HEGM: A hierarchical elastic graph matching for hand gesture recognition

A hierarchical scheme for elastic graph matching applied to hand gesture recognition is proposed. The proposed algorithm exploits the relative discriminatory capabilities of visual features scattered on the images, assigning the corresponding weights to each feature. A boosting algorithm is used to determine the structure of the hierarchy of a given graph. The graph is expressed by annotating the nodes of interest over the target object to form a bunch graph. Three annotation techniques, manual, semi-automatic, and automatic annotation are used to determine the position of the nodes. The scheme and the annotation approaches are applied to explore the hand gesture recognition performance. A number of filter banks are applied to hand gestures images to investigate the effect of using different feature representation approaches. Experimental results show that the hierarchical elastic graph matching (HEGM) approach classified the hand posture with a gesture recognition accuracy of 99.85% when visual features were extracted by utilizing the Histogram of Oriented Gradient (HOG) representation. The results also provide the performance measures from the aspect of recognition accuracy to matching benefits, node positions correlation and consistency on three annotation approaches, showing that the semi-automatic annotation method is more efficient and accurate than the other two methods.     

Telerobot-enabled HUB-CI model for collaborative lifecycle management of design and prototyping

The HUB-CI model is investigated in a telerobotic system in a client/server network to manage the lifecycle of engineering design and prototyping. The purpose of this platform is to support collaborative engineering design and proof of concept to enhance distributed team collaboration and resource utilization. The suggested platform is exemplified in two collaboration support tools and a physical prototyping platform. Structured Co-Insight Management is developed to support innovative idea exchanges and the consensus decision-making during the design process. Conflict/error detection management helps preventing conflicts and errors during the lifecycle of design and development. Physical collaboration over the network occurs when a team controls the telerobot operation during prototyping and testing in design cycles. A pilot system is implemented with a group project for the design of an electronic circuit (including both hardware and software designs). The functional assessment method is used to compare this platform to other collaborative design tools. The system presented offers unique qualitative advantages as an integrated collaboration support system.     

Straightforward modeling of dynamic I–V characteristics for microwave FETs

This work presents a straightforward approach aimed at modeling the dynamic I–V characteristics of microwave active solid‐state devices. The drain‐source current generator represents the most significant source of nonlinearity in a transistor and, therefore, its correct modeling is fundamental to predict accurately the current and voltage waveforms under large‐signal operation. The proposed approach relies on using a small set of low‐frequency time‐domain waveform measurements combined with numerical optimization‐based estimation of the nonlinear model parameters. The procedure is applied to a gallium nitride HEMT and silicon FinFET. The effectiveness of the modeling procedure in terms of prediction accuracy and generalization capability is demonstrated by validation of the extracted models under operating conditions different than the ones used for the parameters estimation. Good agreement between measurements and model simulations is achieved for both technologies and in both low‐ and high‐frequency range. © 2013 Wiley Periodicals, Inc. Int J RF and Microwave CAE 24:109–116, 2014.     

Smoothing: A natural way to detect contour features

This work presents a dominant point detector. The angles of the contour are characterized through local entropy produced by a rotationally symmetric smoothing. The proposed scheme uses a punctual multi-scale approach in which only the candidates are analyzed in higher scales. To preserve the angle-entropy relationship in higher scales, we propose a smoothing kernel which presents special features that ensure its steepness in every scale. It is built from the sum of two Gaussians with different openings resembling center-surround receptive fields. The outputs of the proposed method are confronted to a ground-truth found in the literature, and to popular boundary based corner detectors that used the same set of images. Results reveal that the proposed detector performs extremely well.     

A multipurpose audio aggregation watermarking based on multistage vector quantization

Watermarking technology can achieve multipurpose such as copyright protection, copy protection, and integrity authentication. We propose a novel watermarking approach which involves robust watermark and fragile watermark in a two-stage quantization technique. Traditional watermarking algorithms mostly have poor performance in imperceptibility as the codeword selected from the modified codebook is not optimal. In our method, we select the codeword in the original codebook to ensure optimality. Furthermore, we use Huffman encoding to pick up property which is sensitive to many attacks in the entire aggregation. For copyright protection, the proposed scheme can resist attacks such as lossy compression, noise addition, and normalization. As for content authentication, the proposed scheme is sensitive to various attacks provided by Stirmark Benchmark for Audio. Experimental results show that the proposed method can be used respectively for protecting the copyright and authenticating the integrity of the audio aggregation.     

Using self-organizing maps with complex network topologies and coalitions for time series prediction

A Self-organizing Map (SOM) is a competitive learning neural network architecture that make available a certain amount of classificatory neurons, which self-organize spatially based on input patterns. In this paper we explore the use of complex network topologies, like small-world, scale-free or random networks; for connecting the neurons within a SOM, and apply them for Time Series Prediction. We follow the VQTAM model for function prediction, and consider several benchmarks to evaluate the quality of the predictions. Afterwards, we introduce the CASOM algorithm (Coalitions and SOM) that uses coalitions to temporarily extend the neighborhood of a neuron, and to provide more accuracy in prediction problems than classical SOM. The results presented in this work suggest that the most regular the network topology is, the better results it provides in prediction. Besides, we have found that not updating all the neurons at the same time provides much better results. Finally, we describe how the use of coalitions can enhance the capability of SOM for Time Series Prediction.     

Binary and m-ary encoding in applications of tree-based genetic algorithms for QoS routing

Ubiquitous and Pervasive Computing (UPC) applications often have Quality of Service (QoS) requirements. These become constraints for the UPC network infrastructure. In this paper, we refer to Mobile ad Hoc Networks, one of the most important technologies supporting UPC, and investigate on Genetic Algorithms (GAs) for QoS routing. GAs are part of the soft computing paradigm and can solve the NP search of QoS routes with multiple constraints. We elaborate on tree-based GAs, which represent the set of paths from source to destination as a tree and encode them through the crossed junctions. While their most well-known applications use m-ary encoding representing single paths in the chromosomes, in this paper we discuss a binary encoding with the objective of improving the convergence speed. The binary encoding represents classes of paths in the chromosomes and allows local search on classes of paths. These classes are both collectively exhaustive and mutually exclusive. Simulation results compare convergence speed and scalability of GA applications with binary and m-ary encoding in networks with an increasing number of nodes and links per node. As the per-class processing is reason of additional computational cost, an hybrid GA application that uses both binary and m-ary encoding is introduced.     

基于A-VFSA粒子发生器的PSO算法的研究

对超快速模拟退火算法进行改进(A-VFSA),并以此为粒子发生器,提出了PSO-PG算法.该算法通过粒子发生器对搜索域内粒子进行改进,从而生成一个精英粒子池;并根据随机策略从粒子池中选择粒子,作为初始种群,采用PSO算法进行优化,得到全局最优解.相比于标准PSO算法和LDW算法,PSO-PG算法拥有更好的稳定性和优化精度,能够更加快速地收敛到全局最优解,在一定范围内几乎不依赖于初始参数的选择.     

快速鲁棒特征算法的CUDA加速优化

提出一种基于统一计算设备架构(Compute Unified Device Architecture,CUDA)的快速鲁棒特征(Speed-up Robust Feature,SURF)图像匹配算法。分析了SURF算法的并行性,在图像处理单元(Graphics Processing Unit,GPU)的线程映射和内存模型方面对算法的构建尺度空间、特征点提取、特征点主方向的确定、特征描述子的生成及特征匹配5个步骤进行CUDA加速优化。实验表明,相比适用于CPU的SURF算法,文中提出的适用于GPU的SURF算法在处理30MB的图片时性能提高了33倍。适用于GPU的SURF算法拓展了SURF算法在遥感等领域的快速应用,尤其是大影像的快速配准。     

改进的DV-Hop算法在节点定位中的应用研究

基于传统DV-Hop算法的定位技术,通过分析该算法误差产生的主要原因,提出了节点间跳数的修正方法。通过在广播通信中采用双通信频率,将节点间的跳数缩减为0.5 R,减少了平均跳距的误差。通过Matlab软件的实验验证,本算法可以提高节点定位精度和网络通信频率,可进一步提高定位精度,定位结果达到了预期的目的。