Facial expression recognition based on shape and texture

In this paper, an efficient method for human facial expression recognition is presented. We first propose a representation model for facial expressions, namely the spatially maximum occurrence model (SMOM), which is based on the statistical characteristics of training facial images and has a powerful representation capability. Then the elastic shape-texture matching (ESTM) algorithm is used to measure the similarity between images based on the shape and texture information. By combining SMOM and ESTM, the algorithm, namely SMOM-ESTM, can achieve a higher recognition performance level. The recognition rates of the SMOM-ESTM algorithm based on the AR database and the Yale database are 94.5% and 94.7%, respectively.     

基于纹理和形状的图像相关反馈检索

综合图像的纹理和形状特征进行图像检索时,采用了基于灰度共生矩阵的纹理特征提取 和基于不变矩的形状特征提取方法,并在此基础上加入了基于权重调整的相关反馈机制,使用户可以 参与检索过程,通过调整权重使得检索结果最终满足用户的检索要求。实验表明,这种算法获得了较 好的检索结果。     

基于局部纹理ASM模型的人脸表情识别

针对主动形状模型(ASM)迭代过程容易陷入局部最优解的不足,提出了一种基于局部纹理模型的改进ASM算法,即EWASM.在局部纹理模型构建中,以每个特征点的中垂线方向搜索其邻域信息以确定最佳匹配位置,对衡量匹配程度的马氏距离加以推广,进而得到改进的扩展加权局部纹理模型,它由中心局部纹理模型、前局部纹理模型和后局部纹理模型共3个子模型加权组成,并对加权参数进行实验优化,使各个特征点之间的联系更加紧密,模型的鲁棒性更好.通过表情识别实验对提出的EWASM算法和传统ASM算法进行对比,选用RBF神经网络分类器进行表情分类,实验结果表明EWASM算法收敛速度更快,识别率也得以提高,并解决了局部最小问题,能更有效地表征表情.     

综合颜色、纹理、形状和相关反馈的图像检索*

提出了一种综合颜色、纹理和形状三种特征并进行多层检索的方法;同时,将相关反馈技术融合到算法中,通过调整特征间和特征内的权重来提高检索准确率.实现了一个图像检索原型系统,将不同实验结果进行了比较和分析.实验结果表明,提出的方法具有良好的检索效果.     

Incorporating haptic feedback in simulation for learning physics

The purpose of this study was to investigate the effectiveness of a haptic augmented simulation in learning physics. The results indicate that haptic augmented simulations, both the force and kinesthetic and the purely kinesthetic simulations, were more effective than the equivalent non-haptic simulation in providing perceptual experiences and helping elementary students create multimodal representations of the movements of gears. However, in most cases, force feedback was needed to construct a fully loaded multimodal representation that helps students to comprehend later instruction with less sensory modalities. In addition, the force and kinesthetic simulation was effective in helping to transfer knowledge to new learning situations. These findings suggest that it is important to help elementary students make a solid cognitive grounding with the use of a perceptual anchor.     

Methods for haptic feedback in teleoperated robot-assisted surgery

Teleoperated minimally invasive surgical robots can significantly enhance a surgeon's accuracy, dexterity and visualization. However, current commercially available systems do not include significant haptic (force and tactile) feedback to the operator. This paper describes experiments to characterize this problem, as well as several methods to provide haptic feedback in order to improve surgeon's performance. There exist a variety of sensing and control methods that enable haptic feedback, although a number of practical considerations, e.g. cost, complexity and biocompatibility, present significant challenges. The ability of teleoperated robot-assisted surgical systems to measure and display haptic information leads to a number of additional exciting clinical and scientific opportunities, such as active operator assistance through "virtual fixtures" and the automatic acquisition of tissue properties.     

Usability of optically simulated haptic feedback

In this contribution a method will be described to optically simulate haptic feedback without resorting to mechanical force feedback devices. This method exploits the domination of the visual over the haptic modality. The perception of haptic feedback, usually generated by force feedback devices, was simulated by tiny displacements on the cursor position relative to the intended force. The usability of optically simulated haptic feedback (OSHF) was tested experimentally by measuring effectiveness, efficiency and satisfaction of its users in a Fitts’ type target-acquisition task and comparing the results with the usability of mechanically simulated force feedback and normal feedback. Results show that OSHF outperforms mechanically simulated haptic feedback and normal feedback, especially in the case of small targets.     

An empirical evaluation of extreme learning machine: application to handwritten character recognition

Multimedia Tools and Applications - Extreme learning machine (ELM), a randomized learning paradigm for single hidden layer feed-forward network, has gained significant attention for solving...     

An auto tuned noise resistant descriptor for dynamic texture recognition

Multimedia Tools and Applications - This paper presents a scheme for the representation and recognition of the dynamic texture in the noisy environment. Dynamic texture is the sequence of images of...     

The texture gradient equation for recovering shape from texture

Studies the recovery of shape from texture under perspective projection. We regard shape from texture as a statistical estimation problem, the texture being the realization of a stochastic process. We introduce warplets, which generalize wavelets over the 2D affine group. At fine scales, the warpogram of the image obeys a transport equation, called texture gradient equation. In order to recover the 3D shape of the surface, one must estimate the deformation gradient, which measures metric changes in the image. This is made possible by imposing a notion of homogeneity for the original texture, according to which the deformation gradient is equal to the velocity of the texture gradient equation. By measuring the warplet transform of the image at different scales, we obtain a deformation gradient estimator     

Multi-modal virtual environments for education with haptic and olfactory feedback

It has been suggested that immersive virtual reality (VR) technology allows knowledge-building experiences and in this way provides an alternative educational process. Important key features of constructivist educational computer-based environments for science teaching and learning, include interaction, size, transduction and reification. Indeed, multi-sensory VR technology suits very well the needs of sciences that require a higher level of visualization and interaction. Haptics that refers to physical interactions with virtual environments (VEs) may be coupled with other sensory modalities such as vision and audition but are hardly ever associated with other feedback channels, such as olfactory feedback. A survey of theory and existing VEs including haptic or olfactory feedback, especially in the field of education is provided. Our multi-modal human-scale VE VIREPSE (virtual reality platform for simulation and experimentation) that provides haptic interaction using a string-based interface called SPIDAR (space interface device for artificial reality), olfactory and auditory feedbacks is described. An application that allows students experiencing the abstract concept of the Bohr atomic model and the quantization of the energy levels has been developed. Different configurations that support interaction, size and reification through the use of immersive and multi-modal (visual, haptic, auditory and olfactory) feedback are proposed for further evaluation. Haptic interaction is achieved using different techniques ranging from desktop pseudo-haptic feedback to human-scale haptic interaction. Olfactory information is provided using different fan-based olfactory displays (ODs). Significance of developing such multi-modal VEs for education is discussed.     

Optimizing low-order controllers for haptic systems under delayed feedback

In this paper, a PD controller design for haptic systems under delayed feedback is considered. More precisely, a complete stability analysis of a haptic system where local dynamics are described by some second-order mechanical dynamics is presented. Next, using two optimization techniques ($H_{\infty }$ and stability, margin optimization) an optimal choice for the controller gains is proposed. The derived results are tested on a three degree-of-freedom real-time experimental platform to illustrate the theoretical results.     

A teleoperation system for micro positioning with haptic feedback

This paper presents the research work on a 1 Degree of Freedom (DOF) macro-micro teleoperation system which enables human operator to perform complex task in micro environment such as cell insertion with the capability of haptic feedback. To reach submicron resolution, a nano-motion piezo actuator was used as the slave robot and a servo DC motor was used as the master robot. Force sensors were implemented at both ends for haptic feedback and a microscope equipped with camera was employed for real-time visual feedback. The hysteresis nonlinearity of the piezo motor was modeled using LuGre friction model and compensated for. A Sliding Mode Based Impedance Controller (SMBIC) was designed at the slave side to ensure position tracking while an impedance force controller was designed at the master side to ascertain tracking of the force. Control parameters were chosen based on Llewellyn stability criteria such that the entire system stays stable against parameter uncertainties and constant time delay. The experimental results demonstrated capability of the proposed control frameworks in desirable tracking of the position and force signals while the entire system remained stable. The results of this study can be used for complex tasks in micron environment such as cell insertion.     

An empirical evaluation of kernels for time series

There exist a variety of distance measures which operate on time series kernels. The objective of this article is to compare those distance measures in a support vector machine setting. A support vector machine is a state-of-the-art classifier for static (non-time series) datasets and usually outperforms k-Nearest Neighbour, however it is often noted that that 1-NN DTW is a robust baseline for time-series classification. Through a collection of experiments we determine that the most effective distance measure is Dynamic Time Warping and the most effective classifier is kNN. However, a surprising result is that the pairing of kNN and DTW is not the most effective model. Instead we have discovered via experimentation that Dynamic Time Warping paired with the Gaussian Support Vector Machine is the most accurate time series classifier. Finally, with good reason we recommend a slightly inferior (in terms of accuracy) model Time Warp Edit Distance paired with the Gaussian Support Vector Machine as it has a better theoretical basis. We also discuss the reduction in computational cost achieved by using a Support Vector Machine, finding that the Negative Kernel paired with the Dynamic Time Warping distance produces the greatest reduction in computational cost.

     

An investigation on the impact of auditory and haptic feedback on rhythmic walking interactions

This paper presents a system capable of rhythmic walking interaction by auditory and haptic display. Likewise, it summarizes the results of the research on the influence of the audio and haptic stimulation on rhythmic walking interaction. The system detects user’s footsteps and either provides interactive real-time feedback or suggests a pace using a synthetic walking sound or vibration. This pace is either a constant tempo or adapts to the walker. Auditory and haptic feedback signals are either ecological physically-based synthetic walking signals or simple sinusoidal beeps. In the experiment, the different auditory and haptic feedback and interaction modes are studied with respect to their effect on the walking tempo. The results show that participants synchronise equally well with the tempo with either audio or haptic cues, but indicate the audio–haptic conditions as the easiest to synchronise with. Moreover, results indicate that multimodal audio–haptic feedback provide the most natural feeling. These results have implications on the design of interactive entertainment or therapeutical applications.     

Neural network models of haptic shape perception

Three different models of tactile shape perception inspired by the human haptic system were tested using an 8 d.o.f. robot hand with 45 tactile sensors. One model is based on the tensor product of different proprioceptive and tactile signals and a self-organizing map (SOM). The two other models replace the tensor product operation with a novel self-organizing neural network, the Tensor-Multiple Peak Self-Organizing Map (T-MPSOM). The two T-MPSOM models differ in the procedure employed to calculate the neural activation. The computational models were trained and tested with a set of objects consisting of hard spheres, blocks and cylinders. All the models learned to map different shapes to different areas of the SOM, and the tensor product model as well as one of the T-MPSOM models also learned to discriminate individual test objects.     

An empirical evaluation of weak mutation

Mutation testing is a fault-based technique for unit-level software testing. Weak mutation was proposed as a way to reduce the expense of mutation testing. Unfortunately, weak mutation is also expected to provide a weaker test of the software than mutation testing does. This paper presents results from an implementation of weak mutation, which we used to evaluate the effectiveness versus the efficiency of weak mutation. Additionally, we examined several options in an attempt to find the most appropriate way to implement weak mutation. Our results indicate that weak mutation can be applied in a manner that is almost as effective as mutation testing, and with significant computational savings     

Workspace analysis for haptic feedback manipulator in virtual cockpit system

To obtain natural space experience of haptic interaction for users in virtual cockpit systems (VCS), a haptic feedback system and a workspace analysis framework for haptic feedback manipulator (HFM) are presented in this paper. Firstly, improving the classical three-dimensional workspace obtained by the Monte Carlo method, a novel workspace representation method, oriented workspace, is presented, which can indicate both the position and the orientation of the end-effector. Then, aimed at the characters of HFMs, the oriented workspace is divided into the effective workspace and the prohibited area by extracting the control panel area. At last, the effective workspace volume and the control panel area are calculated by the double-directed extremum method, with the accuracy improved by repeatedly adding and extracting boundary points. By simulation, the area in which interactions between the manipulator and users hand performed is determined and accordingly the effective workspace along with its boundary and volume are obtained in a relative high precision, which lay a basis for haptic interaction in VCS.     

An empirical evaluation of XQuery processors

This paper presents an extensive and detailed experimental evaluation of XQuery processors. The study consists of running five publicly available XQuery benchmarks—the Michigan benchmark (MBench), XBench, XMach-1, XMark and X007—on six XQuery processors, three stand-alone (file-based) XQuery processors (Galax, Qizx/Open, Saxon-B) and three XML/XQuery database systems (BerkeleyDB/XML, MonetDB/XQuery, X-Hive/DB). Next to assessing and comparing the functionality, performance and scalability for the various systems, the major focus of this work is to report in detail about the experiences made while performing such an exhaustive study, to discuss all the problems that we encountered and how we solved them, and hence to hopefully provide some guidelines (or even a recipe) for performing reproducible large-scale experimental research and system evaluation.