cloud service. For this reason, this paper proposes a services recommendation model based on performance prediction. The model uses improved artificial neural network based on conjugate gradient to predict the performance of cloud services. Factors such as performance and prices can be compared and calculated by using AHP (Analytic Hierarchy Process), and then the most suitable cloud service would be recommended to the users. The experimental results show that the prediction accuracy would be higher by using improved neural network predicting service performance method, and AHP can recommend the most suitable service to the user according to comprehensively considering the performance and price of services.

Abstract:The data processing algorithm is studied to improve Wi-Fi fingerprint indoor positioning performance. Firstly, Wi-Fi fingerprint samples are collected and then are put into MySQL database and R project. Secondly, the Wi-Fi fingerprint data is divided into several clusters, and the K-mean clustering (K-Means) and fuzzy C-means clustering (FCM) are used to cluster the Wi-Fi fingerprint respectively. Finally, an enhanced clustering strategy (ECS) is proposed to for Wi-Fi fingerprint matching. Experimental results show that ECS reduces the positioning time-consuming about 50%-80% than that consumed by only using FCM and the positioning accuracy is also improved; ECS improves about 20%-40% than that obtained by only using K-Means in terms of positioning accuracy and it proves positioning stability and can automatically update the Wi-Fi fingerprint database.

Abstract:The ECG monitor is an indispensable medical instrument for modern hospitals, communities and families. In this paper, GUI development software Qtopia Core is used to design a simple and human-friendly interface for ECG monitoring system. This interface is realized the functions of data acquisition, processing, display, analysis, storage and transmission of ECG data etc. Finally, the experiments verify the effectiveness of the system.

Abstract:In this paper, we propose a vertical handover algorithm based on speed prediction for mobility users, such as the scenario of the straight highway. Combining with the dwell time algorithm and the straight road scenario, this paper proposes a handover algorithm based on speed prediction. In this algorithm, speed matrices and the corresponding weight matrices are created. Then through mathematics, the speeds and the next interval position of terminals would be determined. Finally terminals make a handover decision with the dwell time algorithm. The simulation results indicate that, the proposed algorithm would increase the handover times, but effectively reduce the handover latency with the same environment and reliable signal strength, comparing to the traditional dwell time algorithm.

Abstract:In order to meet the demand of coal quality analysis in power industry, this paper proposes a new technology of coal analysis based on Hadamard near infrared spectroscopy. On the basis of comprehensive consideration of five aspects, such as development of near infrared spectrometer, the design of control and analysis software, the acquisition of coal spectral signal, the extraction of characteristic information and the establishment of quantitative model, the Hadamard near infrared coal quality analysis system is developed. In the study, 41 standard coal samples with different quality indicators are quantitatively analyzed and predicted. The ability of Hadamard near infrared spectroscopy to predict coal indicators in conditions of same particle size is investigated. Based on LS-SVM algorithm, a coal indicators prediction method is proposed, which has a good correlation with the coal index, and the correlation coefficients is above 0.9.

Abstract:The LT-H10 hobbing machine numerical control system is an embedded system based on ARM processor. The CPU occupancy in ARM-based NC system is always higher than that in x86-based systems for some characteristic of ARM processor itself limits. A μC/OS-II system porting scheme based on MicroBlaze soft-core processor is presented in this paper, to reduce the occupancy of ARM processor by executing some driver programs of master station on the MicroBlaze soft-core processor of FPGA chip on the bottom plate. The experimental test results show that the μC/OS-II system ported on the MicroBlaze processor can run stably. The rapid context-switch benefits to the real-time of the NC system. We design two different tasks scheduling algorithm to measure and research the context-switch time in MicroBlaze-based μC/OS-II system and ARM-based Linux system. This study can not only meet the application needs of CNC system based on ARM, but also can been applied to CNC system based on X86, to achieve the purpose of reducing CPU occupancy, which has important research meaning and application value in Embedded CNC system.

Abstract:At present, Chinese speech recognition has made some achievements. However, due to regional differences in China, different place has different dialect, the Chinese recognition system has low recognition rate and poor performance in the dialect recognition. In order to solve the shortcomings of speech recognition system in dialect recognition, an isolated word recognition system of Hengyang dialect based on HTK is proposed. This method constructs the Hidden Markov Models (HMM), using phoneme as the basic recognition unit and using the HTK3.4.1 toolbox to extract the speech feature parameters of 39-dimensional Mel frequency cepstral coefficients (MFCC). Viterbi algorithm is used to train and match the model to achieve the isolated word speech recognition system of Hengyang dialect. The system's performances are compared under the different phoneme models and different Gaussian mixture numbers. The experimental results show that the system performance can be greatly improved by combining the 39-dimensional MFCC with 5 Gauss mixed numbers and HMM model.

Abstract:Aiming at the problem of poor reusability of the 3D visual scene's repeated development in data center, a visualized editor system for three dimension visualization of data center is proposed. With the aid of 3D engine technology, the system provides a complete, visual and networked environment design platform for managers in the way of simulation, three-dimension and interaction. With high efficiency and flexibility, it can create 3D scene for data center for scene modeling and auxiliary design of the data center. The visual editor is structured on Unity3D engine and web technologies, which not only provides intuitive browser interface and real time environment monitoring system for users, but also solves the problem of remote real-time monitoring of data center monitoring in the way of three dimensional scenes. Users can use it to build a three-dimensional visual scene quickly, reducing duplication of development. Based on other 3D models, this system can be also widely used to build and edit scene of three dimension visualization for other non data center.

Abstract:With the continuous development of Android system, the Android system has been used in a variety of devices, including applying to wire telephones. Therefore, we need to develop a contact which has the real-time synchronization function between wire telephone and mobile phone equipped with Android system. The contact establishes connection between client and server by using Socket, and using Handler mechanism to send data and read data, as a result, it realizes the function of the real-time synchronization for contact data. The test shows the contact achieves the goal of the real-time synchronization function of calling records, adding, modifying and deleting contacts between mobile phone and wire telephone.

Abstract:With the continuous development of computer numerical control (CNC) system, there are higher requirements of accuracy and speed for CNC system, which brings grave challenge for the current most widely used single-core processor platform. In this paper, based on isomorphic quad core ARM processor platform which is under the Linux system environment with the real-time patch, according to the characteristic of hard real-time tasks, soft real-time tasks and non real-time tasks running parallel in the CNC system, the CNC system tasks are distributed to multi-core ARM reasonably. The improved RM algorithm is used to schedule. Finally execute the real-time performance test for the new scheduling algorithm. This scheduling model can solve the problems of mixed tasks scheduling of CNC system admirably, can well meet the real-time requirement of real-time control of CNC system.

Abstract:This paper proposes a two-step fixed audio retrieval algorithm based on zero crossing rate and audio fingerprinting. The iterative calculation of the histogram and the sliding step of the observation time window are used in preliminary retrieval based on the zero crossing rate histogram to reduce the amount of calculation and speed up the search, fast filtering out candidate audio segments with high similarity; Then based on the dimension reduction Philips audio fingerprint, accurate retrieval of the candidate audio is carried out, further improving the retrieval accuracy. The experimental results show that the audio retrieval algorithm can improve the retrieval speed greatly and has good robustness, ensuring good retrieval accuracy.

Abstract:On this paper, the research mainly focuses on the technology of indoor location. First, we collect coordinates of sample spots by handheld smart device in the specified range and information of wifi wireless hotspot, with which the location fingerprint technique is applied to absolute positioning; In order to improve the real-time performance and accuracy of indoor positioning, the pedestrian dead reckoning algorithm which will measure stride frequency, estimate step size and detect the direction with data collected and processed by the mobile phone sensor is used to achieve relative position estimation. The pedestrian dead reckoning algorithm can overcome the instability of the location fingerprint technique and the latter can reduce the cumulative error that the former brings. The result shows the combination of the two technologies is able to effectively improve the accuracy of indoor location and be fully used in the real world.

Abstract:Due to the high degree of informatization in cigarette factory and the large amounts of collected production data, this thesis focuses on how to effectively use these production data to monitor the abnormal information during the process of production. In previous monitoring, managers make adjustments to abnormal situations by watching the production board or artificial feedback. In this way, the response speed of abnormal situations is slow and leaky. Therefore, this paper proposes a method of abnormal information analysis to obtain effective abnormal information during the production of rolling and packaging workshop, meanwhile using mobile platform to push these effective abnormal information to relevant manager. By this way, the response ability and the control level of production can be improved.