Earticle

To reduce the impacts of harmonics and noises on frequency measurements, a novel method of frequency tracking is presented based on improved EKF algorithm and wavelet transform. Firstly, the power signal was preprocessed to extract the fundamental wave by wavelet filtering. Then, an improved Extended Kalman Filter was introduced to achieve fast and accurate frequency tracking. By adding a robust factor and modifying the covariance matrix in real time, it enhanced the stability for detecting distorted signals. The proposed tracking algorithm can effectively solve the problems of traditional Extended Kalman Filter such as lack of adaptability, instability and divergence. And it improved the tracking speed and accuracy. Finally, Detailed simulations on several typical models of power signals have been carried out and the Simulation results shown that the new algorithm increases the speed, convergence and accuracy. It effectively suppresses the impacts of harmonics and noises on the frequency tracking results.

In this paper, we propose an effective error and erasure decoding scheme with multi-level erasure-decision threshold for Reed-Solomon (RS) codes in frequency-hopped multiple-access communications with M-ary FSK signaling, and analyze its performance.RS coding is used to correct erasures and errors caused by other-user interference and thermal noise. The proposed decoder decides whether to erase the received RS code symbol based on the erasure decision threshold. The decision formulas for the optimal erasure decision threshold are derived in such a way that packet error probability can be minimized. Numerical results show that the proposed multi-level threshold control decoding scheme yields a higher throughput over a conventional decoding cases.

The introduction of digital video technology into basketball trainings can greatly increase the efficiency of training; this thesis proposes a behavior prediction method based on the vision judgment of ball carrier to the computer technology used for the behavior prediction of ball carrier. The method analyzes the characteristics in basketball matches like complexity background, fast speed, and low resolution and so on, which integrate the multiple effective head vision characteristics, form the Riemannian manifold, and then map them into tangent space. Well-trained classifier is used to recognize the head pose of the ball carrier to ensure the visual range. This thesis makes prediction experiments for shooting, passing and dribbling of the ball carrier, and the simulation experiments shows that the proposed method is correct and efficient.

There are several methods proposed for detection and segmentation of object effectively. However, this algorithm struggles to detect an object with a lot of noise and shadows. Therefore, it is difficult to segment the accurate region and information of the object using background modeling only. To solve these problems, this paper introduced a more effective method of object segmentation based on interest point detection and description, which are core SURF theories. As a result, the feature extracted from the region of interest (ROI) was detecTable even with changes in scale, noise, and illumination. We then made the adaptive search window by this feature for ROI. After object detection, we applied the SVM to train the information of the feature from the detected object, and a classifier was built to estimate whether a result was a pedestrian. Therefore, if the result is a pedestrian, we would employ the Camshift algorithm to track the motion of this pedestrian. The experimental results showed the effectiveness of our method through comparison with others.

A new framework based on multi- modal semantic clues and HCRF (Hidden Conditional Random Field) for soccer wonderful event detection. Through analysis of the structural semantics of the wonderful event videos, define nine kinds of multi- modal semantic clues to accurately describe the included semantic information of the wonderful events. After splitting the video clips into several physical shots, extract the multi- modal semantic clues from the key frame of each shot to get the feature vector of the current shots, and compose the observed sequence of the feature vectors of all shots in the test video clips. Using the above observed sequence as HCRF model input in the case of small-scale training samples, establish wonderful event detection HCRF model effectively. Experiments show that the average recall of this paper reaches 95.3%, the average precision rate reaches 96%, the performance of this paper is obviously superior to the contrast method.

In order to improve the recognition performance of the tradition KNN, we combined the contributions of each feature parameter for different emotions with the tradition KNN and proposed a Fuzzy KNN algorithm for speech emotion recognition. Four kinds of emotions’ recognition experiment has shown that the Fuzzy KNN we proposed not only keeps the tradition KNN’s advantages of fast recognition and easy realization, but also improves the recognition performance.

DNA barcoding involve rapidly sequencing the standardized region from the genome for species identification and this homologous sequence information plays an important role to resolve the phylogenetic issues. However in case of complex groups, incorporation of insertion and deletion (Indel) informative sites in DNA barcode sequences are becoming more important due to low substitution rate. Indels are largely ignored in phylogenetics analysis and removed from sequence alignment assigning them a missing data. We review the current trends on mining the indels analysis, focusing especially on the topics of rapidly evolving indel containing loci and methods of indel treatment for phylogenetic relationship. Among the indel coding methods, Simple indel coding is easy to implement in indels contain sequences. But, this method does not utilized the all the available information and Complex Indel Coding rules are very difficult to translate into a clearly formulate algorithm for determining the values. But this coding method suffers from internal inconsistence when a long indel has a number of shorter subset indel and triangular in equality exists. However, SIDIER software package infer evolutionary relationships based on both the indels and substitutions. SIDIER is promising software for intra and inter specific calculation in DNA barcode studies as well as to infer phylogenetic relationships.

In this article, a multi-tone signal generated by micro-speaker is adopted as the acoustic stimulation, and two microphones are used to collect the DPOAE data from human ear and background noise respectively. Otoacoustic emission is modeled based on Volterra kernel. The feature of human ear’s DPOAE feature model is extracted intelligently by improved stimulation annealing genetic mixed algorithm. In order to apply this model feature to identification, its feasibility is verified by BP neural network. This provides a new biometric method for identity authentication.

In this paper, another strategy for immediate features based image recovery is proposed. Image database is developed with low level texture features got from Gray Level Co- Occurrence Matrix (GLCM) and measurable techniques for Tamura. Semantic level inquiries from the user mapped to the low level peculiarities at recovery time to recover the required images. Images with more than one moderate features can be recovered by utilizing intersection of images recovered by each of the queried feature. Artificial Neural Network (ANN) is utilized as a part of the following steps in the wake of accepting user inputs. In spite of the fact that semantics are utilized as search key as a part of the beginning steps, low level features are utilized as a part of the ANN based searching in later steps. Back propagation Algorithm is utilized as a part of learning step. This ANN based relevance feedback technique enhances accuracy of immediate feature based image retrieval method. Decision tree (DT) can likewise be connected in relevance feedback stage. Decision tree is framed in training stage and images will be tested by of the decision tree. Relation storing ontology related information is utilized as a part of every phase of retrieval procedure to evacuate ambiguities identified with synonyms and hypernym-homonym sets.

Crowd motion analysis in public places is an important research subject in the monitoring field. This paper proposes an approach for detecting abnormal crowd motion using Hidden Conditional Random Fields Model (HCRF). This approach derives variations of motion patterns from direction distribution of the crowd motion obtained by the optical flow and these variations are encoded with HCRF to allow for the detection of abnormal crowd motion. Modeling the temporal neighborhood relations in a video sequence based on HCRF can incorporate hidden states and label the video depending on long range observations. The experimental results show that this proposed algorithm can achieve better results than HMM and CRF.

With the increasing threats on multimedia information sharing, degradation of security is one of the most important concerns. Visual Cryptography is an emerging field for improving the security level by encoding the secret video using encryption scheme and performs computational free decoding process. The secret sharing through visual cryptographic techniques for multiple stack users is not sufficient in providing high percentage result on video sharing domain. Video sharing did not develop source key properties for individual frames and as a result the quality estimation strategy on video frame was not achieved. Moreover, the target rate constraint for video quality under the encryption-decryption was not achieved. This encryption-decryption target key rate declines on arranging multiple sequences of frames from multiple stack users. To improve the quality of visual secret video sharing scheme, a Discrete Cosine Transform Chaos Cryptographic (DCTCC) mechanism is proposed in this paper. The main objective of DCTCC is to enhance the security level of different level video frame using the encryption key properties. Initially, sender takes the video of differing frame length and compresses the video on image using the Discrete Cosine Transformer. The Discrete Cosine Transformer based visual secret sharing of video in the image takes the luminance values and provides high lossless compression. Second, the embedded video frame is encrypted to perform the secret sharing (i.e., visual cryptographic) using the chaos method. The Chaos method uses the enhanced RSA key properties to improve the video quality estimation strategy. Chaos method in DCTCC works effectively with non-linear dynamic set of video frames from users and also with the discrete time dynamical video cryptographic system. On the decryption side, the chaos method is applied in the reverse procedural order to fetch the video quality of video frames from original image. Experiment is conducted on factors such as mean square error rate, and secret video sharing efficacy level and encryption time.

This paper mainly adopts the literature research method, which analyzes the application progress and development overview of the eye tracking technology, as well as the technical features of the eye tracking devices. Combined with the research needs in landscape visual perception, eye tracking technology have application values in the following aspects: landscape attractiveness analysis, safety design of landscape environment, usability design of landscape facilities, landscape perception in motion, quality improvement of landscape information interface, cognitive analysis of physiological vulnerable groups and quality improvement of landscape education. Eye- tracking technology should be used reasonably in the study so as to exploit its advantages to the full and correct its deficiencies. The development of hardware technology and software technology of eye-tracking offers a new perspective and study platform for the study on landscape visual perception, which contributes to improving the scientificity of the study on landscape visual perception.

In the latest stock systems, a variety of technologies have been introduced to minimize low latency and accelerate data processing, and a massive amount of stock information has become interoperable. Only with the support of expensive network equipment, however, it is limited to increase the speed. Besides, it is not very effective to induce such equipment. Also, such stock information is hard to be shared in cloud environment for many reasons due to the heterogeneity of metadata. This makes it harder to optimize the proper time of data among stock trading systems as well. This study is going to suggest a stock trading system using X-DBaaS (XMDR-Database as a Service) in cloud environment. The suggested system applies SPT (Safe Proper Time) which can satisfy the proper time of data transmission and also minimize the CPI (Cycle Per Instruction) in order to reduce the CPU overhead based on the understanding of kernels. Moreover, XMDR-DAI is utilized to solve the heterogeneity issues among metadata and schemata taking place in the interoperation of information among stock systems.

This paper proposes a conceptual framework for a safety driving system that can pre- vent secondary collision due to car accidents ahead on a road. The proposed framework uses sensors (shock and velocity), GPS information, video information using a Drone system and multicast transmission in connected-vehicle communication network includ- ing V2V and V2I communication network. The proposed safety driving mechanism con- sists of five parts, the 1st part is a sensor part that detects shock and velocity of crashed vehicle due to collision accidents, and the 2nd one is a GPS system that detects the posi-tion of accident. The 3rd one is a Drone system that captures the situation video of the accidents and transmits it, and the 4th part is a V2X communication network including multicast transmission. The last part is a safety driving server system that manages all of the information for the safety driving and controls all of devices and communication mechanism in connected-vehicle communication network.

At present, the 3D object model technique is being applied in the simulation, virtual reality, 3D film industry and culture protection, and many other fields which can be widely applied. However, the existing 3D reconstruction technology is based on the complex and often expensive sensors. Therefore, we conduct research on computer graphics theory and mathematical modelling based 3D reconstruction methodology with Kinect. We firstly introduce the basic concepts and application of Kinect system which serves as the preliminary of our research. Then, we conduct mathematical modelling for our proposed system with theoretical analysis. Later, experiment and numerical simulation is conducted for verifying the robustness and effectiveness of our proposed approach. The visualized and numerical simulation result indicates that our method outperforms and holds better accuracy compared with other popular adopted algorithms. In the future, we plan to combine more optimization techniques to polish and modify our approach.

In this paper, a new watermark algorithm that based on wavelet packet transform and hybrid filter is proposed for watermark of digital images, which include many high frequency components. The proposed watermark algorithm is used in the high frequency component of image and applies watermark to the overall subband that include the lowest frequency band. And the watermark is embedded on input image and hybrid filtering concept has been adopted to remove the noise that was added to the original image in watermark embedding stage. The quality of the watermarked image has been analyzed with PSNR. From the simulation results, the proposed algorithm shows better invisibility and robustness performance compare with conventional watermark methods. Especially, it demonstrates better robustness for high image compression in the remote sensing images application.

The impermeability layer has decreased the surface storage ability, groundwater level and the time of concentration, and increased the peak flow of storm water runoff and urban floods. Due to the fact that developing the drainage functions in urbanized areas requires a considerable amount of burden in public expenses, which in reality makes it difficult to proceed with, it is necessary to apply the Low Impact Development (LID) concept and transform the drainage-oriented water management to water use and environment- friendly management. As the direction of recent design methods has been shifted to BIM (Building Information Modeling) and GIS (Geographic Information System), it is expected that the LID design will also be shifted to the same direction. Therefore, the related analytical solution is needed, and by using Autodesk® Storm and Sanitary Analysis (SSA) it will enable an interpretation of the city outflow and analysis by making connections with BIM and GIS, and this study’s analysis included making a connection with the runoff model of another city SWMM. This study helped to minimize data loss and rework the process of constructing the network data using the BIM design method and the public GIS followed by the analytical work, and to quickly construct the current network writing by utilizing the public GIS.

As the use of smart devices is being activated in the education sector the introduction of smart education using smart devices is coming into full swing. For universities in 2012 the Ministry of Education, Science and Technology is establishing a smart campus establishment plan and promoting it so that students can use electronic devices including smart devices to learn anywhere and anytime. For it is the study designed and implemented a convenient and practical attendance management system. The suggested system automatically uses the Bluetooth 4.0 communication of the students’ smart phones when the student enters the lecture hall of the course to check the location of the Beacon and automatically acknowledges attendance if it is valid location. It is characterized by the fact that there is no action to be taken on the part of the student or professor for checking attendance. Also the system recognizes the student and professor modes using the smart phones’ number and registers students and professors. Thus it can be used more conveniently because there are no other actions required.

This paper has clear understanding on features of experiencing laboratories which is based on ERP on account of the ERP simulation with sand table experiment. Then we established strategic position to applied personnel majoring in vocal music training on account of ERP, so as to establish the practicing platform of experiments, practical training as well as practice for students majoring in vocal music training, to improve the application ability of students. We put forward practicing discovery on model of applied personnel majoring in vocal music training.

This paper proposes an extended diagonal mesh (XDMesh) topology for network-on- chip (NoC) architectures to reduce latency and energy consumption for fast and low- power communication among remote nodes by including diagonal links in the network. In addition, we compare the performance of the proposed XDMesh with conventional state-of-the art topologies, including mesh, extended-butterfly fat tree (EFTI), and diametrical mesh, in terms of throughput, latency, energy consumption, and area overhead. Experimental results indicate that XDMesh outperforms the conventional topologies in terms of throughput and latency by varying the number of virtual channels and injection rate. Moreover, XDMesh achieves 46.28%, 35.29% and 19.37% lower energy consumption than EFTI, mesh, and diametrical mesh topologies, respectively, and 9.29%, 31.28%, and 15.23% lower silicon area, respectively.

This paper proposes a human action recognition algorithm which can be efficiently applied to a real-time intelligent surveillance system. This method models the background, obtains the difference image between input image and the modeled background image, extracts the silhouette of human object from input image, and recognizes human action by using coordinates of object, directions of that and accumulated moving regions of that. The human actions recognized in this study amount to a total of 8 type of actions, which include walking, raising an arm (left, right), raising a leg (left, right), sitting and crouching. The proposed method has been experimented for 8 different movements using 4 people using video input of a webcam and it has shown good results in terms of recognizing human action.

Since social activities of the modern people have increased continuously in recent year, many people frequently communicate with each other online and/or offline, exchanging a variety of information. Thanks to the development of smartphones and a wide range of mobile applications helping social activities of people, people can communicate more easily beyond their time and space. But, there are not many mobile applications specially designed for effective communication and collaboration among co-located people, most of the existing applications only provide limited functions. In this paper, we present an android mobile application DIO specially designed for co-located people to collaborate in a rich way. DIO enables users to share and exchange their profiles, favorite photos, musics, videos and documents according to one of the three open levels - public, business and community level. In addition, through DIO, co-located people can chat, vote, and survey. DIO is implemented on the top of AllJoyn platform - an open source IoT platform developed by the AllSeen Alliance. So, it can be easily done to extend DIO to communicate with other smart things running on the same platform.

To deal with the issues like existing common data sparseness in weibo social network and the phenomena of cold start, this paper puts forward a two-stage clustering based on the recommendation algorithm GCCR. The algorithm firstly selects users’ focused nodes which have higher number, so as to extract a dense subset of sparse data, and by using the method of graph paper, similar concerned interested core clustering is formed to this dense subset. Then, it is extracted that weibo content features of seed clustering and the whole data set other users. Then the entire user group is clustered based on content similarity. Finally the clustering results are used in subject recommendation. Through clustering the two phases of dense data subset and the whole data set, the clustering effect of extreme sparse data sets are improved. At the same time, because of fuzziness of graph clustering, this thesis retains a certain diversity in the process of user interest clustering, so as to avoid convergence too fast when cold start. This method is verified through the real social network data, and the experimental results show that this algorithm can effectively solve the problems such as data sparseness and cold start phenomenon.

Bio-toxicity using luminescence of PFOA and PFOS was assessed by photomultiplier tube. The assessment was evaluated based on EC50 and transient profile in time with varying concentration of 0.0~250 mg/L in 6 steps. As a result, the toxicity was found out to be 150.3 and 237.7 in EC50 for PFOA and PFOS, respectively. Besides, the profiles depicted that PFOA is more toxic than PFOS. As a result of applying 1CFOK model where actual toxic kinetics were applied, high kinetics parameters showing biological concentration or damage were drawn in case of PFOA. Also, it is determined that PFOA has a significant effect of chronic toxicity on the residual concentration. PFOS does not show a higher toxicity than PFOA, but it shows a fatal toxicity at near 250 mg/L and it can be concentrated, so measures for this substance are necessary

In a growing web services environment, the usage, design and composition methods of web services require proper and reliable information about web services quality. Quality- of- Service (QoS) is usually referred to the non-functional characteristics of web services which can be used as an important role to differentiate between different web services. In order to select the best web services among those with similar functionality, it is important for the web requester to have a mechanism to rank them and also from the service providers side to ensure that their web services are according to the agreed QoS. Currently there is no standard manner to define the QoS characteristics of web services. Many studies have shown different categorization of QoS which led to the issue of semantic interoperability of QoS. One way to understand the QoS, is to identify all its possible requirements for web services. This paper is an attempt to explore some of the QoS issues which are essential for web services. It first proposes a QoS framework which suggests considering QoS for web services along four different views. Each view is representing a particular relevant facet of QoS models. The main contribution of this paper is to show that there is a big quantity and variety of interpretations of various service quality concepts and models and gives the reasons why there is no holistic QoS modeling approach.

Location-awareness is important for healthcare, and can be applied to the various consumer applications. The received signal strength (RSS) based localization technique has advantages of needing no additional hardware and simple to be implemented in- building applications. Received signal strength indication fingerprinting (RSSIFP) is an indoor localization technique. However, the RSS is affected by radio signals’ reflections, shadowing, and fading. To solve this problem, an effective indoor localization method of revised RSSIFP is proposed to reduce the deviation during indoor RSSIFP localization. The proposed algorithm uses the RSSIFP based on the position probability grid. Before position, the RSSIFP data are revised according to anchor node signal and time tag. The K-nearest neighbor (KNN) and weighted centre localization method is adopted in position prediction. A test-bed only including common consumer electronic equipments such as wireless access point (AP), Zigbee node and smart cell-phone is deployed. Performance results show that the proposed algorithm outperforms other algorithms in the healthcare environments.

Recent advancement in mobile miniature information equipment has led to a rapid growth in the usage of storage device based on flash memory. Also, Flash memory is being used as the next generation’s storing device and shows rapid growth in massive storage device such as the SSD due to the distinguishable features such as low power consumption, strong durability, approach velocity, non-vibration, non-noise etc. These Flash drives uses B+-Tree index universally. However, B+-tree index has a flaw which downgrades the performance due to the repetitive writing request caused by the insertion of node. This dissertation suggested the F-ISLD (F-Tree Index Segment Log Directory) method which uses the F-Tree index. To prove the method’s superiority the suggested method and BISLD (B+-Tree Index Segment Log Directory) was compared. According to the assessment, the overall performance was increased by 29%.

Today, a “co-creation” culture for creating shared values plays an important role in our society. According to this cultural trend, co-creation is needed in developing smart objects. However, existing IoT (Internet of Things) platforms do not support a co- creation environment for users. Thus, this research aims to develop a co-creation environment that can allow users to create with others by sharing different ideas in the process of developing and testing smart objects. In this paper, we report our experience in the development of a co-creation environment for an open IoT platform. This paper contributes to the provision of developmental approaches to developers and researchers who develop IoT platforms that can support co-creation environments.

The purpose of this study is to analyze how the consumers perceive information disclosure systems and how they are utilizing the information provided. To achieve the study purposes, this study analyzed the user database of the College Alimi website, conducted recognition and satisfaction surveys on information disclosure with the college students and high school teachers. In addition, interviews with college professors were conducted as well. Based on the study results, how information users such as students, teachers, and professors perceive and utilize the information and suggestions were made for the better use of the system.