IRDC-Net: An Inception Network with a Residual Module and Dilated Convolution for Sign Language Recognition Based on Surface Electromyography
Uloženo v:
| Název: | IRDC-Net: An Inception Network with a Residual Module and Dilated Convolution for Sign Language Recognition Based on Surface Electromyography |
|---|---|
| Autoři: | Xiangrui Wang, Lu Tang, Qibin Zheng, Xilin Yang, Zhiyuan Lu |
| Zdroj: | Sensors, Vol 23, Iss 13, p 5775 (2023) |
| Informace o vydavateli: | MDPI AG |
| Rok vydání: | 2023 |
| Sbírka: | Directory of Open Access Journals: DOAJ Articles |
| Témata: | sign language recognition, surface electromyogram, inception network, residual module, dilated convolution, Chemical technology, TP1-1185 |
| Popis: | Deaf and hearing-impaired people always face communication barriers. Non-invasive surface electromyography (sEMG) sensor-based sign language recognition (SLR) technology can help them to better integrate into social life. Since the traditional tandem convolutional neural network (CNN) structure used in most CNN-based studies inadequately captures the features of the input data, we propose a novel inception architecture with a residual module and dilated convolution (IRDC-net) to enlarge the receptive fields and enrich the feature maps, applying it to SLR tasks for the first time. This work first transformed the time domain signal into a time–frequency domain using discrete Fourier transformation. Second, an IRDC-net was constructed to recognize ten Chinese sign language signs. Third, the tandem CNN networks VGG-net and ResNet-18 were compared with our proposed parallel structure network, IRDC-net. Finally, the public dataset Ninapro DB1 was utilized to verify the generalization performance of the IRDC-net. The results showed that after transforming the time domain sEMG signal into the time–frequency domain, the classification accuracy (acc) increased from 84.29% to 91.70% when using the IRDC-net on our sign language dataset. Furthermore, for the time–frequency information of the public dataset Ninapro DB1, the classification accuracy reached 89.82%; this value is higher than that achieved in other recent studies. As such, our findings contribute to research into SLR tasks and to improving deaf and hearing-impaired people’s daily lives. |
| Druh dokumentu: | article in journal/newspaper |
| Jazyk: | English |
| Relation: | https://www.mdpi.com/1424-8220/23/13/5775; https://doaj.org/toc/1424-8220; https://doaj.org/article/3b2b3373faec4f0bbd5c30335b2b2465 |
| DOI: | 10.3390/s23135775 |
| Dostupnost: | https://doi.org/10.3390/s23135775 https://doaj.org/article/3b2b3373faec4f0bbd5c30335b2b2465 |
| Přístupové číslo: | edsbas.2E34B0F2 |
| Databáze: | BASE |
Nájsť tento článok vo Web of Science | Abstrakt: | Deaf and hearing-impaired people always face communication barriers. Non-invasive surface electromyography (sEMG) sensor-based sign language recognition (SLR) technology can help them to better integrate into social life. Since the traditional tandem convolutional neural network (CNN) structure used in most CNN-based studies inadequately captures the features of the input data, we propose a novel inception architecture with a residual module and dilated convolution (IRDC-net) to enlarge the receptive fields and enrich the feature maps, applying it to SLR tasks for the first time. This work first transformed the time domain signal into a time–frequency domain using discrete Fourier transformation. Second, an IRDC-net was constructed to recognize ten Chinese sign language signs. Third, the tandem CNN networks VGG-net and ResNet-18 were compared with our proposed parallel structure network, IRDC-net. Finally, the public dataset Ninapro DB1 was utilized to verify the generalization performance of the IRDC-net. The results showed that after transforming the time domain sEMG signal into the time–frequency domain, the classification accuracy (acc) increased from 84.29% to 91.70% when using the IRDC-net on our sign language dataset. Furthermore, for the time–frequency information of the public dataset Ninapro DB1, the classification accuracy reached 89.82%; this value is higher than that achieved in other recent studies. As such, our findings contribute to research into SLR tasks and to improving deaf and hearing-impaired people’s daily lives. |
|---|---|
| DOI: | 10.3390/s23135775 |