Sign language recognition using the Extreme Learning Machine

Sole, Mosiuoa M.; Tsoeu, Mohohlo

doi:10.1109/afrcon.2011.6072114

Cited by 16 publications

(4 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Each gesture was mapped onto the corresponding alphabet letter by comparing three different classifiers, namely support vector machines, Mahalanobis distances and Euclidean distances, obtaining an accuracy of >90%. A complete set of 26 characters were classified, with an accuracy of 95%, by means of a glove equipped with fourteen RFSs and a neural network classifier, in [207]. A set of 31 static gestures were recognized, with a classification accuracy ranging between 77.42% and 99.61%, from a South African sign language dataset, adopting the Levenberg-Marquardt training algorithm and neural network classificatory [158].…”

Section: Gesture Recognition Assistance and Interpretationmentioning

confidence: 99%

Resistive flex sensors: a survey

Saggio

Riillo

Sbernini

et al. 2015

Smart Mater. Struct.

140

View full text Add to dashboard Cite

Resistive flex sensors can be used to measure bending or flexing with relatively little effort and a relativelylow budget. Their lightness, compactness, robustness, measurement effectiveness and low power consumption make these sensors useful for manifold applications in diverse fields.Here, we provide a comprehensive survey of resistive flex sensors, taking into account their working principles, manufacturing aspects, electrical characteristics and equivalent models, useful front-end conditioning circuitry, and physic-bio-chemical aspects. Particular effort is devoted to reporting on and analyzing several applications of resistive flex sensors, related to the measurement of body position and motion, and to the implementation of artificial devices. In relation to the human body, we consider the utilization of resistive flex sensors for the measurement of physical activity and for the development of interaction/interface devices driven by human gestures. Concerning artificial devices, we deal with applications related to the automotive field, robots, orthosis and prosthesis, musical instruments and measuring tools. The presented literature is collected from different sources, including bibliographic databases, company press releases, patents, master's theses and PhD theses.

show abstract

Section: Gesture Recognition Assistance and Interpretationmentioning

confidence: 99%

Resistive flex sensors: a survey

Saggio

Riillo

Sbernini

et al. 2015

Smart Mater. Struct.

140

View full text Add to dashboard Cite

show abstract

“…As a result, the system had lower accuracy than the trained system. Sole et al used Extreme Learning Machine (ELC) algorithm to classify the Auslan (Australian Sign Language) alphabet [8]. The reason they chose ELC because it is a simplified version of the Neural Network (NN) [8].…”

Section: Cyber Glovesmentioning

confidence: 99%

American Sign Language Recognition system by using surface EMG signal

Savur

Sahin

2016

2016 IEEE International Conference on Systems, Man, and Cybernetics (SMC)

View full text Add to dashboard Cite

Sign Language Recognition (SLR) system is a novel method that allows hard of hearing people to communicate with society. In this study, an American Sign Language (ASL) recognition system was proposed by using the surface Electromyography (sEMG). The objective of this study is to recognize the American Sign Language alphabet letters and allow users to spell words and sentences. For this purpose, sEMG signals are acquired from subject's right forearm for 27 American Sign Language gestures, 26 English alphabet letters, and one for home position. Time domain, frequency domain (band power), power spectral density (band power), and average power features were used as the feature extraction methods. After feature extraction, Principal Component Analysis (PCA) was applied to obtain uncorrelated features. As a classification method, Support Vector Machine and Ensemble Learning algorithm were used and their performances were compared with tabulated results. In conclusion, the results of this study show that sEMG signal can be used for SLR systems. vi List of Contributions • Implementation of a new input method for users who know American Sign Language. • Design of a smart phone application that allows hard of hearing people to communicate with hearing people. • Comparison of results from the Electromyography signal collected for Sign Language Recognition System using four different feature extraction methods: Time domain features, frequency fomain band powers, power spectral density(PSD) band power, and average power in each channel in combination with two different classification metods: Support Vector Machine (SVM) and Ensemble Learning (Bagged Trees).

show abstract

“…The ELM-based artificial neural network was chosen by the present study due to its capacity to process real-time pattern recognition in applications [9], as well as the fact that it is used in gesture recognition systems [36] [35]. In [37], the ELM technique was compared to other machine learning techniques to recognize language when applied to LIBRAS sign language and demonstrated recognition rates very similar to state of the art results, except up to 5 times faster, as presented in Table 1 Training [37].…”

Section: System Architecture Using Elm Neural Networkmentioning

confidence: 99%