Earticle

In coded-structured light three dimensional measurement system, system calibration plays an important role for the measurement accuracy. Projector calibration is an important part of the system calibration. Therefore, this paper proposes a projector calibration method with simple calibration process and high accuracy. This method combines Zhang’s plane model calibration method with orthogonal Gray code + trapezoid phase shift code. During the calibration process, this paper uses Gray code pattern and trapezoid phase shift code pattern to establish the relation of projector image and camera point coordinates, the corner extraction accuracy of this paper’s method is higher than the traditional orthogonal Gray code. According to the image coordinates in the projector’s perspective, we program and calculate the projector’s internal and external parameters matrix based on the Zhang’s plane model calibration toolbox. The results show that this paper’s method is simple and flexible, the maximum relative error of the calibration parameters is 0.027%, and it meets the requirements of system calibration in engineering or industrial fields.

The digital revolution is occurring in the publishing market due to the development of IT industry. This environmental change is highly influencing on the changes of publishing market and the culture as well as on the consumption and reading patterns of consumers. Apparently the sentimental old-media “paper book” is rapidly being replaced by the e-book which is the digital new-media. Despite the many useful features of the e-book, the current situation shows that it has stopped from developing to the mature stage due to the disincentive factors occurring during the process of embracement. Therefore, the aim for this thesis is to suggest strategic direction for the stable substructure while approaching to the fundamental problem that occurs during the introduction of e-book in an overall perspective of the publication. As a result, the activation and the stable settlement of a local e-book market will take in place.

Risk dominance and payoff dominance strategy are two complementary parts of the game theory decision strategy. While payoff dominance is still the basic principle in perfect information, two player games, risk dominance has shown its advantages in imperfect information conditions. In this paper, we first review the related work in the area of estimation methods and the influence of risk factors on computing game equilibrium. Then a new algorithm, UCT-Risk is proposed in this paper, which is a modification of UCT (UCB apply to Trees) algorithm based on risk estimation methods. Finally, we implement the proposed algorithm in SiGuo game, a popular imperfect information game in China. The experimental result of the new algorithm shows it correctness and effectiveness.

The melody in music can only be assessed by the assessor’s emotions so it can only be subjective and cannot be expressed in standardized numbers. Therefore nobody can know for sure if two songs are plagiarized or not. In this paper, we attempt to realize a plagiarism analysis system of musical content that can provide judging grounds for plagiarism – usually analyzed in consideration of human emotions - using IT technology. To protect music copyright, both knowledge of music and IT technology is compulsory to create a music plagiarism analysis system. First, by understanding factors of the musical content and analyzing plagiarism between two separate but similar voices. We tried to invent a system that measures the level of similarity. For this we considered the digital music factors for the analysis of voices and searched for methods to use IT technology for judging plagiarism. For an efficient analysis of plagiarism, we utilized the music string provided by JFugue, and designed a plagiarism analysis system for musical content and proposed an abstract syntax tree (AST).

At present, the higher education pays more attention to the improvement of the teaching quality, not the scale expansion. Constructing the scientific, reasonable evaluation index system of college physical education teaching is the basis and prerequisite for the education administrative department to launch the appraisal work. The rapid development of the computer technology not only brings the convenience in the daily life, but also provides the new ideals and methods for the college physical education teaching evaluation. The combination between the college physical education teaching evaluation and the computer technology becomes the development trend in current college physical education evaluation work. AHP method is an important assessment method. However, in traditional AHP method, human subjectivity may cause a certain impact on the assessment results. In this paper, we propose the Fuzzy AHP-Entropy method. This method can reduce the influence of the human subjectivity on the results. We apply this method to college physical education teaching evaluation. Then we use the computer simulation to obtain the evaluate results. Finally, the computer simulation experiment shows that the method is scientific, operability and efficient.

Data mining is the process of extracting hidden patterns of data. Association rule mining is an important data mining task that finds interesting association among a large set of data item. It may disclose pattern and various kinds of sensitive information. Such information may be protected against unauthorized access. Association rule hiding is one of the techniques of privacy preserving data mining to protect the association rules generated by association rule mining. This paper adopts data distortion technique for hiding sensitive association rules. Algorithms based on this technique either hide a specific rule using data alteration technique or hide the rules depending on the sensitivity of the items to be hidden. In the proposed technique, positions of sensitive items are altered while maintaining the support. The proposed technique uses the idea of representative rules to prune the rules first and then hides the sensitive rules.

Support vector machine is an effective pattern classification method. Its time complexity and space complexity is O(n3) and O(n2) respectively for training samples with scale of n. Meanwhile, for core vector machine, the relation between time complexity and the scale of training samples is linear and space complexity is independent with the scale of training samples. In this paper, for the problem of big data classification, the concept and principle of support vector machine are described and support vector machine is converted into the form of minimum enclosing ball, consequently core vector machine is used to efficiently obtain the approximate optimal solution. Experiments confirm that core vector machine algorithm can classify the big data quickly and efficiently.

In this paper we propose a hybrid maximum likelihood decoding (MLD) for linear block codes. For the reliable data transmission over noisy channels, convolutional and block codes are widely used in most digital communication systems. Much more efficient algorithms have been found for using channel measurement information in the decoding of convolutional codes than in the decoding of block codes. Word correlation method can be utilized to use channel measurement information in the decoding of block codes. However as the number of code words becomes larger, the decoding complexity increases dramatically to the power of the number of information bits. The hybrid maximum likelihood decoding can solve the problem of the hardware complexity as well as the computational time. Simulation result for Reed Muller code is presented to demonstrate the effectiveness of the algorithm.

This paper considers the problem of ghosting artifacts and stitching seam caused by image stitching and fusion directly, which influenced by the registration and geometric transformation, and the differences between the two images overlap region. This paper proposed maximum flow / minimum cut in graph cut algorithm to obtain the optimal stitching line. Considering the relationship between color difference and structural changes of pixels in the neighborhood, achieving costs or energy division minimize, thereby eliminating the stitching seams. And using the Poisson fusion technology, further eliminate ghosting artifacts caused by the brightness and color differences or moving objects in the image fusion process. The experiment proved that the algorithm of searching optimal stitching line and Poisson fusion algorithm in this paper, can eliminate ghosting artifacts and also be able to maximize avoid stitching seams appears, it presents good performance in applications.

Content Centric Network (CCN) is proposed to facilitate future multicast services. The special attributes of the CCN are the named-data scheme, receiver-initiated transmission, in-network caching, and support of multicasting. The CCN node has three main components: Content Store (CS), Pending Interest Table (PIT), and Forwarding Information Base (FIB). Among the components, the content store caches some portion of content expecting to reuse the content in near future by other clients. This paper describes that when CCN nodes reserve a reasonable amount of the content store memory for the shared content of multiple users, the network provides stable transmission performance due to in-network caching and multicast transmission. On the contrary, if the CCN nodes do not reserve a proper amount of content store space for the shared content, it results in degrading the transmission performance. The simulation results indicate that it is better to employ a FIFO-based replacement algorithm for a shared content flow than to choose usage-based algorithms such as LRU and LFU.

For multi-level semantic structure, the asymmetry of similarity between semantic concepts, as well as the different correlation of semantics between the children nodes and father node, this dissertation proposed a novel similarity calculation method of image semantic based on the probabilistic weighting. This method combines the image feature mapping the visual characteristics of the underlying semantic with the domain ontology description to build a tree-like hierarchical semantic model. According to what the posterior probability and conditional probability were gained by Bayesian network learning, and further for those semantic similarity who are based on semantic distance took the weighted processing so as to get the final similarity of image semantic. Moreover, taking the medical image semantics as the experiment object in weighted method can improve the retrieval performance compared with the non-weighted similarity calculation method.

According to the error approximation problem the sparse preserving projections (SPP) reconstruct the original sample. This paper proposes the algorithms based on orthogonal sparse preserving projections of kernel. In order to get sparse representation coefficients that contain more identification information by kernel method, it mapped samples to high-dimensional feature space to. Then, reconstructing sparse coefficient of kernel sparse representation increase the similar non neighbor sample weight, and reduce heterogeneous neighbor sample weight. Finally, the whole orthogonal constraint transformation improve the ability of sparse retain sample. The algorithm experiments were carried out on the YALE_B and ORL face database, and the recognition rate reached 96.3%, and the results verify the effectiveness and robustness of the algorithm.

The Internet of Things (IoT) is a new evolution in technological advancement taking place in the world today. This paradigm allows physical world objects in our surroundings to be connected to the Internet. This idea comes to life by utilizing two architecture; the Sensing Entity in the environment that collects data and connects itself to the cloud and the Cloud Service that hosts the data from the environment. The combination of wireless sensor networks and cloud computing is becoming a popular strategy for the IoT era. The cold chain requires controlled environment for sensitive products in order for them to be fit for use. The monitoring process is the only assurance which tells if a certain process has been carried out successfully. Taking advantage of IoT and its benefits to monitor cold chain logistics will result in better management and product handling. This paper looks at a system comprising of Arduino wireless sensor network and Xively sensor cloud which can be an ideal system to monitor temperature and humidity of cold chain logistics.

Divisible e-cash allows the user to withdraw a single divisible coin and spends the any sub-coins by dividing the value of the coin. The first divisible e-cash system in the standard model was proposed by Izabachene and Libert. However, the efficiency of the spending protocol and the deposit protocol is very low. In this paper, we construct an efficient divisible e-cash scheme without random oracle by using the Groth-Sahai (GS) proof system and bound accumulators. Our scheme is on-line and truly anonymous without a trusted third party. Comparing to Izabachene and Libert's work, we improve the efficiency of the spending protocol and deposit protocol by introducing a new generational algorithm. Moreover, the bank only needs to look up the coin's serial number in a table of previously spent coins. We give the NIZK proofs of bounded accumulator in the standard model. Some security properties of our scheme, such as anonymity, unforgeability and exculpability, are proved in the standard model.

Through to analyze the characteristics of the digital cartoon image, proposed a based on discrete cosine transform (DCT) and singular value decomposition (SVD) combined contourlet transform with zero watermarking algorithm, the algorithm first for image contourlet transform and then make the low-frequency subband DCT and SVD decomposition again after operation block, the largest singular value as the image feature extracting for zero watermarking structure. In order to improve the security of the algorithm, introducing into visual cryptography processing the watermark image, the zero watermarking features obtained and secret sharing for the logic operation is zero watermarking characteristics of the final value.. The simulation results show that the algorithm has strong robustness.

Existing Android applications for streaming video in real time are dependent on the codec, which composes the encoding function, and the version of Android operating system. Also, for streaming video in real time, most applications should be connected with a separate desktop PC. In this paper, we propose a new application, which overcomes these disadvantages and thus, streams video in real time. Specifically, to overcome these disadvantages, the proposed application uses the flash video file format, which is the common media file format supported by various versions of Android operating system. Through experiments, we show that it is possible for the proposed application to stream video in real time while using the existing video encoding methods.

As competition of Chinese securities industry becomes increasingly fierce, customer loyalty management has become the core elements of enterprise competition. Combining cross industry standard process for data mining (CRISP-DM) process model, flow visualization, this article adopts the mining method of large data to find the key factors to describe customer loyalty, and establishes customer loyalty prediction model of two dimensions, according to the model can effectively predict customer loyalty degree, support enterprise decision makers develop corresponding retention strategies.

Climate data, agricultural crop related data, etc. are being provided in real time as the national statistics by using the public data portal infrastructure. On this account, it is possible to process and provide customized data including those individually important elements and geographic information by personalizing them. This thesis designed and implemented the mobile integration system that can push customized data that combined the cultivation information as to the selected crops and the local climate forecast information based on the local based service of a smartphone.

Skin color detection is the process that classifies unknown colors into skin or non-skin classes. Skin color detection is preliminary step for facial or gesture analysis and can reduce search space for next higher level processing. In this paper, we show that proposed PCA-based color representation can give better performance than other frequently used color spaces such as XYZ and Luv. For skin detection, we use two classification models – histogram model and elliptical boundary model from skin and non-skin colors. The experimental results show the PCA-based color representation is more efficient than other color representation and the existing PCA-based representation based on the two classification models.

Construct an image retrieval method based on feature fusion of color features of RGB component and edge feature. The execution of this method is divided into the following steps: First, the original image is decomposed into three components images of RGB; Secondly, wavelet decomposition is carried out on each component image; Third, color and edge features are obtained on the low-pass components from each component image; Finally, these two features are combined to complete the image retrieval. Experimental results show that the proposed method has better accuracy than the retrieval method using single feature, and retrieval speed is relatively fast.

This paper discusses our work with the design and implementation of real-time face recognition for cloud robot. Face recognition has been widely used for surveillance and airport security. Robot systems which support face recognition are available too. As for robots, real-time feedbacks from sensors are always required. In order to recognize a face in real time, the images captured by camera have to be stored and then processed by face recognition algorithm. This processing procedure runs several times per second which is computationally intensive tasks for usual robot systems. Benefit from cloud computing technology, robot can offload computational tasks to cloud server which composes of several computational units (PCs) and is capable of executing distributed tasks. In our approach, real-time images will be captured and sent back to the cloud server. PCs in cloud side process the images in parallel. Face detection and recognition algorithm will be applied to each image and then the recognition results will be returned to the robot as the response. Our experimental results demonstrate that as a implementation of cloud robotics system, the proposed cloud robot is capable of doing real-time face recognition.

Today, as the supply of smart phones becomes active and a variety of functions of smart phones and services based on internet is utilized, a wide range of applications are being developed and deployed. Despite the rapid growth of smart phones and the application market, the attendance book which is currently being used is still in its primitive forms. Previously developed automatic attendance or absence check system requires additional equipments such as RFID reader and fingerprint reader, with higher deployment costs and has problems of management such as mechanical failure and dishonest attendance. In this paper, an automatic attendance check system based on smart phones utilizing the QR code for recognizing the code using the built-in camera function of smart phones and accessing the information included in the recognized code in real-time basis is proposed. The proposed system, by utilizing the built-in camera function of smart phones, can reduce the initial deployment costs because it does not require additional equipments when building the system and has partially solved the problems of dishonest attendance and improved the recognition rate.

Now a day’s Electronics plays a vital role in our daily life. It is a key in making this world a small and making everything happen at our finger tips, today’s most used and flexible is USB flash drives, which is used to store the data and data transfer between them related with computer, and we are not able to share files between two USB flash drives when user is away from computer. So to overcome this problem we are designing a complete blend of hardware and software With the help of this project we can not only transfer the data but also we can see the transfer of the particular file which we want to send by using LCD display. In our proposal we are transferring the data between two USB flash drives using two ways. One is Wireless Technology either Wi-Fi technology which we can send/receive File or Bluetooth Technology which we can send/receive data to USB flash drive with Bluetooth enabled devices like Mobiles, laptops etc.. Another is in-built USB slots to perform data transmission to other pen drives directly without the use of the computer system. The implementation of an operating system and a processor to the USB Flash drive are the main challenges in this proposed system.

Holography is the optimal and the final choice among several techniques for imaging and displaying the three-dimensional (3D) objects. In this paper, we propose a new denoising algorithm for digital hologram using Fresnel transform (FT). The proposed algorithm initially transforms a digital hologram into frequency-domain data through a FT, and separates the object from the background in the transformed image. Experimental results show that the technique to reduce noise in a natural 2D image such as a smoothing filter can reduce the noise to improve the image quality for a digital hologram. We expect that the conclusions of this paper will be a very useful technique for further work in the area of signal processing scheme for electronic holographic video service.

Sound travels in acoustics. There are numerous audio effects and sound plug-ins in the digital music industry based on the principles of acoustics. In addition, sound is minute fluctuations in pressure. These fluctuations are rapid and are caused by pressure and velocity of the air. Pressure pushes air molecules which move, and therefore have velocity. Human hearing is capable of responding to fluctuations within the range of 20 fluctuations per second to 20,000 fluctuations per second. Standing waves are generated wherever waves are reflected towards oncoming waves. They are generated when two loudspeakers produce exactly the same sound. The waves pass through each other, but the interference nodes and antinodes seem motionless. The Doppler Effect is observed as a phenomenon in which the pitch of a sound drops as the sound source passes by the observer. The direction and relative speed of the source and observer are important. This paper describes the concept of Doppler Effect plug-ins and explains how they work and how they can be utilized in music production.

Based on support vector machine (SVM) and analysis of optical flow, the paper presents a new method for recognizing player motions in broadcast sports video. The video often has problems like bad-quality image, non-static video cameras and low-resolution image of player. To address them, from the perspective of movement analysis and according to the spatial distribution features of optical flow field of tracked members, the grid classification method, a kind of local analysis idea, is used to extract descriptive characteristics of movement recognition. With different idea from the traditional flow analysis, the method regards optical flow vectors in the traced areas as a kind of spatial distribution information in the mode of mobility, improving robustness of optical flow features. With SVM as model classifier and the application of time-sequential voting strategy, the type of player actions is identified. Compared with existing recognition methods based on apparent characteristics, the proposed recognition technique which fetches and depends on motional descriptive feature achieves better recognition effects.

The purpose of this study is designing and implementing a context aware service model that can collect and process learners’ learning context in real time on mobile LMS. With the spread of so-called ubiquitous computing environment based on today’s wireless mobile communication technology, hardware technology, sensor network technology, etc., e-learning platform is also entering into a new phase. As the next-generation e-learning platform called e-learning 2.0 is evolving from PC-based to mobile device-based, research is continued for the application of new technologies such as cloud computing and big data processing. In the field of school education, on the other hand, there is an increasing need for developing experiential education contents to stimulate students’ creative synthetic thinking required in the age of creative economy. Thus, this study aimed to provide elaborative knowledge organized by teachers based on the learners’ context recognized using various types of context aware information collected from the learners’ mobile device. With the outcome of this study, we expect to enhance the timeliness of learning by providing contents of education organized in advance by teachers at any time and in any place, even without the learner’s request.