作者机构:
[Wang, Xiang; Zhang, Zhaoli; Liu, Hai] Cent China Normal Univ, Natl Engn Res Ctr E Learning, Wuhan 430079, Peoples R China.;[Wang, Xiang; Zhang, Wei] Cent China Normal Univ, Natl Engn Lab Educ Big Data, Wuhan 430079, Peoples R China.;[Li, You-Fu; Liu, Hai] City Univ Hong Kong, Dept Mech Engn, Hong Kong, Peoples R China.
通讯机构:
[Zhang, Wei] C;Cent China Normal Univ, Natl Engn Lab Educ Big Data, Wuhan 430079, Peoples R China.
摘要:
Head pose estimation (HPE) has been widely applied in human attention recognition, robot vision and assistant driving. Infrared (IR) images bear unique advantages of being still effective under visible scenarios, which are resistance to illumination changing and strong penetration. However, the lack of public IR database hinders the research progress in the low illumination environment. In this paper, we establish a first-of-its-kind infrared head pose (IRHP) database and propose a novel convolutional neural network architecture IRHP-Net on the IRHP database. The IRHP database contains 145 kinds of IR head pose images of subjects, and benchmark evaluations are conducted on our database by the facial features based standard HPE classification methods to prove the usability and effectiveness of IRHP database. To extract the adaptive features for the IR images, a novel multi-scale feature fusion descriptor is developed in the proposed IRHP-Net model. Quantitative assessments of the proposed method on the IRHP images demonstrate the significant improvements over the traditional methods. The new proposed IRHP-Net model can be utilized in human attention recognition and intelligent driving assistant system. (c) 2020 Elsevier B.V. All rights reserved.
作者机构:
[Zhang, Wei; Yi, Baolin; Qin, Shiming] Cent China Normal Univ, Natl Engn Res Ctr E Learning, Wuhan, Hubei, Peoples R China.;[Tian, Peng] Cent China Normal Univ, Coll Publ Adm, Wuhan, Hubei, Peoples R China.
通讯机构:
[Tian, Peng] C;Cent China Normal Univ, Coll Publ Adm, Wuhan, Hubei, Peoples R China.
摘要:
Automatic recommendation has become an increasingly relevant problem to industries, which allows users to discover new items that match their tastes and enables the system to target items to the right users. In this paper, we propose a deep learning (DL) based collaborative filtering framework, namely, deep matrix factorization (DMF), which can integrate any kind of side information effectively and handily. In DMF, two feature transforming functions are built to directly generate latent factors of users and items from various input information. As for the implicit feedback that is commonly used as input of recommendation algorithms, implicit feedback embedding (IFE) is proposed. IFE converts the high-dimensional and sparse implicit feedback information into a low-dimensional real-valued vector retaining primary features. Using IFE could reduce the scale of model parameters conspicuously and increase model training efficiency. Experimental results on five public databases indicate that the proposed method performs better than the state-of-the-art DL-based recommendation algorithms on both accuracy and training efficiency in terms of quantitative assessments.
摘要:
Radio-frequency Identification (RFID) grouping proof protocol is widely used in medical healthcare industry, transportation industry, crime forensics and so on,it is a research focus in the field of information security. The RFID grouping proof protocol is to prove that some tags belong to the same group and exist simultaneously. To improve the applicability of the RFID grouping proof protocol in low cost tag applications, this paper proposes a new scalable lightweight RFID grouping proof protocol. Tags in the proposed protocol only generate pseudorandom numbers and execute exclusive-or(XOR) operations. An anti-collision algorithm based on adaptive 4-ary pruning query tree (A4PQT) is used to identify the response message of tags. Updates to secret information in tags are kept synchronized with the verifier during the entire grouping proof process. Based on these innovations, the proposed protocol resolves the scalability issue for low-cost tag systems and improves the efficiency and security of the authentication that is generated by the grouping proof. Compared with other state-of-the art protocols, it is shows that the proposed protocol requires lower tag-side computational complexity, thereby achieving an effective balance between protocol security and efficiency.
作者:
Liu, Zhi;Zhang, Wenjing;Cheng, Hercy N. H.;Sun, Jianwen(孙建文);Liu, Sannyuya
期刊:
International Journal of Distance Education Technologies,2018年16(2):37-50 ISSN:1539-3100
作者机构:
[Sun, Jianwen; Zhang, Wenjing; Liu, Sannyuya; Liu, Zhi; Cheng, Hercy N. H.] National Engineering Research Center for E-Learning, Central China Normal University, Wuhan, China
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.
摘要:
With the rapid development of information technology, blended learning theory in higher education has become more and more maturing, many experts and scholars at home and abroad are interested in this field. In this paper, we selected the information economics course as the observation, and observed the students during the period from 2014 to 2016,. In the experiment, some students learned in traditional class and the others learned in blended environment using Hstar teaching platform as the supporting tool every year. The results show that using Hstar teaching platform as blended learning tool made teaching and learning more effective. So we can conclude that Hstar teaching platform is an effective platform in blended learning in higher education.
摘要:
With the continuous development of online learning platforms, educational data analytics and prediction have become a promising research field, which are helpful for the development of personalized learning system. However, the indicator's selection process does not combine with the whole learning process, which may affect the accuracy of prediction results. In this paper, we induce 19 behavior indicators in the online learning platform, proposing a student performance prediction model which combines with the whole learning process. The model consists of four parts: data collection and pre-processing, learning behavior analytics, algorithm model building and prediction. Moreover, we apply an optimized Logistic Regression algorithm, taking a case to analyze students' behavior and to predict their performance. Experimental results demonstrate that these eigenvalues can effectively predict whether a student was probably to have an excellent grade.
摘要:
Here we propose a trajectory privacy model to solve privacy and security problems with radio-frequency identification (RFID) systems. The model first formalizes an Adversary Model and then defines an adversary indistinguishability privacy game and interval security privacy game according to the ability of the adversary. Based on the privacy game between adversary and challenger, the author gives the definition of weak trajectory privacy and strong trajectory privacy. Finally, we analyzed the privacy protection level of present RFID systems with the help of this trajectory privacy model. It can be seen that the trajectory privacy model can effectively analyze and find the privacy vulnerabilities of RFID security protocols.