Myoelectric patternrecognition (PR) can provide a more intuitive control for upper limb amputees in using multi-function prosthesis than direct control. Accuracy of a patternrecognition system has been shown to impr...
详细信息
ISBN:
(纸本)9781467319690
Myoelectric patternrecognition (PR) can provide a more intuitive control for upper limb amputees in using multi-function prosthesis than direct control. Accuracy of a patternrecognition system has been shown to improve with increasing number of EMG channels. However, increasing the number of channels comes with a drawback of increased weight, cost and complexity of the prosthesis. this paper presents the concept and design of a novel EMG acquisition system to acquire higher number of channels without increasing the number of electrodes placed or the complexity of the prosthetic device. A prototype of the device was developed and tested on able-bodied subjects to evaluate its performance in patternrecognition. Subjects were requested to perform 9 different hand movements while EMG data was collected into training and test groups. Test results indicate a 15% improvement in classification accuracy withthe new system when compared to conventional systems. A system like this is valuable for patients with higher level amputations where placing higher number of electrodes is not feasible due to limited availability of appropriate muscle sites.
the long-term electromyography (EMG) signal can make significant effect on prosthesis control based on patternrecognition. In this paper, we collected myoelectric signals lasting twelve days and compared the performa...
详细信息
Face is the key component in understanding emotions which play significant roles in many areas from security and entertainment to psychology and education. In this paper, we propose a method to detect facial action un...
详细信息
ISBN:
(纸本)9781479903108
Face is the key component in understanding emotions which play significant roles in many areas from security and entertainment to psychology and education. In this paper, we propose a method to detect facial action units in 3D face data by combining novel geometric properties and a new descriptor based on the Local Binary pattern (LBP) methodology. the proposed method enables person and gender independent facial action unit detection. the decision level fusion is used by employing the Random Forests classifiers to combine geometric and LBP based features. Unlike the previous methods which suffer from the diversity among different persons and normalize features utilizing neutral faces, our method extracts features on a single 3D face data. Besides, we show that orientation based 3D LBP descriptor can be implemented efficiently in terms of size and time without degrading the performance. We tested our method on the Bosphorus database and present comparative results withthe existing methods. Our results outperform those of existing methods, achieving a mean receiver operating characteristic area under curve of 97.7%.
pattern-recognition-based control using surface electromyography (EMG) from the extrinsic hand muscles has shown great promise for providing control of multiple prosthetic functions. However, it is not clear how these...
详细信息
ISBN:
(纸本)9781467319690
pattern-recognition-based control using surface electromyography (EMG) from the extrinsic hand muscles has shown great promise for providing control of multiple prosthetic functions. However, it is not clear how these systems will perform when the user possesses a functional wrist;an attribute unique to the population of partial-hand amputees. Fortunately, partial-hand amputees may have remaining intrinsic hand muscles, from which additional information-rich EMG data may be extracted and used for prosthetic control. We investigated the effect of statically and dynamically varying wrist position on a patternrecognition system's ability to classify hand grasp patterns in able-bodied individuals. We found that varying wrist position significantly degraded the system's performance (p<0.001). the system performed worse when trained only with EMG data from the extrinsic hand muscles than when trained with EMG data from the intrinsic hand muscles. the system's performance significantly improved when trained in all static wrist positions (p<0.001) and with all dynamic wrist motions (p<0.001).
In this paper, we explore the utility of Local Binary pattern (LBP) descriptors and variance measure towards the development of efficient techniques in order to segment a large collection of historical machine printed...
详细信息
Computational Analysis of gene expression data is extremely difficult, due to the existence of a huge number of genes and less number of samples (limited number of patients). thus,it is of significant importance to pr...
详细信息
A natural and intuitive operation of multifunctional upper limb prostheses involves the concurrent activation of multiple degrees of freedom in a proportional way. Several approaches to simultaneous and proportional c...
详细信息
ISBN:
(纸本)9781467319690
A natural and intuitive operation of multifunctional upper limb prostheses involves the concurrent activation of multiple degrees of freedom in a proportional way. Several approaches to simultaneous and proportional control strategies have been investigated;provided outcome measures however were offline accuracy or error rates and lacked the functional component of a preclinical assessment. this study evaluated a simultaneous proportional patternrecognition control strategy with two parallel classifiers in a two-dimensional Fitts' law style test and compared it to a sequential patternrecognition approach. the proposed test allowed for a complete evaluation through different performance metrics such as throughput (TP, bits/sec), path efficiency (PE, %), completion rate (%), overshoot (%) and reaction time (sec). We found that the simultaneous approach presented with numerous advantages with respect to the sequential alternative through significantly higher TP and PE for combined-motion targets (p<0.001) and significantly less overshooting in both combined and discrete targets (p<0.01). For discrete motions, the TP was significantly lower for the simultaneous approach (p<0.001) but PE was similar. there was no difference in either completion rate or reaction time. these results support the potential of simultaneous patternrecognition for the control of multifunctional prostheses and underline the usefulness of a simple functional test in a preclinical framework.
We introduce a new rough set inspired approach to attribute selection. We consider decision systems with attributes specified by means of two layers: 1) general meta-attribute descriptions, and 2) their specific reali...
详细信息
the proceedings contain 105 papers. the topics discussed include: efficient stereo matching using histogram aggregation with multiple slant hypotheses;pose-invariant face recognition with a two-level dynamic programmi...
ISBN:
(纸本)9783642386275
the proceedings contain 105 papers. the topics discussed include: efficient stereo matching using histogram aggregation with multiple slant hypotheses;pose-invariant face recognition with a two-level dynamic programming algorithm;one-shot learning for real-time action recognition;human body segmentation with multi-limb error-correcting output codes detection and graph cuts optimization;modeling pose/appearance relations for improved object localization and pose estimation in 2D images;consensus clustering using partial evidence accumulation;efficient optimization algorithm for space-variant mixture of vector fields;modality combination techniques for continuous sign language recognition;genetic programming of prototypes for pattern classification;bias and variance multi-objective optimization for support vector machines model selection;and hybrid grammar language model for handwritten historical documents recognition.
this paper presents an objective evaluation of previously unexplored biometric techniques utilizing patterns identifiable in human eye movements to distinguish individuals. the distribution of primitive eye movement f...
详细信息
ISBN:
(纸本)9781479903108
this paper presents an objective evaluation of previously unexplored biometric techniques utilizing patterns identifiable in human eye movements to distinguish individuals. the distribution of primitive eye movement features are compared between eye movement recordings using algorithms based on the following statistical tests: the Ansari-Bradley test, the Mann-Whitney U-test, the two-sample Kolmogorov-Smirnov test, the two-sample t-test, and the two-sample Cramer-von Mises test. Score-level information fusion is applied and evaluated by: weighted mean, support vector machine, random forest, and likelihood ratio. the accuracy of each comparison/fusion algorithm is evaluated, with results suggesting that, on high resolution eye tracking equipment, it is possible to obtain equal error rates of 16.5% and rank-1 identification rates of 82.6% using the two-sample Cramer-von Mises test and score-level information fusion by random forest, the highest accuracy results on the considered dataset.
暂无评论