Perbandingan Arsitektur LeNet dan AlexNet Pada Metode Convolutional Neural Network Untuk Pengenalan American Sign Language
DOI:
https://doi.org/10.35143/jkt.v7i1.4489Keywords:
Classification, ASL, CNN, Lenet, AlexNetAbstract
American Sign Language (ASL) is a sign language used to communicate for deaf people. The method used to identify ASL is Convolutional Neural Network (CNN). The architecture used by LeNet and AlexNet. The results of each architecture are then compared. The research was conducted with 2 schemes of the amount of data used, namely the first scheme of 100 data per letter and the second scheme of 1,000 data per letter to test the performance of the two architectures. The research results after being tested with new data, the first scheme for the LeNet architecture produces an overall accuracy of 48.332% and the AlexNet architecture produces an overall accuracy of 32.584%. The second scheme for the LeNet architecture produces an overall accuracy of 92.468% and the AlexNet architecture produces an overall accuracy of 91.618%. Overall comparison can be said that the LeNet architecture is the best architecture in this study.Downloads
References
Shivashankara and Srinath, “American Sign Language Recognition System: An Optimal Approach,†Int. J. Image, Graph. Signal Process., vol. 10, no. 8, pp. 18–30, 2018, doi: 10.5815/ijigsp.2018.08.03.
T. Hunt, J. Carper, T. Lasley, C. Raisch, and E. Drasgow, “American Sign Language,†Encycl. Educ. Reform Dissent, 2013, doi: 10.4135/9781412957403.n31.
M. E. Al Rivan, H. Irsyad, K. Kevin, and A. T. Narta, “Pengenalan Alfabet American Sign Language Menggunakan K-Nearest Neighbors Dengan Ekstraksi Fitur Histogram Of Oriented Gradients,†J. Tek. Inform. dan Sist. Inf., vol. 5, no. 3, pp. 328–339, 2020, doi: 10.28932/jutisi.v5i3.1936.
I. Fareza, R. Busdin, M. E. Al Rivan, and H. Irsyad, “Pengenalan Alfabet Bahasa Isyarat Amerika Menggunakan Edge Oriented Histogram dan Image Matching,†J. Tek. Inform. dan Sist. Inf., vol. 4, no. 1, pp. 82–92, 2018, doi: 10.28932/jutisi.v4i1.747.
M. Ezar, A. Rivan, and M. T. Noviardy, “Klasifikasi American Sign Language Menggunakan Ekstraksi Fitur Histogram of Oriented Gradients dan Jaringan Syaraf Tiruan,†vol. 6, pp. 442–451, 2020.
M. Zufar and B. Setiyono, “Convolutional Neural Networks Untuk Pengenalan Wajah Secara Real-Time,†J. Sains dan Seni ITS, vol. 5, no. 2, p. 128862, 2016, doi: 10.12962/j23373520.v5i2.18854.
M. Bagus, S. Bakti, and Y. M. Pranoto, “Pengenalan Angka Sistem Isyarat Bahasa Indonesia Dengan Menggunakan Metode Convolutional Neural Network,†pp. 11–16, 2019.
M. Swapna, Y. K. Sharma, and B. M. G. Prasadh, “CNN Architectures: Alex Net, Le Net, VGG, Google Net, Res Net,†Int. J. Recent Technol. Eng., vol. 8, no. 6, pp. 953–960, 2020, doi: 10.35940/ijrte.f9532.038620.
S. Kulkarni and S. Harnoorkar, “Comparative Analysis of CNN Architectures,†vol. 7, no. June 6, pp. 1459–1464, 2020.
M. Kayed, A. Anter, and H. Mohamed, “Classification of Garments from Fashion MNIST Dataset Using CNN LeNet-5 Architecture,†Proc. 2020 Int. Conf. Innov. Trends Commun. Comput. Eng. ITCE 2020, no. February, pp. 238–243, 2020, doi: 10.1109/ITCE48509.2020.9047776.
S. Ilahiyah and A. Nilogiri, “Implementasi Deep Learning Pada Identifikasi Jenis Tumbuhan Berdasarkan Citra Daun Menggunakan Convolutional Neural Network,†JUSTINDO (Jurnal Sist. dan Teknol. Inf. Indones., vol. 3, no. 2, pp. 49–56, 2018, doi: 10.32528/JUSTINDO.V3I2.2254.
V. Maeda-Gutiérrez et al., “Comparison of convolutional neural network architectures for classification of tomato plant diseases,†Appl. Sci., vol. 10, no. 4, 2020, doi: 10.3390/app10041245.
Akash, “ASL Alphabet,†https://www.kaggle.com/grassknoted/asl-alphabet/, 2018.
Downloads
Published
Issue
Section
License
Copyright (c) 2021 Jurnal Komputer TerapanÂ
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
Copyright info for authors
1. Authors hold the copyright in any process, procedure, or article described in the work and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
2. Authors retain publishing rights to re-use all or portion of the work in different work but can not granting third-party requests for reprinting and republishing the work.
3. Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) as it can lead to productive exchanges, as well as earlier and greater citation of published work.
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License 
