China Internet Network Information Center

Face antispoofing has now attracted intensive attention, aiming to assure the reliability of face biometrics. We notice that currently most of face antispoofing databases focus on data with little variations, which may limit the generalization performance of trained models since potential attacks in real world are probably more complex. In this paper we release a face antispoofing database which covers a diverse range of potential attack variations. Specifically, the database contains 50 genuine subjects, and fake faces are made from the high quality records of the genuine faces. Three imaging qualities are considered, namely the low quality, normal quality and high quality. Three fake face attacks are implemented, which include warped photo attack, cut photo attack and video attack. Therefore each subject contains 12 videos (3 genuine and 9 fake), and the final database contains 600 video clips. Test protocol is provided, which consists of 7 scenarios for a thorough evaluation from all possible aspects. A baseline algorithm is also given for comparison, which explores the high frequency information in the facial region to determine the liveness. We hope such a database can serve as an evaluation platform for future researches in the literature.

This book develops a framework that shows how uncertainty in Artificial Intelligence (AI) expands and generalizes traditional AI. It explores the uncertainties of knowledge and intelligence. The authors focus on the importance of natural language – the carrier of knowledge and intelligence, and introduce efficient physical methods for data mining amd control. In this new edition, we have more in-depth description of the models and methods, of which the mathematical properties are proved strictly which make these theories and methods more complete. The authors also highlight their latest research results.

<h3>Objective</h3> It is well documented that both work stress and work motivation are key determinants of job satisfaction. The aim of this study was to examine levels of work stress and motivation and their contribution to job satisfaction among community health workers in Heilongjiang Province, China. <h3>Design</h3> Cross-sectional survey. <h3>Setting</h3> Heilongjiang Province, China. <h3>Participants</h3> The participants were 930 community health workers from six cities in Heilongjiang Province. <h3>Primary and secondary outcome measures</h3> Multistage sampling procedures were used to measure socioeconomic and demographic status, work stress, work motivation and job satisfaction. Logistic regression analysis was performed to assess key determinants of job satisfaction. <h3>Results</h3> There were significant differences in some subscales of work stress and work motivation by some of the socioeconomic characteristics. Levels of overall stress perception and scores on all five work stress subscales were higher in dissatisfied workers relative to satisfied workers. However, levels of overall motivation perception and scores on the career development, responsibility and recognition motivation subscales were higher in satisfied respondents relative to dissatisfied respondents. The main determinants of job satisfaction were occupation; age; title; income; the career development, and wages and benefits subscales of work stress; and the recognition, responsibility and financial subscales of work motivation. <h3>Conclusions</h3> The findings indicated considerable room for improvement in job satisfaction among community health workers in Heilongjiang Province in China. Healthcare managers and policymakers should take both work stress and motivation into consideration, as two subscales of work stress and one subscale of work motivation negatively influenced job satisfaction and two subscales of work motivation positively influenced job satisfaction.

BACKGROUND: Magnifying endoscopy with narrow band imaging (M-NBI) has been applied to examine early gastric cancer by observing microvascular architecture and microsurface structure of gastric mucosal lesions. However, the diagnostic efficacy of non-experts in differentiating early gastric cancer from non-cancerous lesions by M-NBI remained far from satisfactory. In this study, we developed a new system based on convolutional neural network (CNN) to analyze gastric mucosal lesions observed by M-NBI. METHODS: A total of 386 images of non-cancerous lesions and 1702 images of early gastric cancer were collected to train and establish a CNN model (Inception-v3). Then a total of 341 endoscopic images (171 non-cancerous lesions and 170 early gastric cancer) were selected to evaluate the diagnostic capabilities of CNN and endoscopists. Primary outcome measures included diagnostic accuracy, sensitivity, specificity, positive predictive value, and negative predictive value. RESULTS: The sensitivity, specificity, and accuracy of CNN system in the diagnosis of early gastric cancer were 91.18%, 90.64%, and 90.91%, respectively. No significant difference was spotted in the specificity and accuracy of diagnosis between CNN and experts. However, the diagnostic sensitivity of CNN was significantly higher than that of the experts. Furthermore, the diagnostic sensitivity, specificity and accuracy of CNN were significantly higher than those of the non-experts. CONCLUSIONS: Our CNN system showed high accuracy, sensitivity and specificity in the diagnosis of early gastric cancer. It is anticipated that more progress will be made in optimization of the CNN diagnostic system and further development of artificial intelligence in the medical field.

Smoke detection plays an important role in industrial safety warning systems and fire prevention. Due to the complicated changes in the shape, texture, and color of smoke, identifying the smoke from a given image still remains a substantial challenge, and this has accordingly aroused a considerable amount of research attention recently. To address the problem, we devise a new deep dual-channel neural network (DCNN) for smoke detection. In contrast to popular deep convolutional networks (e.g., Alex-Net, VGG-Net, Res-Net, and Dense-Net and the DNCNN that is specifically devoted to detecting smoke), our proposed end-to-end network is mainly composed of dual channels of deep subnetworks. In the first subnetwork, we sequentially connect multiple convolutional layers and max-pooling layers. Then, we selectively append the batch normalization layer to each convolutional layer for overfitting reduction and training acceleration. The first subnetwork is shown to be good at extracting the detailed information of smoke, such as texture. In the second subnetwork, in addition to the convolutional, batch normalization, and max-pooling layers, we further introduce two important components. One is the skip connection for avoiding the vanishing gradient and improving the feature propagation. The other is the global average pooling for reducing the number of parameters and mitigating the overfitting issue. The second subnetwork can capture the base information of smoke, such as contours. We finally deploy a concatenation operation to combine the aforementioned two deep subnetworks to complement each other. Based on the augmented data obtained by rotating the training images, our proposed DCNN can promptly and stably converge to the perfect performance. Experimental results conducted on the publicly available smoke detection database verify that the proposed DCNN has attained a very high detection rate that exceeds 99.5% on average, superior to state-of-the-art relevant competitors. Furthermore, our DCNN only employs approximately one-third of the parameters needed by the comparatively tested deep neural networks. The source code of DCNN will be released at https://kegu.netlify.com/.

The main task of future networks is to build, as much as possible, intelligent networking architectures for intellectualization, activation, and customization. Software-defined networking (SDN) technology breaks the tight coupling between the control plane and the data plane in the traditional network architecture, making the controllability, security, and economy of network resources into a reality. As one of the important actualization methods of artificial intelligence (AI), machine learning (ML), combined with SDN architecture will have great potential in areas, such as network resource management, route planning, traffic scheduling, fault diagnosis, and network security. This paper presents the network applications combined with SDN concepts based on ML from two perspectives, namely the perspective of ML algorithms and SDN network applications. From the perspective of ML algorithms, this paper focuses on the applications of classical ML algorithms in SDN-based networks, after a characteristic analysis of algorithms. From the other perspective, after classifying the existing network applications based on the SDN architecture, the related ML solutions are introduced. Finally, the future development of the ML algorithms and SDN concepts is discussed and analyzed. This paper occupies the intersection of the AI, big data, computer networking, and other disciplines; the AI itself is a new and complex interdisciplinary field, which causes the researchers in this field to often have different professional backgrounds and, sometimes, divergent research purposes. This paper is necessary and helpful for researchers from different fields to accurately master the key issues.

Scene text detection and recognition have been well explored in the past few years. Despite the progress, efficient and accurate end-to-end spotting of arbitrarily-shaped text remains challenging. In this work, we propose an end-to-end text spotting framework, termed PAN++, which can efficiently detect and recognize text of arbitrary shapes in natural scenes. PAN++ is based on the kernel representation that reformulates a text line as a text kernel (central region) surrounded by peripheral pixels. By systematically comparing with existing scene text representations, we show that our kernel representation can not only describe arbitrarily-shaped text but also well distinguish adjacent text. Moreover, as a pixel-based representation, the kernel representation can be predicted by a single fully convolutional network, which is very friendly to real-time applications. Taking the advantages of the kernel representation, we design a series of components as follows: 1) a computationally efficient feature enhancement network composed of stacked Feature Pyramid Enhancement Modules (FPEMs); 2) a lightweight detection head cooperating with Pixel Aggregation (PA); and 3) an efficient attention-based recognition head with Masked RoI. Benefiting from the kernel representation and the tailored components, our method achieves high inference speed while maintaining competitive accuracy. Extensive experiments show the superiority of our method. For example, the proposed PAN++ achieves an end-to-end text spotting F-measure of 64.9 at 29.2 FPS on the Total-Text dataset, which significantly outperforms the previous best method. Code will be available at: git.io/PAN.

The development and current status of BeiDou Navigation Satellite System are briefly introduced. The definition and realization of the system time scales are described in detail. The BeiDou system time (BDT) is an internal and continuous time scale without leap seconds. It is maintained by the time and frequency system of the master station. The frequency accuracy of BDT is superior to 2 × 10−14 and its stability is better than 6 × 10−15/30 days. The satellite synchronization is realized by a two-way time transfer between the uplink stations and the satellite. The measurement uncertainty of satellite clock offsets is less than 2 ns. The BeiDou System has three modes of time services: radio determination satellite service (RDSS) one-way, RDSS two-way and radio navigation satellite service (RNSS) one-way. The uncertainty of the one-way time service is designed to be less than 50 ns, and that of the two-way time service is less than 10 ns. Finally, some coordinate tactics of UTC from the viewpoint of global navigation satellite systems (GNSS) are discussed. It would be helpful to stop the leap second, from our viewpoint, but to keep the UTC name, the continuity and the coordinate function unchanged.

Authentication is the first entrance to kinds of information systems; however, traditional centered single-side authentication is weak and fragile, which has security risk of single-side failure or breakdown caused by outside attacks or internal cheating. In the edge and Internet-of-Things (IoT) environment, blockchain can apply edge devices to better serve the IoT and provide decentralized high security service solutions. In this article, we proposed a blockchain-based decentralized authentication modeling scheme (named BlockAuth) in edge and IoT environment to provide a more secure, reliable, and strong fault tolerance novel solution, in which each edge device is regarded as a node to form a blockchain network. We designed secure registration and authentication strategy, blockchain-based decentralized authentication protocol, and developed the blockchain consensus, smart contract, and implemented a whole blockchain-based authentication platform for the feasibility, security, and performance evaluation. The analysis and evaluation show that the proposed BlockAuth scheme provides a more secure, reliable, and strong fault tolerance decentralized novel authentication with high-level security driven configuration management. The proposed BlockAuth scheme is suitable for password-based, certificate-based, biotechnology-based, and token-based authentication for high-level security requirement system in edge and IoT environment.

People counting is one of the key components in video surveillance applications, however, due to occlusion, illumination, color and texture variation, the problem is far from being solved. Different from traditional visible camera based systems, we construct a novel system that uses vertical Kinect sensor for people counting, where the depth information is used to remove the affect of the appearance variation. Since the head is always closer to the Kinect sensor than other parts of the body, people counting task equals to find the suitable local minimum regions. According to the particularity of the depth map, we propose a novel unsupervised water filling method that can find these regions with the property of robustness, locality and scale-invariance. Experimental comparisons with mean shift and random forest on two databases validate the superiority of our water filling algorithm in people counting.

Cyber-physical systems (CPS) have received much attention from both academia and industry. An increasing number of functions in CPS are provided in the way of services, which gives rise to an urgent task, that is, how to recommend the suitable services in a huge number of available services in CPS. In traditional service recommendation, collaborative filtering (CF) has been studied in academia, and used in industry. However, there exist several defects that limit the application of CF-based methods in CPS. One is that under the case of high data sparsity, CF-based methods are likely to generate inaccurate prediction results. In this paper, we discover that mining the potential similarity relations among users or services in CPS is really helpful to improve the prediction accuracy. Besides, most of traditional CF-based methods are only capable of using the service invocation records, but ignore the context information, such as network location, which is a typical context in CPS. In this paper, we propose a novel service recommendation method for CPS, which utilizes network location as context information and contains three prediction models using random walking. We conduct sufficient experiments on two real-world datasets, and the results demonstrate the effectiveness of our proposed methods and verify that the network location is indeed useful in QoS prediction.

Demand side management (DSM) is an important way to achieve smart energy management. Herein, a dynamic price (DP)-based demand response (DR) model is developed for DSM in smart grid. The proposed DR model can shift the peak electricity demand, thereby improving the power system stability and reliability. In a district-scale smart grid with a high photovoltaic power penetration, the energy service provider (ESP) optimizes the DP to maximize its utilities and reduce the load fluctuation while minimizing the bills and dissatisfaction of electricity consumers (ECs). The game theory model is used to explore the interaction between ESP and ECs, and the existence of Nash equilibrium is proved. The proposed DR model is validated with real-world data from a commercial and residential cluster in Suzhou City, Jiangsu Province, China. The results show that the peak electricity demands of commercial and residential ECs decreased by 4.99% and 9.99%, respectively, through the proposed DR model. Meanwhile, the ESP's net profits increased by 7.13% and 2.37%, respectively, while ensuring the ECs’ benefits. The results also demonstrate that the proposed DR model is robust in different scenarios. This article contributes to the effectiveness and efficiency of energy engineering management.

End-to-end network monitoring is essential to ensure transmission quality for Internet applications. However, in large-scale networks, full-mesh measurement of network performance between all transmission pairs is infeasible. As a newly emerging sparsity representation technique, matrix completion allows the recovery of a low-rank matrix using only a small number of random samples. Existing schemes often fix the number of samples assuming the rank of the matrix is known, while the data features thus the matrix rank vary over time. In this paper, we propose to exploit the matrix completion techniques to derive the end-to-end network performance among all node pairs by only measuring a small subset of end-to-end paths. To address the challenge of rank change in the practical system, we propose a sequential and information-based adaptive sampling scheme, along with a novel sampling stopping condition. Our scheme is based only on the data observed without relying on the reconstruction method or the knowledge on the sparsity of unknown data. We have performed extensive simulations based on real-world trace data, and the results demonstrate that our scheme can significantly reduce the measurement cost while ensuring high accuracy in obtaining the whole network performance data.

Mobile Service selection is an important but challenging problem in service and mobile computing. Quality of service (QoS) predication is a critical step in service selection in 5G network environments. The traditional methods, such as collaborative filtering (CF), suffer from a series of defects, such as failing to handle data sparsity. In mobile network environments, the abnormal QoS data are likely to result in inferior prediction accuracy. Unfortunately, these problems have not attracted enough attention, especially in a mixed mobile network environment with different network configurations, generations, or types. An ensemble learning method for predicting missing QoS in 5G network environments is proposed in this paper. There are two key principles: one is the newly proposed similarity computation method for identifying similar neighbors; the other is the extended ensemble learning model for discovering and filtering fake neighbors from the preliminary neighbors set. Moreover, three prediction models are also proposed, two individual models and one combination model. They are used for utilizing the user similar neighbors and servicing similar neighbors, respectively. Experimental results conducted in two real-world datasets show our approaches can produce superior prediction accuracy.

Failure prediction for hard disk drives is a typical and effective approach to improve the reliability of storage systems. In a large-scale data center environment, the various brands and models of drives serve diverse applications with different input/output workload patterns, and non-ignorable differences exist in each type of drive failures, which make this mechanism much challenging. Although many efforts are devoted to this mechanism, the accuracy still needs to be improved. In this article, we propose a failure prediction method for hard disk drives based on a part-voting random forest, which differentiates prediction of failures in a coarse-grained manner. We conduct groups of validation experiments on two real-world datasets, which contain the SMART data of 64,193 drives. The experimental results show that our proposed method can achieve a better prediction accuracy than state-of-the-art methods.

Smart grid can greatly improve the efficiency, reliability, and sustainability of the traditional grids. In industrial smart grid, real-time user-side metering data may be frequently collected for monitoring and controlling electricity consumption. However, the procedure of frequently metering data collection may lead to sensitive information leakage. To address the security issues in industrial smart grid, in this article, we construct an efficient identity-based metering data aggregation scheme supporting batch verification by collector and electricity service provider, respectively, which guarantees the privacy and integrity of metering data. In our scheme, collectors are allowed to collect and aggregate the metering data of users in their respective administrative domain without compromising the confidentiality of metering data. Security analysis demonstrates that our proposed scheme is provably secure in the random oracle and satisfies the above security requirements. Performance analysis indicates that our scheme outperforms existing solutions in terms of communication and computation costs.

In the mobile edge computing environment, there are a large number of mobile edge services which are the carriers of various mobile intelligent applications. So how to recommend the most suitable candidate from such a huge number of available services is an urgent task, especially the recommendation task based on quality-of-service (QoS). In traditional service recommendation, collaborative filtering (CF) has been studied in academia and industry. However, due to the mobility of users and services, there exist several defects that limit the application of the CF-based methods, especially in an edge computing environment. The most important problem is the cold-start. In this paper, we propose an ensemble model which combines the model-based CF and neighborhood-based CF. Our approach has two phases, i.e., global features learning and local features learning. In the first phase, to alleviate the cold-start problem, we propose an improved auto-encoder which deals with sparse inputs by pre-computing an estimate of the missing QoS values and can obtain the effective hidden features by capturing the complex structure of the QoS records. In the second phase, to further improve prediction accuracy, a novel computation method is proposed based on Euclidean distance that aims to address the overestimation problem. We introduce two new concepts, common invocation factor and invocation frequency factor, in similarity computation. Then we propose three prediction models, containing two individual models and one hybrid model. The two individual models are proposed to utilize user similar neighbors and service similar neighbors, and the hybrid model is to utilize all neighbors. The experiments conducted in a real-world dataset show that our models can produce superior prediction results and are not sensitive to parameter settings.

The Internet architecture has evolved over time, adapting to the needs of its users and incorporating new technology as it has been developed. The introduction of multiprotocol label switching (MPLS) as a part of the Internet forwarding architecture has immediate applications in traffic engineering and virtual private networks. In the longer term, MPLS may affect how traffic transits the Internet and the services that the Internet delivers.

Hybrid services use different protocols on various networks, such as WIFI networks, Bluetooth networks, 5G communications systems, and wireless sensor networks. Hybrid service compositions can be varied, representing an effective method of integrating into wireless scenarios context-aware applications that can sense mobility via changes in user location and combining services to support target functions. In this article, improved particle swarm optimization is introduced into the quality service evaluation of dynamic service composition to meet the mobility requirements of hybrid networks. First, this work classifies hybrid services into different task groups to generate candidate sets and then interface matching is used to compare the operations of candidate services with user requirements to select the appropriate services. Second, the service composition is determined by the particle swarm optimization simulation process, which aims to identify an optimal plan based on the calculated value from quality of service. Third, considering a change of service repository, when the quality of a composite service is lower than a predefined threshold, the local greedy algorithm and global reconfiguration method are adopted to dynamically restructure composite services. Finally, a set of experiments is conducted to demonstrate the effectiveness of the proposed method for determining the dynamic service composition, particularly when the scale of hybrid services is large. The method provides a technical reference for engineering practice that will fulfill mobile computing needs.

RFID technology holds the promise of real-time identifying, locating and monitoring physical objects. To achieve these goals, RFID events need to be collected efficiently and composed expressively. Furthermore, these events have unique characteristics, such as locomotive, temporal and history oriented which should be considered and integrated into an event engine model. The diversity of RFID applications poses further challenges to a generalized framework for RFID events processing. In this paper, the Expressive Stream Language is utilized to collect vast number of primitive events efficiently. Moreover, we introduce a novel semantics to meet requirement of expressive event composition. At last, we use Timed Petri Net to model our newly RFID complex event engine. By introducing typical applications scenarios, we evaluate the validity and effectiveness of our RFID event processing system.