An efficient approach for face recognition based on common eigenvalues

In this paper, a simple technique is proposed for face recognition among many human faces. It is based on the polynomial coefficients, covariance matrix and algorithm on common eigenvalues. The main advantage of the proposed approach is that the identification of similarity between human faces is carried out without computing actual eigenvalues and eigenvectors. A symmetric matrix is calculated using the polynomial coefficients-based companion matrices of two compared images. The nullity of a calculated symmetric matrix is used as similarity measure for face recognition. The value of nullity is very small for dissimilar images and distinctly large for similar face images. The feasibility of the propose approach is demonstrated on three face databases, i.e., the ORL database, the Yale database B and the FERET database. Experimental results have shown the effectiveness of the proposed approach for feature extraction and classification of the face images having large variation in pose and illumination.     

智能手机人脸检测系统的设计与实现

改进了肤色检测方法,设计和实现了智能手机上的人脸检测系统.由于移动平台计算能力弱、内存空间小,该系统对算法进行了优化:在去除非人脸阶段提出排队编号算法以避免多次重复遍历图像,使用静态查找表、位运算、迭代法和浮点化整法等.实验结果表明,该系统能在Pocket PC上快速且较为准确的定位出复杂背景下彩色图像中的多个正面、有一定偏转的人脸.     

An efficient feature selection method for mobile devices with application to activity recognition

This paper presents a feature selection method for data classification, which combines a model-based variable selection technique and a fast two-stage subset selection algorithm. The relationship between a specified (and complete) set of candidate features and the class label is modeled using a non-linear full regression model which is linear-in-the-parameters. The performance of a sub-model measured by the sum of the squared-errors (SSE) is used to score the informativeness of the subset of features involved in the sub-model. The two-stage subset selection algorithm approaches a solution sub-model with the SSE being locally minimized. The features involved in the solution sub-model are selected as inputs to support vector machines (SVMs) for classification. The memory requirement of this algorithm is independent of the number of training patterns. This property makes this method suitable for applications executed in mobile devices where physical RAM memory is very limited.An application was developed for activity recognition, which implements the proposed feature selection algorithm and an SVM training procedure. Experiments are carried out with the application running on a PDA for human activity recognition using accelerometer data. A comparison with an information gain-based feature selection method demonstrates the effectiveness and efficiency of the proposed algorithm.     

An efficient fire detection method based on orientation feature

This paper proposes a novel method for reliable fire detection. The burning fire usually causes rich moving features in terms of directions, which can offer the best chance to distinguish between the fire region and the non-fire one. Motivated by this observation, we design a novel orientation feature to represent this characteristic. Based on this feature, a method is proposed to detect the fire efficiently. First, fire color is utilized to extract the fire candidate areas from the surveillance video. Then, the direction is obtained by computing the optical flow for each pixel in the candidate area. The directions are discretized to four parts. By counting the percentage of pixels whose moving directions fall into these four parts in a period of time, and combining with the two parameters, i.e., both of the number of frames without the moving directions and the number of consecutive frames in the candidate area, we use these six parameters as the fire orientation feature. In the end, by training a support vector machine (SVM) classifier with the input of our fire orientation feature, the candidate area is judged whether it is a fire. Our main contribution is that we design the novel fire orientation feature. The feature can not only characterize the fire intrinsic dynamic properties accurately but also is very efficient. Compared with the art-of-state methods, the experimental results confirm that our approach significantly improves the accuracy of fire detection and impressively decreases the false alarm rate. The detection speed of our approach is also very competitive with the art-of-state fire detection methods.     

An efficient face anti-spoofing and detection model using image quality assessment parameters

Multimedia Tools and Applications - Biometric authentication poses a significant problem as reconstructed sample or fake self-manufactured samples used by intruders for accessing the actual real...     

An efficient IoT based smart farming system using machine learning algorithms

Multimedia Tools and Applications - This paper suggests an IoT based smart farming system along with an efficient prediction method called WPART based on machine learning techniques to predict crop...     

Random oracle-based anonymous credential system for efficient attributes proof on smart devices

Attributes proof in anonymous credential systems is an effective way to balance security and privacy in user authentication; however, the linear complexity of attributes proof causes the existing anonymous credential systems far away from being practical, especially on resource-limited smart devices. For efficiency considerations, we present a novel pairing-based anonymous credential system which solves the linear complexity of attributes proof based on aggregate signature scheme. We propose two extended signature schemes, BLS+ and BGLS+, to be cryptographical building blocks for constructing anonymous credentials in the random oracle model. Identity-like information of message holder is encoded in a signature in order that the message holder can prove the possession of the input message along with the validity of a signature. We present issuance protocol for anonymous credentials embedding weak attributes which are referred to what cannot identify a user in a population. Users can prove any combination of attributes all at once by aggregating the corresponding individual credentials into one. The attributes proof protocols on AND and OR relation over multiple attributes are also given. The performance analysis shows that the aggregation-based anonymous credential system outperforms both the conventional Camenisch–Lysyanskaya pairing-based system and the accumulator-based system when prove AND and OR relation over multiple attributes, and the size of credential and public parameters are shorter as well.     

Kernel ICA: An alternative formulation and its application to face recognition

This paper formulates independent component analysis (ICA) in the kernel-inducing feature space and develops a two-phase kernel ICA algorithm: whitened kernel principal component analysis (KPCA) plus ICA. KPCA spheres data and makes the data structure become as linearly separable as possible by virtue of an implicit nonlinear mapping determined by kernel. ICA seeks the projection directions in the KPCA whitened space, making the distribution of the projected data as non-gaussian as possible. The experiment using a subset of FERET database indicates that the proposed kernel ICA method significantly outperform ICA, PCA and KPCA in terms of the total recognition rate.     

基于人脸超分的网络视频人脸检测

针对网络视频质量低导致人脸检测准确率低的问题,提出一种基于人脸超分辨率重建的SR Face Detection模型.使用去掉自监督分支且以Resnet50为基础网络的RetinaFace进行帧图片人脸的粗提取;在人脸检测器后增加一个人脸超分辨率重建网络,剔除粗提取人脸中的非人脸.该超分网络的生成网络使用残差密集块进行特征提取,加入注意力损失和热图,更好地还原面部细节;根据实际需求设计一个多判别功能的判别网络.实验结果表明,SR Face Detection模型在WID-ER FACE数据集上取得了令人信服的结果,提高了人脸检测准确率,且人脸检测场景越复杂,效果提升越明显.     

An efficient and low-error mesh simplification method based on torsion detection

To preserve the major characteristics of the simplified model, this study proposes the use of torsion detection to improve the quadric error metric of vertex-pair contraction, and retain the physical features of the models. Besides keeping the physical features of the models, the proposed method also decreases the preprocessing time cost associated with analysis. To verify the conclusion, this research not only presents the effects of simplification and compares them with the vertex-pair contraction, but also employs Metro detection and image comparison to verify the error measurements. The experimental results demonstrate that the proposed method improves the error rate and keeps the precision of the object features efficiently.     

An EM based probabilistic two-dimensional CCA with application to face recognition

Recently, two-dimensional canonical correlation analysis (2DCCA) has been successfully applied for image feature extraction. The method instead of concatenating the columns of the images to the one-dimensional vectors, directly works with two-dimensional image matrices. Although 2DCCA works well in different recognition tasks, it lacks a probabilistic interpretation. In this paper, we present a probabilistic framework for 2DCCA called probabilistic 2DCCA (P2DCCA) and an iterative EM based algorithm for optimizing the parameters. Experimental results on synthetic and real data demonstrate superior performance in loading factor estimation for P2DCCA compared to 2DCCA. For real data, three subsets of AR face database and also the UMIST face database confirm the robustness of the proposed algorithm in face recognition tasks with different illumination conditions, facial expressions, poses and occlusions.     

一种基于像素差值特征的车辆检测方法

为了快速定位监控场景中的车辆位置,提出了一种基于像素差值特征的车辆检测方法。首先提取图像的归一化像素差值特征(NPD),之后使用深度二次树(DQT)学习最优的特征子集,最后使用Ada Boost算法筛选最具区分力的特征构建强分类器。以含有正面、侧面及背面三个角度超过3 500个样本为测试集进行了快速车辆检测测试,并与梯度方向直方图(HOG)和Haar的组合特征进行了对比。对比实验表明,基于NPD的车辆检测方法最优,其检测率和检测时间分别为85.47%和200 ms。     

WLAN based energy efficient smart city design

The rapid growth of new technologies resulted in a new city model, known as the famous “Smart City”. The main aim in this paper is to create a paradigm for building an energy efficient smart city. Wireless local area network (WLAN) controller that will be used by the city will be constructed in such a manner that when there will be no request from any node to the access point (AP), the AP will be send from active mode to sleep mode. In Qualnet7.2, with the help of three types of energy model generic, mica z and mica motes the energy consumption in three modes transmit, receive and sleep mode is analysed, where it is seen that energy consumption in sleep mode is much less than in any other modes. In this paper, we propose an algorithm where it is shown that the energy consumption in sleep mode is less than in any other modes.

     

一种快速高效的人脸检测方法

介绍了一种建立在改进型Adaboost算法基础上的人脸检测方法,整个方法分为训练和检测两个阶段。训练阶段包含提取类Haar_Like矩形特征、利用改进型Adaboost算法生成强分类器、级联强分类器生成人脸检测器三步。检测阶段,采用金字塔式的穷举搜索法将对待检测图像进行人脸检测。为了解决传统Adaboost算法在训练过程中可能出现退化现象的问题,在Adaboost每轮训练中,定义一个阈值HWt,结合样本是否被错误分类以及当前权值是否大于HWt来给样本更新权值,该方法可以避免训练中可能出现的权重分布严重扭曲的退化现象,提高检测效率。经过编程实践,结果证明该方法检测效率高、检测精度较好。     

An efficient spherical mapping algorithm and its application on spherical harmonics

The sphere is a natural and seamless parametric domain for closed genus-0 surfaces. We introduce an efficient hierarchical optimization approach for the computation of spherical parametrization for closed genus-0 surfaces by minimizing a nonlinear energy balancing angle and area distortions. The mapping results are bijective and lowly distorted. Our algorithm converges efficiently and is suitable to manipulate large-scale geometric models. We demonstrate and analyze the effectiveness of our mapping in spherical harmonics decomposition.     

An evidential reasoning based classification algorithm and its application for face recognition with class noise

For classification problems, in practice, real-world data may suffer from two types of noise, attribute noise and class noise. It is the key for improving recognition performance to remove as much of their adverse effects as possible. In this paper, a formalism algorithm is proposed for classification problems with class noise, which is more challenging than those with attribute noise. The proposed formalism algorithm is based on evidential reasoning theory which is a powerful tool to deal with uncertain information in multiple attribute decision analysis and many other areas. Thus, it may be more effective alternative to handle noisy label information. And then a specific algorithm—Evidential Reasoning based Classification algorithm (ERC) is derived to recognize human faces under class noise conditions. The proposed ERC algorithm is extensively evaluated on five publicly available face databases with class noise and yields good performance.     

Weighted maximum scatter difference based feature extraction and its application to face recognition

An improved maximum scatter difference (MSD) algorithm based on weighted scheme is proposed in this paper. The existing MSD model and its improved method only highlight the role which within-class scatter matrix plays while they pay little attention to the action of between-class scatter matrix. Another weakness of the existing MSD model is that it is difficult to select an appropriate weight for within-class scatter matrix because the range of weight is usually too large. In order to make MSD more suitable for classification, different weights are assigned to both between-class and within-class scatter matrices, respectively. This scheme is more convenient for operation than original MSD because it confines the range of parameters to a small range. Finally, the results of experiments conducted on AR and FERET face database indicate the effectiveness of the proposed approach.     

An efficient moving object detection and tracking system based on fractional derivative

Multimedia Tools and Applications - Video shadowing is a blooming system with the intention of conserving the tangible and also capital resources in an organization. Simultaneously, the necessity...     

Evolutionary pursuit and its application to face recognition

Introduces evolutionary pursuit (EP) as an adaptive representation method for image encoding and classification. In analogy to projection pursuit, EP seeks to learn an optimal basis for the dual purpose of data compression and pattern classification. It should increase the generalization ability of the learning machine as a result of seeking the trade-off between minimizing the empirical risk encountered during training and narrowing the confidence interval for reducing the guaranteed risk during testing. It therefore implements strategies characteristic of GA for searching the space of possible solutions to determine the optimal basis. It projects the original data into a lower dimensional whitened principal component analysis (PCA) space. Directed random rotations of the basis vectors in this space are searched by GA where evolution is driven by a fitness function defined by performance accuracy (empirical risk) and class separation (confidence interval). Accuracy indicates the extent to which learning has been successful, while separation gives an indication of expected fitness. The method has been tested on face recognition using a greedy search algorithm. To assess both accuracy and generalization capability, the data includes for each subject images acquired at different times or under different illumination conditions. EP has better recognition performance than PCA (eigenfaces) and better generalization abilities than the Fisher linear discriminant (Fisherfaces)     

An incremental decision tree algorithm based on rough sets and its application in intrusion detection

As we know, learning in real world is interactive, incremental and dynamical in multiple dimensions, where new data could be appeared at anytime from anywhere and of any type. Therefore, incremental learning is of more and more importance in real world data mining scenarios. Decision trees, due to their characteristics, have been widely used for incremental learning. In this paper, we propose a novel incremental decision tree algorithm based on rough set theory. To improve the computation efficiency of our algorithm, when a new instance arrives, according to the given decision tree adaptation strategies, the algorithm will only modify some existing leaf node in the currently active decision tree or add a new leaf node to the tree, which can avoid the high time complexity of the traditional incremental methods for rebuilding decision trees too many times. Moreover, the rough set based attribute reduction method is used to filter out the redundant attributes from the original set of attributes. And we adopt the two basic notions of rough sets: significance of attributes and dependency of attributes, as the heuristic information for the selection of splitting attributes. Finally, we apply the proposed algorithm to intrusion detection. The experimental results demonstrate that our algorithm can provide competitive solutions to incremental learning.