%0 Conference Paper %1 oliveira2019recognizing %A Oliveira, Anne de Souza %A Filho, Cicero Ferreira Fernandes Costa %A Costa, Marly Guimarães Fernandes %B XXVI Brazilian Congress on Biomedical Engineering %C Singapore %D 2019 %E Costa-Felix, Rodrigo %E Machado, João Carlos %E Alvarenga, André Victor %I Springer %K artificial-neural-network brazilian-sign-language brazilian-sign-language-recognition bsl bslr convolutional-neural-network deaf-community gesture-recognition hand-configuration libras real sign-language sign-language-recognition slr %P 421-427 %R 10.1007/978-981-13-2517-5_64 %T Recognizing Hand Configurations of Brazilian Sign Language Using Convolutional Neural Networks %U https://doi.org/10.1007/978-981-13-2517-5_64 %X This paper proposes evaluating three convolutional neural network architectures for recognizing hand configurations of the Brazilian Sign Language (Libras). To improve the generalization of neural networks, two techniques were employed: dropout and L2 regularization. A proprietary database consisting of 12.200 depth images, captured with the Kinect® sensor was used. Two hundred images were captured for each one of 61 Hand Configurations (HC) of Libras. The training and testing subsets were compounded using an interleave technique. An accuracy of 98% was achieved. This value is better than previous results obtained, with the same dataset, using the k-Nearest Neighbor (kNN) and Novelty classifiers, 95.41% and 96.31%, respectively. %@ 978-981-13-2517-5

@inproceedings{oliveira2019recognizing, abstract = {This paper proposes evaluating three convolutional neural network architectures for recognizing hand configurations of the Brazilian Sign Language (Libras). To improve the generalization of neural networks, two techniques were employed: dropout and L2 regularization. A proprietary database consisting of 12.200 depth images, captured with the Kinect® sensor was used. Two hundred images were captured for each one of 61 Hand Configurations (HC) of Libras. The training and testing subsets were compounded using an interleave technique. An accuracy of 98% was achieved. This value is better than previous results obtained, with the same dataset, using the k-Nearest Neighbor (kNN) and Novelty classifiers, 95.41% and 96.31%, respectively.}, added-at = {2019-09-15T05:58:17.000+0200}, address = {Singapore}, author = {Oliveira, Anne de Souza and Filho, Cicero Ferreira Fernandes Costa and Costa, Marly Guimarães Fernandes}, biburl = {https://www.bibsonomy.org/bibtex/2bf8adb08a380cdb5ddeb9f878d7930c7/jpmor}, booktitle = {XXVI Brazilian Congress on Biomedical Engineering}, doi = {10.1007/978-981-13-2517-5_64}, editor = {Costa-Felix, Rodrigo and Machado, João Carlos and Alvarenga, André Victor}, interhash = {0a4d274a6197ae6e3cf40d2ae42dc910}, intrahash = {bf8adb08a380cdb5ddeb9f878d7930c7}, isbn = {978-981-13-2517-5}, keywords = {artificial-neural-network brazilian-sign-language brazilian-sign-language-recognition bsl bslr convolutional-neural-network deaf-community gesture-recognition hand-configuration libras real sign-language sign-language-recognition slr}, language = {English}, month = {05}, pages = {421-427}, publisher = {Springer}, school = {Universidade Federal do Amazonas (UFAM)}, timestamp = {2020-10-07T13:36:50.000+0200}, title = {Recognizing Hand Configurations of Brazilian Sign Language Using Convolutional Neural Networks}, url = {https://doi.org/10.1007/978-981-13-2517-5_64}, year = 2019 }