ANALISA PERFORMA ARSITEKTUR MOBILENETV1 DAN RESNET MENGGUNAKAN META-LEARNING DALAM MENDETEKSI OBJEK HEWAN KUCING

Authors

  • Faiz Octa Reynaldi Sekolah Tinggi Manajemen Informatika dan Komputer Nusa Mandiri
  • Omar Pahlevi
  • Indah Suryani

DOI:

https://doi.org/10.21927/ijubi.v4i1.1686

Keywords:

Object Detection, Transfer Learning, Cloud Computing, Few-Shot Learning, Hewan Kucing

Abstract

Object Detection memiliki beberapa kendala saat proses training seperti banyaknya data yang harus dilatih, menggunakan waktu cukup lama untuk dilatih dan lain-lain. Pada penelitian ini, peneliti melakukan komparasi akurasi dan average loss training arsitektur SSD MobileNetV1 dan SSD ResNet menggunakan Pre-Trained model dengan metode Few-Shot Learning menggunakan Hold-Out Cross Validation untuk mendeteksi Objek Hewan Kucing Hitam dan Objek Hewan Kucing Putih dengan pengambilan data secara rill dari metode observasi Jakarta Vet Shop dan hanya membutuhkan sedikit data untuk dilakukannya proses training. Penelitian ini dilakukan dengan cara menggunakan Cloud Computing seperti Google Colab sebagai media untuk membandingkan akurasi arsitektur SSD MobileNetV1 dan SSD ResNet. Hasil analisa dalam penelitian ini adalah SSD ResNet memiliki akurasi yang tinggi dengan nilai rata-rata 100% pada kucing hitam dan nilai rata-rata 97.9% pada kucing putih sementara untuk SSD MobileNetV1 memiliki nilai rata-rata 99.66666667% pada kucing hitam dan 78.733% pada kucing putih. Kemudian SD MobileNetV1 memiliki Train Loss lebih besar dengan nilai rata-rata 0.003923 pada Kucing Hitam dan nilai rata-rata 0.0059 Kucing Putih jika dibandingkan dengan SSD ResNet dengan nilai rata-rata 0.030263 pada Kucing Hitam dan nilai rata-rata 0.00413 pada Kucing Putih. 

References

E. Wijaya, “Analisis Penggunaan Algoritma Breadth First Search Dalam Konsep Artificial Intellegencia,†Time, vol. II, no. 2, pp. 18–26, 2013.

A. P. Mahardika, “Sistem Pakar Mendeteksi Penyakit Dalam dengan Metode Backward Chaining Menggunakan Visual Basic 2010,†pp. 23–34, 2010.

A. Syaikhoni & A. Ariyadi, “Deteksi Objek Dengan Tensorflow Object Detection Api,†Binus, 2016. https://mti.binus.ac.id/2018/12/26/deteksi-objek-dengan-tensorflow-object-detection-api/.

J. Deng, X. Li, and Y. Fang, “Few-shot Object Detection on Remote Sensing Images,†arXiv, pp. 1–12, 2020.

A. Wibowo, “10 Fold-Cross Validation,†2017,https://mti.binus.ac.id/2017/11/24/10-fold-cross-validation/ (accessed Jul. 29, 2019).

K. Zhang, M. Sun, T. X. Han, X. Yuan, L. Guo, and T. Liu, “Residual Networks of Residual Networks: Multilevel Residual Networks,†IEEE Trans. Circuits Syst. Video Technol., vol. 28, no. 6, pp. 1303–1314, 2018, doi: 10.1109/TCSVT.2017.2654543.

P. R. Aningtiyas, A. Sumin, and S. Wirawan, “Pembuatan Aplikasi Deteksi Objek Menggunakan TensorFlow Object Detection API dengan Memanfaatkan SSD MobileNet V2 Sebagai Model Pra - Terlatih,†J. Ilm. Komputasi, vol. 19, no. 3, pp. 421–430, 2020, doi: 10.32409/jikstik.19.3.68.

F. E. Ramadhan, “Penerapan Image Classification Dengan Pre-Trained Model Mobilenet Dalam Client-Side Machine Learning,†2020.

F. G. Mohammadi, M. H. Amini, and H. R. Arabnia, “An Introduction to Advanced Machine Learning : Meta Learning Algorithms, Applications and Promises,†Adv. Intell. Syst. Comput., vol. 1123, pp. 129–144, 2020, doi: 10.1007/978-3-030-34094-0_6.

M. J. Garbade, “Understanding few-shot learning in machine learning,†Medium, vol. 28, no. 2. pp. 1–43, 2018, [Online]. Available: https://medium.com/quick-code/understanding-few-shot-learning-in-machine-learning-bede251a0f67.

P. Chyan, “Penerapan Image Enhancement Algorithm Untuk Meningkatkan Kualitas Citra Tak Bergerak,†Maj. Ilm. INTI, vol. 12, no. 2, pp. 278–281, 2017.

E. M. Learners, “Mengenal Machine Learning,†Medium.Com. 2019, [Online].Available:https://medium.com/evolve-machine-learners/mengenal-machine-learning-6c4a48db48b0.

T. Hospedales, A. Antoniou, P. Micaelli, and A. Storkey, “Meta-Learning in Neural Networks : A Survey,†pp. 1–20, 2020.

M. A. Pangestu and H. Bunyamin, “Analisis Performa dan Pengembangan Sistem Deteksi Ras Anjing pada Gambar dengan Menggunakan Pre-Trained CNN Model,†J. Tek. Inform. dan Sist. Inf., vol. 4, no. 2, pp. 337–344, 2018.

“Mengenal Google Colab - Structilmy.†[Online].Available:https://structilmy.com/2019/05/mengenal-google-colab/.

K. He, X. Zhang, S. Ren, and J. Sun, “Deep residual learning for image recognition,†Proc. IEEE Comput. Soc. Conf. Comput. Vis. Pattern Recognit., vol. 2016-Decem, pp. 770–778, 2016, doi: 10.1109/CVPR.2016.90.

Y. Wang, Q. Yao, J. T. Kwok, and L. M. Ni, “Generalizing from a few examples: A survey on few-shot learning,†arXiv, vol. 1, no. 1, pp. 1–34, 2019.

T. Lin, C. L. Zitnick, and P. Doll, “Microsoft COCO : Common Objects in Context,†pp. 1–15.

T. Beysolow, Applied Natural Language Processing with Python Learning and Deep Learning Language Processing. 2018.

Downloads

Published

2021-06-27

Issue

Vol. 4 No. 1 (2021): Indonesian Journal of Business Intelligence (IJUBI)

Section

Articles

License

COPYRIGHT TRANSFER FORM

The copyright to this article is transferred to Alma Ata University Press if and when the article is accepted for publication. The undersigned hereby transfers any and all rights in and to the paper including without limitation all copyrights to AAU Press. The undersigned hereby represents and warrants that the paper is original and that he/she is the author of the paper, except for material that is clearly identified as to its original source, with permission notices from the copyright owners where required. The undersigned represents that he/she has the power and authority to make and execute this assignment.

We declare that:

1. This paper has not been published in the same form elsewhere.

2. It will not be submitted anywhere else for publication prior to acceptance/rejection by this Journal.

3. A copyright permission is obtained for materials published elsewhere and which require this permission for reproduction.

Furthermore, I/We hereby transfer the unlimited rights of publication of the above mentioned paper in whole to AAU Press. The copyright transfer covers the exclusive right to reproduce and distribute the article, including reprints, translations, photographic reproductions, microform, electronic form (offline, online) or any other reproductions of similar nature.

The corresponding author signs for and accepts responsibility for releasing this material on behalf of any and all co-authors. This agreement is to be signed by at least one of the authors who have obtained the assent of the co-author(s) where applicable. After submission of this agreement signed by the corresponding author, changes of authorship or in the order of the authors listed will not be accepted.

Retained Rights/Terms and Conditions

 

1. Authors retain all proprietary rights in any process, procedure, or article of manufacture described in the Work.

2. Authors may reproduce or authorize others to reproduce the Work or derivative works for the authors personal use or for company use, provided that the source and the AAU Press copyright notice are indicated, the copies are not used in any way that implies AAU Press endorsement of a product or service of any employer, and the copies themselves are not offered for sale.

3. Although authors are permitted to re-use all or portions of the Work in other works, this does not include granting third-party requests for reprinting, republishing, or other types of re-use.