基于HodgeRank的人脸美貌度预测

随着计算机技术的迅速发展以及人脸识别技术的成熟,人脸美貌度受到越来越多的关注和研究.针对目前的研究方法中存在的对训练数据集的评分过多依赖人工操作,以及对人脸美貌度的预测结果不够详细等问题,本文提出基于HodgeRank的人脸美貌度预测系统,利用数据挖掘方法学习女性人脸的美貌度特征,构造一个模拟预测人脸美貌度的系统.明显区别于之前的研究,该系统训练和测试时采用的人脸数据集放宽了对姿态、光照以及所处环境等条件的限制,评分所需的人工操作大大减少,无需进行大量的人工标定,使用图像的原始像素或纹理特征作为输入,分别采用聚类和改进的BP网络的方法,得到更符合人类特征的美貌度预测结果.     

云计算下基于属性的访问控制方法

云计算作为一种新型的服务模式,近年来受到各个企业的青睐.虽有云有三种部署模式,企业可以自己选择部署方式,但是云中数据存储和传输安全还是企业最为担心的.因此,运用了一种基于属性的控制策略方法(Attribute Based Access Control, ABAC)来保证只有授权用户才有权限来操作云中的数据.另外运用粗糙集属性约简的方法来减除冗余属性保留权威属性减少策略冲突问题的出现,加快访问时间.     

A service recovery method to enhance customer satisfaction with a case study of motion sport game

Service recovery is a concept that involves regaining customer satisfaction and loyalty. Although existing studies have addressed its importance, few service recovery methodologies have been developed. Thus, there is a need to develop a systematic framework that can first detect customer status and then provide appropriate recovery recommendations to ensure customer satisfaction. This study develops such a framework in relation to an individual health maintenance practice. While regular moderate exercise is known to promote good health and help prevent disease, the lack of exercise remains a crucial health issue. This study presents a logistic regression model that uses both physiological data (heart rate and blood pressure) and psychological data (ratings of perceived exertion) to detect the effectiveness of an individual's exercise and to provide sports‐related service recovery suggestions as needed. To implement and assess this service recovery model, the use of a motion gaming system is proposed.     

A divide-and-conquer hole-filling method for handling disocclusion in single-view rendering

Large holes are unavoidably generated in depth image based rendering (DIBR) using a single color image and its associated depth map. Such holes are mainly caused by disocclusion, which occurs around the sharp depth discontinuities in the depth map. We propose a divide-and-conquer hole-filling method which refines the background depth pixels around the sharp depth discontinuities to address the disocclusion problem. Firstly, the disocclusion region is detected according to the degree of depth discontinuity, and the target area is marked as a binary mask. Then, the depth pixels located in the target area are modified by a linear interpolation process, whose pixel values decrease from the foreground depth value to the background depth value. Finally, in order to remove the isolated depth pixels, median filtering is adopted to refine the depth map. In these ways, disocclusion regions in the synthesized view are divided into several small holes after DIBR, and are easily filled by image inpainting. Experimental results demonstrate that the proposed method can effectively improve the quality of the synthesized view subjectively and objectively.     

Iterative quadtree decomposition based automatic selection of the seed point for ultrasound breast tumor images

Based on seed region growing method, lesion segmentation for ultrasound breast tumor images often requires manual selection of the seed point, which is both time-consuming and laborious. To overcome this limit, this paper attempts to explore an automatic method for finding the seed point inside the tumor. Two criteria combining iterative quadtree decomposition (QTD) and the gray characteristics of the lesion are thus designed to locate the seed point. One is to seek the biggest homogenous region and the other is to select the seed region where the seed point is found. Furthermore, this study validates the proposed algorithm through 110 ultrasonic breast tumor images (including 58 malignant tumor images and 52 benign tumor images). According to the needs of the seed region growing algorithm, if the seed point is found inside the tumor, it means the proposed method is correct. Otherwise, it means that the method is a failure. As the quantitative experiment results show, the proposed method in this paper can automatically find the seed point inside the tumor with an accuracy rate of 97.27 %.     

Top-<Emphasis Type="Italic">k</Emphasis> temporal keyword search over social media data

Social media services have already become main sources for monitoring emerging topics and sensing real-life events. A social media platform manages social stream consisting of a huge volume of timestamped user generated data, including original data and repost data. However, previous research on keyword search over social media data mainly emphasizes on the recency of information. In this paper, we first propose a problem of top-k most significant temporal keyword query to enable more complex query analysis. It returns top-k most popular social items that contain the keywords in the given query time window. Then, we design a temporal inverted index with two-tiers posting list to index social time series and a segment store to compute the exact social significance of social items. Next, we implement a basic query algorithm based on our proposed index structure and give a detailed performance analysis on the query algorithm. From the analysis result, we further refine our query algorithm with a piecewise maximum approximation (PMA) sketch. Finally, extensive empirical studies on a real-life microblog dataset demonstrate the combination of two-tiers posting list and PMA sketch achieves remarkable performance improvement under different query settings.     

Efficient entanglement concentration of arbitrary unknown less-entangled three-atom W states via photonic Faraday rotation in cavity QED

We propose an efficient entanglement concentration protocol (ECP) for nonlocal three-atom systems in an arbitrary unknown less-entangled W state, resorting to the Faraday rotation of photonic polarization in cavity quantum electrodynamics and the systematic concentration method. In the first step of the present ECP, one party in quantum communication performs a parity-check measurement on her two atoms in two three-atom systems for dividing the composite six-atom systems into two groups. In the first group, the three parties will obtain some three-atom systems in a less-entangled state with two unknown coefficients. In the second group, they will obtain some less-entangled two-atom systems. In the second step of the ECP, the three parties can obtain a subset of three-atom systems in the standard maximally entangled W state by exploiting the above three-atom and two-atom systems. Moreover, the preserved systems in the failed instances can be used as the resource for the entanglement concentration in the next round. The total success probability of the ECP can therefore be largely increased by iterating the entanglement concentration process several rounds. The distinct feature of our ECP is that we can concentrate arbitrary unknown atomic entangled W states via photonic Faraday rotation, and thus it may be universal and useful for entanglement concentration in future quantum communication network.     

Stability analysis of linear systems with an interval time-varying delay – A delay-range-partition approach

In this paper, the stability analysis problem of linear systems with an interval time-varying delay is investigated. Firstly, an augmented Lyapunov-Krasovskii functional is constructed, which includes more information of the delay’s range and the delay’s derivative. Secondly, based on two improved integral inequalities which are less conservative than Jensen’s integral inequalities, a delay-range-partition (DRP) approach is proposed to estimate the upper bound of the derivative of the augmented Lyapunov-Krasovskii functional. Then, less conservative stability criteria in the form of linear matrix inequality (LMI) are established no matter whether the lower bound of delay is zero or not. Finally, to illustrate the effectiveness of the stability criteria proposed in this paper, two numerical examples are given, and their results are compared with the existing results.     

Reanalysis-based space mapping method,an alternative optimization way for expensive simulation-based problems

The computational cost of evaluation in the design procedure is an essential bottleneck for simulation-based applications. To handle this problem, space mapping(SM) algorithm and reanalysis method are integrated to improve the efficiency of optimization without loss of accuracy. In the suggested method, the SM algorithm is used to construct the projection between coarse and fine spaces. In the coarse space, the reanalysis method is used to analyze finite element (FE) models. Compared with surrogate assisted evaluation, the accuracy of reanalysis method is significantly improved due to introduction of equilibrium equations. Generally, compared with other SM algorithms, coarse and fine spaces are based on solvers instead of models. Due to the high accurate reanalysis method, the reanalysis-based SM algorithm is easier to converge. Moreover, to integrate geometry and finite element models easily, a B-rep based reanalysis method is also introduced. To verify the performance of the suggested method, two examples have been carried out by using reanalysis based space mapping method. According to the results, the efficiency of optimization procedure is improved significantly.     

Variable stiffness composite material design by using support vector regression assisted efficient global optimization method

In this work, a surrogate assisted optimization method is utilized to optimize buckling loads of variable stiffness composites made by fiber steering. To improve the efficiency of optimization procedure, an expected improvement criterion is employed. Moreover, considering uncertainties of the fiber placement, a robust surrogate, least square support vector regression (LSSVR) considering empirical and structural risks is integrated with the expected improvement (EI) criterion and applied to two applications. The first case is the fiber path design of a variable stiffness plate under the compression load. The second one is the fiber path design of a variable stiffness cylinder under the bending load. According to results of the optimization, the buckling load of the variable stiffness plate has 52.63% improvement than the constant stiffness plate and 24.3% improvement than the quasi-isotropic plate. The buckling load of the variable stiffness cylinder has 40.22% improvement than the constant stiffness cylinder and 31.25% improvement than the quasi-isotropic cylinder. Furthermore, to verify the robustness of optimal design variables for the variable stiffness cylinder, the perturbed optimum design is presented and demonstrates that the results are reliable.