Hand Gesture Recognition Research Based on Surface EMG Sensors and 2D-accelerometers

Chen, Xiang; Zhang, Xu; Zhao, Zhong-Qiu; Yang, Jihai; Lantz, Vuokko; Wang, Kongqiao

doi:10.1109/iswc.2007.4373769

Cited by 96 publications

(64 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…While research has been conducted on the use of lower dimensional waveform features for classification, such as Mean Absolute Value [5,6] or Fast Fourier Transform [10], this study has shown that a combination of procedural time-domain techniques including filtering, segmentation and temporal scaling can also be effective. Despite this result, it should be noted that further training and testing is required for subjects with movement disorders to be able to properly gauge the potential for this system.…”

Section: Discussionmentioning

confidence: 99%

“…A multi-layer feed forward artificial neural network (ML-FFNN) was used to classify processed samples: (5) where the activation function is bipolar logistic . Supervised training was done through backpropagated stochastic gradient descent: (6) where is the Least Squared Error (LSE) cost function:…”

Section: Classificationmentioning

confidence: 99%

“…Previous attempts at accelerometer and EMG based gesture classification have been made using Fischer Discriminant Analysis (FDA) [3] and Bayesian Networks (BN) [4,5] with reported accuracies between 85-96%. One study [5] explores the fusion of both accelerometer and EMG sensors, but reports only minor improvements over pure EMG.…”

Section: Introductionmentioning

confidence: 99%

“…One study [5] explores the fusion of both accelerometer and EMG sensors, but reports only minor improvements over pure EMG.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

A system for accelerometer-based gesture classification using artificial neural networks

Stephenson

Naik

Chai

2017

2017 39th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC)

View full text Add to dashboard Cite

Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works. 4Abstract-A great many people suffer from neurological movement disorders that render typical hardware interface devices ineffective. A need exists for a universal interface device that can be trained to accept a wide range of inputs across varying types and severities of movement disorders. In this regard, this paper details the design, testing and optimization of an accelerometer-based gesture identification system. A Bluetooth-enabled IMU mounted on the wrist provides hand motion trajectory information to a local terminal. Several techniques are applied to decrease the intra-class variance and reduce classifier complexity including filtering, segmentation and temporal scaling. Datasets consisted of 520 training samples, 260 validation samples and a further 520 testing samples. A multi-layer feed forward artificial neural network (ML-FFNN) was used to classify the input space into 26 different classes. Initial system accuracy, using arbitrary hyperparameters was 77.69% with final optimized accuracy at 99.42%.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Classificationmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

“…One study [5] explores the fusion of both accelerometer and EMG sensors, but reports only minor improvements over pure EMG.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

A system for accelerometer-based gesture classification using artificial neural networks

Stephenson

Naik

Chai

2017

2017 39th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC)

View full text Add to dashboard Cite

show abstract

“…Recently, Chen et al [2] reported that the combination of EMG sensors and accelerometers achieved 5-10% improvement in the recognition accuracies for various wrist and finger gestures. They used two 2-axis accelerometers and two surface EMG sensors that are attached at the single arm.…”

Section: Related Workmentioning

confidence: 99%