期刊：
ICDTE '17: Proceedings of the 1st International Conference on Digital Technology in Education,2017年Part F131203:Pages 79–83
作者机构：
[Chen, Yun; Zhang, Hao; Huang, Tao] National Engineering Research Center for E-Learning, Central China Normal University(CCNU), Wuhan, 430079, China
会议名称：
978-1-4503-5283-3
会议时间：
August, 2017
会议地点：
Taipei Taiwan
会议论文集名称：
ICDTE '17: Proceedings of the International Conference on Digital Technology in Education
摘要：
The construction of learning object repository is the key infrastructure of learning object organization under the E-learning environment. At present, the learning object repository is redundant construction, scattered and confused, lack the creation of tacit knowledge into explicit knowledge, is not conducive to knowledge sharing. In view of the above problems, this paper proposed a learning object repository fusion method, designed and constructed learning object knowledge units and fusion rule base, and applied fuzzy set theory to knowledge unit similarity fusion. Finally, we verified the effectiveness and feasibility of the method through the experiments, and get the more reliable results than single knowledge source detection, reduced the uncertainty of the fusion results. It has a certain reference value for improving the quality of learning object repository and collaborative work among repositories.
作者机构：
[陈靓影; 苏志铭] National Engineering Research Centre for E-Learning, Central China Normal University, Wuhan, 430079, China;[陈靓影] Collaborative &, Innovative Centre for Educational Technology (CICET), Central China Normal University, Wuhan, 430079, China
通讯机构：
National Engineering Research Centre for E-Learning, Central China Normal University, Wuhan, China
关键词：
face tracking;face detection;face validation;low-cost;real-time;onboard computer
摘要：
It is important to track people's face efficiently and accurately in many Intelligent Transportation Systems (ITSs) and Safety Driving Assistant Systems (SDASs). This paper presents a high-performance and low-cost real-time face tracking system, which runs on general onboard computer with very low CPU consumption. The proposed face tracking system is composed of four modules: the motion detector, face detector, face tracker, and face validator. The motion detector extracts motion areas by using a spatial-temporal bi-differential method with a very low computational cost. The face detector integrates motions into a cascade face detection framework to reject most of non-face scanning-windows to ensure efficient face localization. The face tracker fuses motion feature with color feature to alleviate the drifting problem during tracking. The face validator builds face appearance models online and identifies each specific tracked face to avoid confusion. Experimental results on three challenging video sequences show that the proposed face tracking system outperforms the state-of-the-art face tracker and consumes only 5-13% CPU resources of a low-spec onboard computer while processing in real time. Copyright (c) 2016 John Wiley & Sons, Ltd.
期刊：
ACM International Conference Proceeding Series,2017年Part F131203:64-68
作者机构：
[Lian, Yang; Chen, Jingying; Su, Chunyan] National Engineering Research Center for E-Learning, Central China Normal University, Wuhan;430079, China;[Lian, Yang; Chen, Jingying; Su, Chunyan] 430079, China
会议名称：
2017 International Conference on Digital Technology in Education, ICDTE 2017
会议时间：
August 6, 2017 - August 8, 2017
会议地点：
Taipei, Taiwan
会议论文集名称：
ICDTE '17: Proceedings of the International Conference on Digital Technology in Education
摘要：
Doctoral dissertation is one of the most important parts in the doctoral degree program. This paper analyzes the data of doctoral dissertations in a university for three consecutive years, and investigates the factors related to the quality of doctoral dissertations from the aspects of students' characteristics and cultivation methods. The results show that the enrollment age, study period, mode of study and subject categories, whether cross-specialty, etc., have significant impact on the quality of doctoral dissertations. Then, a prediction model based on the weighted Random Forest (RF) is presented to predict the quality of doctoral dissertation, which is effective for unbalanced data and improves the generalization ability of the original RF, the encouraging result of 81.29% prediction rate has been obtained, which provides objective evidence for doctoral degree program management.
摘要：
Speech interaction is a prominent interaction technology in educational robot, and educational robot with speech interaction can have a more harmonic and natural way of interaction. In some specific application scenarios such as education, the performance of speech recognition is not satisfactory due to the wide range of vocabularies and poor network. This paper trains the language model with professional vocabulary in the specific subject and establishes the corresponding phonetic dictionary in speech recognition based on Sphinx, and combines with speech synthesis based on Ekho to implement the speech interaction of educational robot in the offline state, and designs the relevant application of speech interaction about the teaching scenario of Chinese ancient poetry. The experiment demonstrates that the system designed and implemented can meet the demand of practical application of speech interaction in specific subject, and achieve the purpose of enhancing the student's learning experience and improving the learning effect.
期刊：
International Journal of Distance Education Technologies,2017年15(3):1-14 ISSN：1539-3100
通讯作者：
Liu, SY
作者机构：
[Liu, Sanya; Ni, Cheng; Liu, Zhi; Cheng, Hercy N.H.; Peng, Xian] National Engineering Research Center for E-Learning, Central China Normal University, Wuhan, China
通讯机构：
[Liu, SY ] ;Cent China Normal Univ, Natl Engn Res Ctr E Learning, Wuhan, Hubei, Peoples R China.
关键词：
Sanya Liu;Cheng Ni;Zhi Liu;Xian Peng;Hercy N.H. Cheng;Mining Individual Learning Topics in Course Reviews Based on Author Topic Model: Education;IGI Global;Journal Article;Education;IGI Publishing
摘要：
Nowadays, Massive Open Online Courses (MOOC) has obtained a rapid development and drawn much attention from the areas of learning analytics and artificial intelligence. There are lots of unstructured data being generated in online reviews area. The learning behavioral data become more and more diverse, and they prompt the emergence of big data in education. To mine useful information from these data, we need to use educational data mining and learning analysis technique to study the learning feelings and discussed topics among learners. This paper aims to mine and analyze topic information hidden in the unstructured reviews data in MOOC, a novel author topic model based on an unsupervised learning idea is proposed to extract learning topics for the each learner. According to the experimental results, we will analyze and focuses of interests of learners, which facilitates further personalized course recommendation and improve the quality of online courses.
摘要：
Low-cost tags are widely used, but have very limited storage space and computing power. In this paper, we propose an efficent lightweight radio-frequency identification (RFID) authentication protocol with strong trajectory privacy protection to balance the security and availability of RFID systems. In this protocol, tags only adopt pseudo-random number generator and XOR operation. In the authentication process, tags always use pseudonyms to prevent the exposure of sensitive messages, the pseudonyms and secret numbers of the tags are synchronized with the background server all the time. The analysis shows that the protocol can solve security issues such as desynchronization attack, man in the middle attack, forward security, replay attack, clone and so on effectively, and meet the requirements of low-cost tags. The trajectory privacy model of RFID systems is also used to prove the strong trajectory privacy and security of the protocol. This protocol has a better performance in terms of storage cost, computation cost and communication cost, and the search efficiency of the background server comparing to the existing relevant research results.
作者机构：
[桑农; 高常鑫; 李逢; 颜轶; 王洪智] National Key Laboratory of Science and Technology on Multispectral Information Processing, School of Automation, Huazhong University of Science and Technology, Wuhan, Hubei, 430074, China;[刘乐元] National Engineering Research Center for E-Learning, Central China Normal University, Wuhan, Hubei, 430079, China
通讯机构：
National Key Laboratory of Science and Technology on Multispectral Information Processing, School of Automation, Huazhong University of Science and Technology, Wuhan, Hubei, China
关键词：
hand tracking;multiple features;real-time;human computer interaction
摘要：
Hand tracking in unconstrained environments remains an extremely challenging problem due to several factors, such as background clutter, deformation, and motion blur. In this paper, we combine motion, color, and Haar-like features to construct a real-time and robust hand tracking system. Haar-like features successfully defeat moving skincolored backgrounds, although they are unstable for the whole situation. Three weak trackers are built using each kind of feature and integrated in a boosted cascade. If one stage makes sure of the object position, no other stages is carried out. Otherwise it provides its own point of view to guide the next stage. We realize the proposed approach and demonstrate it on several challenging sequences.
期刊：
Lecture Notes in Computer Science,2017年10309:475-488 ISSN：0302-9743
通讯作者：
Zhou, Zongkui
作者机构：
[Wang, Lu] Cent China Normal Univ, Natl Engn Res Ctr E Learning, Wuhan 430079, Hubei, Peoples R China.;[Tian, Yuan; Lei, Yuju; Zhou, Zongkui] Cent China Normal Univ, Sch Psychol, Wuhan 430079, Hubei, Peoples R China.
通讯机构：
[Zhou, Zongkui] C;Cent China Normal Univ, Sch Psychol, Wuhan 430079, Hubei, Peoples R China.
会议名称：
10th International Conference on Blended Learning (ICBL)
会议时间：
JUN 27-29, 2017
会议地点：
City Univ Hong Kong, Hong Kong, HONG KONG
会议主办单位：
City Univ Hong Kong
会议论文集名称：
Lecture Notes in Computer Science
关键词：
Big five personality traits;Traditional classroom;Online learning;Flipped classroom;Learning achievement
作者机构：
[Lu, Chun; Li, Congcong; Zhou, Wenting] Cent China Normal Univ, Natl Engn Res Ctr E Learning, Wuhan, Hubei, Peoples R China.;[Wu, Di] Cent China Normal Univ, Collaborat & Innovat Ctr Educ Technol, Wuhan, Hubei, Peoples R China.
通讯机构：
[Wu, Di] C;Cent China Normal Univ, Collaborat & Innovat Ctr Educ Technol, Wuhan, Hubei, Peoples R China.
会议名称：
10th International Conference on Blended Learning (ICBL)
会议时间：
JUN 27-29, 2017
会议地点：
City Univ Hong Kong, Hong Kong, HONG KONG
会议主办单位：
City Univ Hong Kong
会议论文集名称：
Lecture Notes in Computer Science
关键词：
Development of ICT in education;Essential elements of school construction;Teachers' ICT competence;Rural-urban divide
摘要：
The development of ICT in schools was influenced by many external environment factors and internal ICT factors. To identify the school-level factors linked with schools ICT situation, and the differences of the ICT implementation stages of rural and urban schools, a survey of 4,357 was conducted. Four subscales including environment, support, application, and fund allocation were extracted from the Features of School ICT Situation (FSIS). Through stepwise regression analysis, it is revealed that the ICT literacy of teacher is the most important predictor for schools’ FSIS. In addition, there are more Essential Elements of School Characteristics (EESC) items that can make significant prediction of FSIS for rural schools than for urban schools, and their predictive power for FSIS of rural schools is stronger than that of urban schools. The result reflects that EESC items which evaluate background and basic situation of ICT development in schools have stronger positive predictive power for schools in rural areas and the number of predictive indicators is larger. EESC items have relatively weaker predictive power and fewer predictive indicators for urban schools.
作者机构：
[Liu, Yuanyuan; Xie, Zhong] China Univ Geosci, Fac Informat Engn, Wuhan, Hubei, Peoples R China.;[Yuan, Xiaohui] Univ North Texas, Dept Comp Sci & Engn, Denton, TX USA.;[Song, Wu; Chen, Jingying] Cent China Normal Univ, Natl Engn Res Ctr E Learning, Wuhan, Hubei, Peoples R China.
通讯机构：
[Yuan, Xiaohui] U;Univ North Texas, Dept Comp Sci & Engn, Denton, TX USA.
关键词：
Head detection;Head pose estimation;Joint detection-estimation;Multi-level structured hybrid forest;Multiple structured features
摘要：
In real-world applications, factors such as illumination variation, occlusion, and poor image quality, etc. make head detection and pose estimation much more challenging. In this paper, we propose a multi-level structured hybrid forest (MSHF) for joint head detection and pose estimation. Our method extends the hybrid framework of classification and regression forests by introducing multi-level splitting functions and multi-structural features. Multi-level splitting functions are used to construct trees in different layers of MSHF. Multi-structured features are extracted from randomly selected image patches, which are either head region or the background. The head contour is derived from these patches using the signed distance of the patch center to the head contour by MSHF regression. The randomly selected sub-regions from the patches within the head contour are used to develop the MSHF for head pose estimation in a coarse-to-fine manner. The weighted neighbor structured aggregation integrates votes from trees to achieve an estimation of continuous pose angles. Experiments were conducted using public datasets and video streams. Compared to the state-of-the-art methods, MSHF achieved improved performance and great robustness with an average accuracy of 90% and the average angular error of 6.6. The averaged time for performing a joint head detection and pose estimation is about 0.44s. In real-world applications, factors such as illumination variation, occlusion, and poor image quality, etc. make head detection and pose estimation much more challenging. In this paper, we propose a multi-level structured hybrid forest (MSHF) for joint head detection and pose estimation. Our method extends the hybrid framework of classification and regression forests by introducing multi-level splitting functions and multi-structural features. Multi-level splitting functions are used to construct trees in different layers of MSHF. Multi-structured features are extracted from randomly selected image patches, which are either head region or the background. The head contour is derived from these patches using the signed distance of the patch center to the head contour by MSHF regression. The randomly selected sub-regions from the patches within the head contour are used to develop the MSHF for head pose estimation in a coarse-to-fine manner. The weighted neighbor structured aggregation integrates votes from trees to achieve an estimation of continuous pose angles. Experiments were conducted using public datasets and video streams. Compared to the state-of-the-art methods, MSHF achieved improved performance and great robustness with an average accuracy of 90% and the average angular error of 6.6. The averaged time for performing a joint head detection and pose estimation is about 0.44s.
期刊：
Multimedia Tools and Applications,2017年76(17):18027-18045 ISSN：1380-7501
通讯作者：
Huang, Tao
作者机构：
[Zheng, Zhigao] Huazhong Univ Sci & Technol, Sch Comp & Technol, Wuhan 430079, Hubei, Peoples R China.;[Jeong, Hwa-Young] Kyung Hee Univ, Humanitas Coll, 1 Hoegi Dong, Seoul, South Korea.;[Shu, Jiangbo; Huang, Tao] Cent China Normal Univ, Natl Engn Res Ctr E Learning, Wuhan 430079, Hubei, Peoples R China.
通讯机构：
[Huang, Tao] C;Cent China Normal Univ, Natl Engn Res Ctr E Learning, Wuhan 430079, Hubei, Peoples R China.
关键词：
Kernel density estimation;Distributed;Data stream;Stream analysis;Exponential decay policy
摘要：
Multimedia networks hold the promise of facilitating large-scale, real-time data processing in complex environments. Their foreseeable applications will help protect and monitor military, environmental, safety-critical, or domestic infrastructures and resources. Cloud infrastructures promise to provide high performance and cost effective solutions to large scale data processing problems. This paper focused on the outlier detection over distributed data stream in real time, proposed kernel density estimation (KDE) based outlier detection algorithm KDEDisStrOut in Storm, firstly formalized the problem of outlier detection using the kernel density estimation technique and update the transported data incrementally between the child node and the coordinator node which reduces the communication cost. Then the paper adopted the exponential decay policy to keep pace with the transient and evolving natures of stream data and changed the weight of different data in the sliding window adaptively made the data analysis more reasonable. Theoretical analysis and experiments on Storm with synthetic and real data show that the KDEDisStrOut algorithm is efficient and effective compared with existing outlier detection algorithms, and more suitable for data streams.
期刊：
The Journal of Engineering,2017年2017(6):212-219 ISSN：2051-3305
通讯作者：
Jerry Z. Xie
作者机构：
[Xie, Jerry Z.; Yang, Zongkai] Cent China Normal Univ, Natl Engn Res Ctr E Learning, 152 Luoyu Rd, Wuhan 430079, Peoples R China.;[Xie, Jerry Z.; Chen, Shaoli] ECCOM Network Syst Co Ltd, 8F South Tower,9 Lujiazui Software Pk,20 Lane, Shanghai 200127, Peoples R China.
通讯机构：
[Jerry Z. Xie] N;National Engineering Research Centre for E-Learning, Central China Normal University, No. 152 Luoyu Road, Wuhan, 430079 People's Republic of China<&wdkj&>ECCOM Network System Co., Ltd., 8F South Tower, No. 9 Lujiazui Software Park, No. 20, Lane 91, E'shan Road, Shanghai, 200127 People's Republic of China
摘要：
Since multicast reduces bandwidth consumption in multimedia grid computing, the middleware for monitoring the performance and topology of multicast communications is important to the design and management of multimedia grid applications. However, the current middleware technologies for multicast performance monitoring are still far from attaining the level of maturity and there lacks consistent approaches to obtain the evaluation data for multicast. In this study, to serve a clear guide for the design and implementation of the multicast middleware, two algorithms are developed for organising all constituents in multicast communications and analysing the multicast performance in two topologies – ‘multicast distribution tree’ and ‘clusters distribution’, and a definitive set of corresponding metrics that are comprehensive yet viable for evaluating multicast communications are also presented. Instead of using the inference data from unicast measurements, in the proposed middleware, the measuring data of multicast traffic are obtained directly from multicast protocols in real time. Moreover, this study makes a middleware implementation which is integrated into a real access grid multicast communication infrastructure. The results of the implementation demonstrate the substantial improvements in the accuracy and real time in evaluating the performance and topology of multicast network.
作者：
Sanya Liu;Zhenfan Hu;Xian Peng;Zhi Liu;Hercy N. H. Cheng;...
期刊：
International Journal of Distance Education Technologies,2017年15(1):15-27 ISSN：1539-3100
作者机构：
[Jianwen Sun; Hercy N. H. Cheng; Zhenfan Hu; Sanya Liu; Zhi Liu; Xian Peng] National Engineering Research Center for E-Learning, Central China Normal University, Wuhan, China
关键词：
Behavioral Pattern;Cloud Classroom;Learning Analytics;Massive Open Online Courses;Sequential Analysis
期刊：
ICDTE '17: Proceedings of the 1st International Conference on Digital Technology in Education,2017年Part F131203:Pages 1–6
通讯作者：
Liu, Tingting(tingtingliu89619@gmail.com)
作者机构：
[Chen, Zengzhao; Wang, Cong; Feng, Xiaochao; Zhang, Chao; Liu, Tingting] National Engineering Research Center for E-Learning, Central China Normal University, Wuhan, 430079, China
会议名称：
978-1-4503-5283-3
会议时间：
August, 2017
会议地点：
Taipei Taiwan
会议论文集名称：
ICDTE '17: Proceedings of the International Conference on Digital Technology in Education
摘要：
With the development of smart education, the computer-assisted teaching system plays an important role in classroom teaching, while there are still some deficiencies in the current teaching systems which hindering the communication between teachers and students sometimes. In the paper, we developed a teaching platform (StarC) with gesture recognition technology, which contains two functions namely lesson preparation and teaching. Furthermore, we proposed 16 kinds of gestures to instantiate the corresponding teaching functions. For a further study, the platform is applied in a teaching reform project in Suzhou city. This paper analyzed the pretest and post-test scores of experimental group and the control group. The results indicated that students' achievement is much more significant in experimental group than that in the control group. It concluded that the StarC platform can better promote teaching than the traditional multimedia used in classroom teaching.
