Frontier Computing

Impulse noise impacts an image, causing the quality of image to be deteriorated in image transmission or capture. In this paper, we propose a gain factor for the removal of the impulse noise. A 3 × 3 fixed-size local window is employed to analyze each extreme pixel (0 or 255 for an 8-bit gray-level image). All non-extreme pixels are sorted in an ascending order and are grouped according to the variation of pixel levels. If the pixel level between adjacent two sorted pixels varies seriously, a new group is created. Hence, the ratio and median value of each group are computed to determine the values of the gain factors. They are multiplied with the median value of each group to obtain the weighted value which is employed to replace the center pixel with an extreme value, enabling noise-corrupted pixels to be restored. Experimental results show that the proposed method can effectively remove salt-and-pepper noise from a corrupted image for various noise corruption densities (from 10% to 90%); meanwhile, the denoised image is freed from the blurred effect.

[Purpose/meaning] The massive information of car under the network environment is of special significance and value to analyze the characteristic of the freight vehicle driving. Through mining vehicle speed, acceleration and other driving data is advantageous to help research vehicle drivers driving behavior and to standard the driver’s driving behavior, and realize the intelligent management of the vehicle. [Methods/processes] In this paper, the part of freight vehicles operating within the bounds of Hebei province as the research object to obtain the characteristic parameters of vehicle driver’s driving behavior, and using data mining method based on factor analysis to convert the parameters as indicators of the K-Means clustering method to analyze driving behavior. [Results/conclusions] According to the analysis results, dangerous driving behavior in the process of freight vehicles on the road exists and there is a certain effect for road transport safety, but they are all very few. By studying the characteristic parameters of velocity and acceleration of the vehicle can better response freight vehicles in operation in the process of driving behavior, and it contributes to the intelligent vehicle management.

A novel iris segmentation technique based on active contour is proposed in this paper. Our approach includes two important issues, pupil segmentation and iris circle calculation. If the correct center position and radius of pupil can be found in the tested image, we can precisely segment the iris. The accuracy of our proposed method for ICE dataset is around 92%, and also reached high accuracy level of 79% for UBIRIS. Our results demonstrate that the proposed iris segmentation method can perform well with high accuracy for Iris segmentation in images.

Facial expression analysis for human-computer interaction, the driver’s state monitoring, or user emotion state monitoring has always been a very important issue in emotion recognition. People have a few emotions, and in different emotions, their facial expressions will have different characteristics. For example, if a person is happy, he/she may be with smiling face or smiling eyes, and if a person is angry or sad, he/she may frown. Once the system identified the user’s facial expression, there will be a corresponding action. This paper presents a framework for the use of raspberry functions to develop emotion recognition systems. We use Raspberry Pi’s camera module to detect the user’s facial expressions and use the Microsoft emotion recognition API to identify the user’s emotion. If the recognition result is angry or sad emotion, the system will broadcast gentle music and tune the light to be soft to smooth the person’s mood. Experiments results indicated the system can interact with users’ emotions.

Prostate-specific antigen (PSA) test is a commonly used clinical examination to evaluate the risk of prostate cancer, with the antibodies used normally as the recognition molecules for measuring PSA levels in serum. Alternatively, aptamers that are able to bind target molecules with high affinity and specificity similar to antibodies could be generated much easier and cheaper than the production of antibodies. In this study, we used computaional and experimental approaches to select truncated PSA-binding aptamers generated from the sequence information of PSA-binding aptamers previously reported in a literature. Genetic algorithm, the analysis of secondary structure, and molecular simulation were utilized in the in silico analysis. The top 4 ranked sequecnes in silico analysis were evaluated through their PSA-binding ability on the quartz crystal microbalance (QCM) biosensor. Finally, We identified a truncated aptamer obtained from the selection showing a nearly 3.5-fold higher measured signal than the response produced by the best known DNA sequence in the QCM measurement.

Conventional BRIR (Binaural Room Impulse Response) measurement uses only miniature microphones plugged on the entrance of the ear canal to pick up acoustic sound entering the subject’s ears. Though simple, this method does not consider the resonance of the ear canal, among other factors. In this paper, the measured BRIRs are refined by using bond conduction headphones to include these factors. Experimental results show that the refined BRIRs produce better distance localization than without the refinement.

To increase the coverage of states that a Web crawler can explore, most crawlers require their users to define a few crawling directives before the crawler starts crawling. A directive can, for example, assign a special input value to a particular input field so that the application performs a specific action and visits some special states. Note that, a crawler is supposedly capable of exploring an unknown application. But, given an unknown application, how could the user possibly prepare all the required directives in advance? This paper proposes a novel, interactive crawling strategy called GUIDE to address this issue. Instead of passively receiving all directives from the user at once, GUIDE actively asks the user for directives when Web pages containing input fields are found. GUIDE frees the user from knowing the target application in advance. In addition, GUIDE offers a hierarchical directive structure, allowing the user to define multiple values for the same input field. Our experiments indicate that GUIDE is easy to use and can also improve the overall Web page coverage.

In recent years, users have more exception for having the customizing goods that users can have it by themselves or be a good gift for their friends. They also want to have an easy system for this purpose and have more materials for produce goods or gifts. The famous technique is using laser machine to print photos on cups or steel plate. However the laser machine accept binary image only, and the photos shoot by users are colored. So converting color image to binary image is the first issue. But traditional binary image conversion mechanism will not be against the photos shoot by users because of the photos will have overexposure beauty shot effect by users. Therefore, in this paper we propose the binary image mechanism based on lightness histogram classification that can be against the effects and propose a practical application for realizing the scenario.

Most digital watermarking schemes using QR factorization suffer from being unable to fully utilize the elements of the R matrix. Thus, these schemes are neither secure nor robust to resist the cropping attack. Besides, these schemes do not deal with the allowable modification ranges of the R elements, thereby causing the damage to the hidden watermark. In this paper, we designed an algorithm to redundantly embed the four copies of the watermark bits to enhance the ability to against the cropping attack. During the embedding process, the property of sign wave is employed to ease the modification of real number coefficients. After the four copies of a watermark bit are extracted, they may be different due to possible attacks. Therefore, we designed a weighted strategy to resolve the watermark bit. The experimental results show that our scheme satisfy the requirements of imperceptibility and robustness. Particularly, our scheme has prominent robustness against cropping attacks.

Along with the development of multimedia system and its relevant technology, the researchers began to explore and attempt to use multimedia technology to assist teaching which is in order to improve the quality of education. The computer-based course is a public compulsory course which has significant implications for cultivating the students’ information literacy. However, there is a big difference between learners’ Computer level and other factors. Since all these unfavorable factors, the Teaching Practice of computer-based courses is not ideal and there is a variety of conflict. This research is trying to use the multimedia system and related technology to serve the daily teaching activities of the computer-based course after sorting out the related research results. This study built the flipped class mode based on SPOC. The study combined two rounds of teaching practice to improve the teaching mode and the study use the questionnaire survey and interview technique to prove the effectiveness of the model. Finally, the study suggested that the teaching model could be improved from three aspects—strengthening the construction of network teaching platform, enhancing the teachers teaching skills and focusing on the creative application of multimedia technology in teaching.

The Internet of Things, which is called the IOT for short, is the third wave of the information industry. Some simple superposition of the existing technology of privacy protection cannot meet the new demand for privacy protection of the Internet of Things in the third wave of the information industry. A timely overview of the development of the Internet of Things security and privacy technology, from numerous existing research results, will serve as a cornerstone of future research. Despite some literatures have given us a detailed summary of the relevant issues of the Internet of Things according to its perception or network, previous studies fail to discuss the terms of the overall framework of the Internet of Things or simply apply the security and privacy protection of the internet to the Internet of Things, which will inevitably affect the directions of future research more or less. This paper first analyzes the system structure and features of the Internet of Things, and then focuses on the security and privacy protection technology of the perceived environment by presenting a systematic overview of the current privacy protection technology concerning the Internet of Things, with a view to laying a foundation for future research.

Traffic sign classification is a significant issue in the intelligent vehicle domain, which helps vehicles to follow the traffic rules and ensure the safety. Feature selection and description are very important and difficult for classification. In this paper, a novel traffic sign classification method is proposed which is based on the Latent Dirichlet Allocation (LDA) model. Feature topics are modeled based on various traffic signs by the LDA automatically. And traffic signs captured onboard are classified according to the modeled features. The experiment results show the efficiency of our work.

In this secure smart grid R&D project issue in Korea, we aimed to develop the smart grid security technology which is the type of support of security vulnerability analysis through the convergence of IT technology. In addition, through securing such cutting edge technologies as smart meter monitoring, vulnerability analysis modeling, privacy mining, light-type mutual authentication service mechanism, and contents-based approach controls, we are going to pursue our research and development activities to make the core technology suitable for domestic environments, and also plan to have these technologies used by the industries by transferring the technology in a timely manner.

In the Internet age, clickbait is an effective method to attract people’s attention, which usually uses some ways to achieve, such as exaggerating, omitting the details or using punctuation exceedingly. In order to attract the readers to click on the link, some news aggregator sites or social media will choose to use the clickbait. If the clickbait applied to scientific articles, not only will affects the article quality, but also will affects the development of relevant subjects. Thus, the purpose of this research is to explore whether there is a clickbait in the data science articles of WeChat official accounts. This paper collects the relevant data by using the shenjianshou platform, and then uses some steps to analyze data, including cleaning data, doing word segmentation, extracting keywords and building a regression model. According to the adjusted r-square value in the regression model, the model can only explains the change of 3.17% page views, which means that the clickbait phenomenon is not prominent in the data science articles of WeChat official accounts. Finally, the regression analysis results are discussed from subject perspective, writer perspective and reader perspective.

Object measurement based on surveillance video-images is a critical task in the field of intelligent video surveillance. In this paper, we proposed an object measurement system based on surveillance video-images. Firstly, considering the scene of the urban video surveillance, a novel calibration tool is designed for such scene, which is portable, collapsible and operational. Next, based on the portable calibration tool, the intrinsic and extrinsic parameters of the surveillance camera can be estimated. Then, an object measurement system based on surveillance video-images is developed, which include the acquisition module of calibration data, the computation module of camera parameters, the selection module of measurement endpoint and the computation module of object. Finally, in the traffic checkpoint, experiments are carried out to verify the effectiveness of the object measurement system. The results show that as long as the surveillance camera is one-time calibrated based on the portable calibration tool, this system can measure the object on the surveillance video-images captured by the calibrated camera. What’s more, experimental results have shown that the object measurement system can be used to estimate the vehicle speed and measure the body height within the acceptable error range, and then have demonstrated the effectiveness of the object measurement system.

Human Action Recognition is becoming more and more important in many fields, especially in sports. However, conventional algorithm are almost camera-based methods, which make it cumbersome and expensive. As the wearable inertial sensor has developed a lot, in this paper, we present a novel human action classification algorithm using in basketball, based on a single inertial sensor, which is a application of multi-label classification. We performed experiment on real world datasets. The AUPRC, AUROC and confusion matrix of our results demonstrated that our novel basketball motion recognizer have a great performance.

A reliable and robust zero velocity points (ZVPs) detection approach is important to restrain the accumulative error in the pedestrian inertial navigation systems. A novel recursive zero-velocity detection (RZVD) approach for smartphone based pedestrian dead reckoning systems is proposed in this paper. It combined the adaptive threshold and context information of the vertical velocity to verify the correctness of ZVP detection and fixed the incorrect ZVPs recursively. The test results show that the performance of the proposed approach is better than original method. It indicates that the proposed approach is helpful to eliminate the serious estimation error caused by false detection of ZVPs.

Along with the widespread use of smartphones, Android has become one of the major platform for multimedia applications (apps). However, due to the fast evolution of Android operating system and the fragmentation of Android devices, it becomes important for an Android multimedia app to be tested on different devices to ensure that the app is compatible with and run well on any of the devices so as to provide consistent user experiences. This paper presents a cloud testing platform (CTP) that allows Android multimedia apps to be tested automatically against a scalable number of physical devices in parallel. Particularly, CTP provides four types of testing to ensure the compatibility of apps from different perspectives. Further, to facilitate identifying the bugs of apps, in addition to test results, CTP also provides the video, screenshots, and performance data corresponding to the tests. The case study shows that CTP can be effective in ensuring the compatibility of Android multimedia apps while saving test time and effort.

Internet of Things is an emerging paradigm to enable easy data collection and exchange among a wide variety of devices. When the scale of Internet of Things enlarges, the cloud computing system could be applied to mine these big data generated by Internet of Things. This paper proposes a task scheduling approach for time-critical data streaming applications on heterogeneous clouds. The proposed approach takes the tasks in critical stages into consideration, and re-schedules these tasks to appropriate resources to shorten their processing time. In general, selecting the time-critical task to give more resources may remove the execution bottleneck. A small-scale cloud system including 3 servers is built for experiments. The performance of the proposed approach is evaluated by three micro-benchmarks. Preliminary experimental results demonstrate the performance improvement of the critical task scheduling approach.

Affective computing (AC) is an emerging research direction to deal with the great challenge of creating emotional intelligence for a machine. Affective computing is a cross-disciplinary research knowledge that integrates recognition, interpretation, and simulation of human emotion into a system. This article describes the design of multimodal perception of affective computing system. Our multimodal physiological channels include facial expression recognition, heart rate monitoring, blood oxygen level (SpO2), skin conductance response (SCR), and electroencephalogram (EEG) signals for building our affective computing system. To solve the various data sampling problem, we developed a concurrent control integration mechanism to automatically average all the sensors’ data into the same data sampling rate (one record per second) and rearrange all the data into the same time. We believed the proposed system design is benefit for helping researchers in collecting and integrating experiment data in affective computing area.

In the inertial sensing unit based indoor positioning systems, the gyroscope drift is the primary source of heading error. To reduce this error, we proposed that the heading drift and the real heading change can be distinguished by the similarity of the gait pattern in the same movement model. It use the curve fitting method to find out the gait pattern in walking straightly. The Frechet distance is used to discriminate the gait of walking in turn and walking straightly. Experiments show that this method can recognize the walking in turn successfully with no mistake and the rate of mismatch walking in straight to walking in turn is less than 17.39%. Although there are some mistakes of match the walking in straight to walking in turn model, it will have few impact because heading drift is little in a short time. The result of test two shows that it get the best result compared with the other two methods when doing heading correction. It indicates that the proposal can promote the performance of heading correction and reduce the effect of sensor drift.

The human intelligence modeling by brain components simulation, such as neurons and their connections, is part of leading smart decision computing paradigms. In Health, artificial neural networks (ANN) have the capacity to adapt to uncertainty situations and learn even with inaccurate data. This paper presents the modeling and performance evaluation of an ANN-based technique, named multilayer perceptron (MLP), for gestational diabetes mellitus (GDM) prediction that is responsible for several severe complications and affects 3 to 7% of pregnancies worldwide. Results show that this approach reached a precision of 0.74, Recall 0.741, F-measure 0.741, and ROC area 0.779. These indicators show that this method is an excellent predictor of this disease. This contribution offers a computational intelligence (CI) tool capable of identifying risk cases during pregnancy and, thus, reduce possible sequels for both pregnant woman and fetus.

GLV method is an important research direction to accelerate the scalar multiplication on classes of elliptic curves with efficiently computable endomorphisms, which can reduce the number of doublings by using Straus-Shamir simultaneous multi-scalar multiplication technique. Researchers explore to generalize the method to higher dimension, and then evaluate the effect of accelerating the scalar multiplication. In this paper, we consider various multi-scalar multiplication algorithms, and analyze the computational cost of scalar multiplication under different dimensions to select the optimal multi-scalar multiplication algorithm and parameters. On this basis, the multi-scalar multiplication algorithm is applied to the GLV method, and the computational cost of scalar multiplication is analyzed. Higher dimension usually means fewer doublings, but more precomputation, there is a trade-off. The analysis results show that the limit of GLV method to accelerate the scalar multiplication is dimension 8, and the GLV method will lose its effect of speedup for higher dimension. In particular, dimension 3 or 4 may be the optimal choice for the case that resource constrained or the cost of endomorphism is large.

In order to improve the accuracy of the Shape Model Active method, we propose a new method to improve the accuracy of ASM (ASM) algorithm in face detection, and propose a new method to construct the local template. In the process of local localization, the paper uses form Closed-algorithm to segment the texture segmentation. Information is effectively improved the performance of the ASM method. The results show that the proposed algorithm can extract the feature points of most forward faces correctly. The proposed algorithm has a wide range of applications in image understanding of face tracking, recognition and facial expression analysis.

For extensive research on the connection and the phase relationships between climate indices and the precipitation time series in Korea, three wavelet image processing are applied to examine the relation between the climate indices (NAO, SOI, NOI, PDO, WP and NP) and precipitation in Korea. The continuous wavelet, cross-wavelet analysis and wavelet coherence is utilized to expand and present regions with common high significant frequency and phase features of the precipitation and these climate indices time series. It is found that the all wavelet frequency spectrum of these climate indices time series have some similar significant frequency features as the spectrum of precipitation time series in Korea have. There are significant variations of around 4–6 years of periodicities in all spectrum analysis. It is illustrated that there is strong underlying connection between precipitation variability and climate indices that implied by the cross wavelet and wavelet coherence analysis. The results clearly demonstrate that the climate indices have the influential consistent correlation relationship with the precipitation variation in Korea.

With the development of modern information technology, people have access to a wider range of information resources and texts, but there are few corpora for local law. Therefore, the computer aided English translation of local legal should meet the basic functional requirements of high efficiency and high quality. The CAT model of parallel corpora WAS constructed to make full use of the storage and computing ability of the software. In order to carry out the practical analysis of the translation method, a concrete evaluation model was constructed. A comparative study of the local laws and regulations in China was conducted in the aspects of the translation time and the quality of the translation by Google, CAT and parallel corpus based computer aided translation. According to the analysis, it can be seen that the computer aided English translation has the characteristics of high quality and high efficiency.

In this paper, an engineering algorithm that can overcome singularity of Euler Equation is adopted to adapt to particularity of tail-sitting aircraft. According to the practical significance and reference to the other algorithms, we expand its definitions and verify numerical calculation. The results demonstrate that the method is simple and quite effective. With the application of this method in attitude computation system for tail-sitting aircraft, satisfactory results are obtained. As a result, there is no requirement for rotatable parts and corresponding control units and the whole aircraft is in simple structure and small mass. Besides, the course of flight is simplified into fixed-wing aircraft maneuvering, thus it is easy to operate and especially suitable for the unmanned vehicles, for which there is no need to consider physical limitations of flight attendants. After redefining the value range of Euler angles, this method can be perfectly applied in attitude computation for tail-sitting aircraft which is also proved to be feasible through using experimental verification with universal application and reference value in engineering practice.

Due to the development of mobile devices and wireless networks, mobile applications are more and more popular in our daily lives. Since most mobile devices have the GPS positioning function, it makes the Location Based Services become an important research issue. The early applications of location based services focused on GIS-related services, such as mobile trajectory records, maps, navigation and other services. Pokemon GO is a well-known LBS related game. In addition, in the past few years, the usage of IT technology in teaching and learning environment has become more popular. Due to the popularity of mobile devices, the design of teaching applications can been implemented through various mobile devices. It provides students with ubiquitous learning area.Therefore, how to combine the LBS application and mobile learning system such that the student can learn anywhere with various learning activities become an important research issue. Furthermore, it will form a whole new learning method for students to learn.The main purples of this paper is to develop an interactive learning platform for LBS learning applications. Based on this platform, the teacher can design different kinds of content, such as teaching activities, AR applications, interactive games, multi-level tournament and so on, for different locations. The corresponding designed content will be triggered when the user or student pass some particular location. Moreover, the nearby users can help the other users to complete the learning activities which is truly a kind of collaboration learning.

The IoV integrates many important features of the IoT in order to provide numerous new services with advantages for humans and society, in which many sensor nodes are developed for the IoT, such as home appliances, including televisions and refrigerators, and health-related equipment, including heart rate sensors and foot pod meters. However, the IoV products are less developed at present. Hence, we should survey related sensor devices to implement on the vehicles in the future works and then integrate the current research on SDMS and MFSS to achieve vehicle-to-sensor interactions. In order to achieve the desired goal of zero traffic accidents, all information in the vehicle is collected through sensor devices, included in safety messages and forwarded to other vehicles or drivers. The future scheme should be extented to gather, share, process, compute, and release secure information to other humans, vehicles, devices and roadside services. The development and deployment of fully connected vehicles requires a combination of various IoT technologies. From the perspective of commercial value we can see that the future of economies of scale is to be expected, and from the perspective of extending technological development we can see that there are further contributions to be made.

Since the National Health Insurance was applied in 1995, the main goal was to offer better medical care to the general public, and to reduce the medical fare as well. Even with the general influenza, the general public would rather choose large medical center instead of local clinic since large hospital is recognized as better and more accurate. As the result, people would directly visit large operation for illness.

Since the first ambulance has been introduced to Taiwan, the members of fire departments have been dedicating themselves to emergency service training. Though all the emergency rescue measures have been improved, the causes and problems of Out-of Hospital-cardiac Arrest (OHCA) caused are still actively being explored by the experts from emergency medicine. The main causes of Out-of-Hospital Cardiac Arrest are classified into two categories: medical causes and surgical causes, the later ones mainly caused by car accidents or falls. There are some chronic disease such as heart disease (myoeardial infaracriton), chronic kidney disease, and diabetes are highly associated with Out-of-Hospital-Cardiac Arrest medical causes (Patel et al. 2014). However, in Taiwan, heart disease, hypertension, and diabetes mellitus are recognized as the three main medical causes for Out-of-Hospital-Cardiac Arrest.

The intention of this paper is to construct the constitutive senses of audition and vision in the design of learning interface for music sight-singing. Studies in the learning interface design for music staff notation, pitch recognition and sight-singing in music education are discussed. The aim is to establish current human-computer interface study results with the senses of audition and vision in music sight-singing acquisition and what is needed for further progress in this field of research. It is argued that the coming digital learning environment provides an effective and efficient field in music sight-singing research. It is also pointed out that the researches on the learning interface design with the combination of audition and vision benefit both children and adults. The paper highlights what learning interface design with the senses of hearing and vision in music sight-singing instruction can learn from research and where future research may provide further advancements.

The first of these seven steps is to eliminate multivalued attributes, composite attributes, and composite operations. The second is to eliminate the partial dependency and transitive dependency among the attributes, as well as shared operations. The third is to eliminate homogeneous operations among the classes to meet the requirements of inheritance and polymorphism. The fourth and fifth steps involve establishing classes for encapsulation, and the sixth and seventh steps eliminate multivalued dependency and operations with multivalued dependency. These steps create normalised concrete classes and control classes in the object-oriented class diagram, and they also maintain favourable consistency, completeness, and accuracy of the data. As such, they can provide effective guidelines and practical reference for system analysis and development.

Customers are difficult to find and sometimes even more difficult to keep. If a company doesn’t pay attention to customer relationship management, it will spend lots of money to acquire them and then let them sleep. The goal of this study was to develop a sleeping customer detection system that collected users’ social network information and shopping behavior, and then classified them into 3 categories (sleeping customer, napping customer, and general customer). We collected the user information from January 01, 2015 to December 31, 2016. Support vector machine based classification was used. In this study, the overall accuracy was 81.7%. The results can help companies to reactivate sleeping customers as soon as possible.

The explosive usage of Internet and telecommunications makes the remote server access mechanism necessary. Multi-server architecture mixes various services into one system. A user can get different services from different providers after registering on one registration center. To protect the information transmitted in the sessions, authentication is naturally considered. In the past decades, many multi-server authentication schemes have been presented. But unfortunately, various sorts of attacks are presented to prove the past schemes insecure. To prevent the common attacks, we give a new two-factor authentication scheme for multi-server systems. Through the analysis of security properties and performance, all can see that the proposed scheme is against the common attacks, such as off-line guessing attacks, tracking attacks, etc. And it is suitable for application in real circumstance. ...

The widespread use of radio frequency identification (RFID) in IoTs makes authentication of RFID systems be widely concerned. In existing encryption schemes (e.g., Hash function) in electronic products, secure chip is hard to be used in high performance RFID system due to high computation complexity and cost. In this work, we consider problems in these performances and propose a PUF-GIMAP protocol by combining GIMAP protocol and physically unclonable functions (PUFs). The response of PUF is added into protocol. The mutual authentication between label and reader is realized by transmitting information such as secret key and dynamical pseudonym, which greatly ensures data security in transmission. After authentication, the reserved data is updated in time. The protocol analysis shows that the proposed scheme has the advantages of high security and high reliability.

Collaborative filtering algorithms help people make choices based on the opinions of other people. User-based and item-based collaborative filtering algorithms predict new ratings by using ratings of similar users or items. Similarity calculation is the key step in the algorithms. This paper compares the prediction quality of four commonly used similarity measures on different datasets. Experimental results show that Adjusted Cosine similarity consistently achieves best prediction accuracy.

Long Term Evolution in unlicensed band extends the benefits of Long Term Evolution and Long Term Evolution - Advanced to deploy in 5 GHz unlicensed spectrum, enabling mobile operators to offload data traffic onto unlicensed frequencies. The License Assisted Access with Long Term Evolution allows co-existence with Wi-Fi through carrier aggregation. The recently evolved intelligent technology viz. Cognitive radio which supports the efficient spectrum utilization is applied in the proposed system model to detect the white spaces in 5 GHz band to accomplish Listen-Before-Talk regulatory requirement of radio communication in Long Term Evolution - unlicensed band. Another major goal of Long Term Evolution - unlicensed to co-existence along with Wi-Fi/Internet of Things users in a non-interference style is also accomplished by the use of Cognitive radio. Simulation results demonstrate their coexistence along with effectiveness of resource allocation in a varying 5 GHz compatible 802.11 wireless local area networks environment.

Spatial data mining refers to the extraction of Geo Spatial Knowledge, maintaining their spatial relationships, along with other interesting patterns not explicitly stored in spatial datasets. The overall objective of this research work is to apply GIS based data mining classification modeling techniques to assess the spatial landslide risk analysis in Nilgris district, Tamilnadu, India. Landslide is one of the most important hazards that affect different parts of India in the every year. Landslides cover broad range impact on the people of the affected area in terms of the devastation caused to material and human resources. Landslide is generated by various factors such as rainfall, soil, slope, land use and land covers, geology, etc. Each landslide factor has a different level of values. The ranking of values and assignment of weight to the landslide factor gives good classification of landslide risk level. Data science and soft computing play major role in landslide risk analysis. The rank and weight are assigned to the landslide factor and its different levels using classification data science techniques. In this paper, we proposed a new model with integration of rough set and Bayesian classification called Hybrid Intelligent Bayesian Model (HIBM) to analyze the possibilities of various landslide risk level. The proposed model is compared with real-time data, and performance is validated with other data science models.

Applications processing massive amount of data demand superior time-performance and data-storage capabilities. Many organizations are exploring Cloud Computing to manage such applications because of scalability and convenience of access from different geographic locations, whereas data security and privacy are few of the major concerns preventing them from fully embracing it. Data security while maintaining time-performance becomes an important consideration for designing data placement strategies. We first present the reader with a quick survey on the recent approaches and solutions in data placement oriented problems that address security concern and then characterize the performance for graph based algorithms that are practically realizable. We evaluated the performance of conventional strategies such as, Random, T-coloring and revisited these algorithms in the view of providing maximum security. We refer to our strategy as Data Security Preferential (DSP) data placement strategy and evaluated via rigorous performance evaluation tests to identify which strategy will best suit the requirements of the user on three state-of-the-art hierarchical cloud platforms viz., FatTree, ThreeTier and DCell and non-hierarchical cloud platforms.

Over the past few years, MOOCs have trigged an education revolution. Clickstream data of user were recorded by MOOCs platform, providing valuable insights about the way user interact with the MOOCs. In this paper, we study user navigation patterns in MOOCs. We propose a metric to measure the similarity between user session-level navigation path, and build an unsupervised clustering model to capture user navigation patterns in MOOCs. Based on the user behavior clustering result, we further explore engagement of each navigation pattern from the perspective of dropout. To measure the effectiveness of our model, we conduct experiment on real world dataset with five weeks of interaction logs of 3,914 users. Through our analysis, clustering model proposed effectively identifies 13 types of user navigation patterns, which help us understand user behavior in MOOCs.

The popularity of Internet of Things (IoT) triggers the rapid development of various IoT platforms, standards, and protocols. However, the differences and heterogeneity among IoT platforms, standards and protocols have become the critical difficulty to interconnect a mass of IoT devices. According to our observation, most IoT standards and platforms utilize similar light-weighted network protocols, i.e., CoAP and MQTT. Thus, in this paper, one bidirectional data exchange scheme based on software-defined network (SDN) is proposed to let two IoT devices with different IoT protocols, i.e., CoAP and MQTT, be able to communicate each other more straightforward. The experimental results show that the proposed scheme costs more time to establish one traffic flow. Once the traffic flow is established, the transmission performance is a little lower than the existing mechanism, i.e., Ponte.

The conventional dual-mode halftoning methods achieve high quality halftone patterns by easing smooth artifact and preserving image details. However, the boundaries between low- and high-frequency image regions still present undesired texture in some particular cases, which significantly degrades the visual quality. In this work, a multi-mode halftoning method is proposed to deal with the object map artifact and boundary artifact simultaneously. In addition, the new absorptance-frequency stacking constraint is also employed to solve the noisy textures of the halftone outputs. As documented in the experimental results, high quality halftone outputs can be obtained, proving that the proposed method can be a competitive candidate for electrophotographic printers.

For narrow-band powerline communications (NB-PLC), the dominant impulsive noise (IN) is cyclostationary noise and periodic noise. However, some existing works still apply the Middle class-A (MCA) IN model in an NB-PLC system. By applying the cyclic spectral analysis technique, we show that the MCA IN model cannot capture the nature of cyclostationary noise and periodic noise.

Fuzzy clustering algorithm as a more successful segmentation algorithm has been successfully applied in the medical field. However, the traditional Fuzzy C-means clustering (FCM) algorithm has the disadvantages of time-consuming, noise-sensitive and non-consideration of neighborhood information in the segmented brain MRI (MRI), and proposes a corresponding solution to these problems. Firstly, Canny operator and morphological processing method is employed to extract the brain MRI of image contour information, reducing the image background brings a series of calculation problem. Secondly, before the FCM image segmentation, the adaptive adjustment of the weight coefficient in the neighborhood is realized by introducing the gradient information to achieve the purpose of eliminating the noise and reducing the initial value of the image objective function. With the experiment proved above, the robustness of the algorithm is improved and effectively shorten the calculation time in the case of constant accuracy.

In recent years, with safe city and the construction and development of the intelligent City project, has become a public security authority security control video surveillance systems, combat crime, and effective means to prevent emergency incidents. With the rapid development of network communication technology and mobile intelligent terminals (such as smart phones, tablets, etc.) the rapid proliferation of smart terminals have been carrying video surveillance, audio, speed sensors and sensing devices. Video equipment parts in high-end smart terminal can carry over parts of the lower end of video surveillance equipment. Intelligent terminal mass popularity makes building a people-centric sensing and computing networks possible in order to achieve the perfect fusion of the physical world and the digital world. Effective integration of different information spaces of information can enhance public safety and effective sensing and detection. According to public safety incidents of multi-source information fusion is proposed surge of data model, and the model is defined. And witness in one system by the model was verified. The unity of witness systems have been developed in several Beijing bus station and train station.

Video surveillance technology has become an indispensable method of public security work. As the number of monitored devices is increasing, resulting in a large amount of video data, resulting in great strength brought by police officers. From the early days of license plate recognition to the nearest human face than, vehicles feature recognition technology, are typical applications of intelligent video analysis, has produced positive results in public security work. With the progress of artificial intelligence techniques, especially the depth of learning in the field of video analysis to continuously refresh the recognition task, can see into the future will have more intelligent video analysis technology penetration into the field of public safety. However, intelligent analysis techniques in the field of public security still faces great challenges and bottlenecks.

This paper proposed a new classification model of information systems. Immune Network algorithm based on classification model is proposed. It can be positive from a large number of security information to access and extract useful information. It also can analyze the consequences of the threat information and effective measures in a timely manner. Classification protection threat information can be shared in a timely manner. Emergency response, announcements and alerts can be completed in a timely manner.

To improve the actual application ability of UAVs for public security, and promote the development of the UAV industry in the police market, the paper analyses the demand and application status of multi axis rotor UAV (unmanned aerial vehicle) for different job categories of public security. It is pointed out that the multi rotor UAV exists the problems in application and the development direction of police multi axis rotor UAV.

The integration system of face and ID card is used to detect person in the important sections such as government place and road. In order to do the effective security of the important region, in this paper, we propose the integration system of face and ID card. The basic framework, modules, and case system are given.