Penerapan Transfer Learning Menggunakan Arsitektur Deep CNN ResNet-50 untuk Klasifikasi Jenis Sampah
DOI:
https://doi.org/10.22441/jitkom.v10i1.009Keywords:
Transfer Learning, ResNet-50, Citra, Jenis sampah, KlasifikasiAbstract
Pengelolaan sampah masih menjadi tantangan besar di Indonesia akibat tingginya timbulan sampah dan rendahnya tingkat daur ulang. Salah satu solusi inovatif yang dapat diterapkan adalah sistem klasifikasi sampah otomatis berbasis citra. Penelitian ini bertujuan membangun model klasifikasi jenis sampah menggunakan arsitektur Deep Convolutional Neural Network ResNet-50 dengan pendekatan transfer learning. Dataset yang digunakan berasal dari Kaggle dan mencakup lima kelas: cardboard, glass, metal, paper, dan plastic. Data telah melalui tahap pra-pemrosesan, augmentasi, dan pembagian secara stratifikasi. Model dilatih selama 50 epoch menggunakan optimizer AdamW dan fungsi kehilangan CrossEntropyLoss. Hasil pelatihan menunjukkan akurasi validasi tertinggi sebesar 85%. Evaluasi dengan confusion matrix dan classification report menunjukkan model mampu melakukan klasifikasi secara akurat dan seimbang pada tiap kelas. Sistem ini juga dilengkapi antarmuka interaktif untuk prediksi gambar tunggal, memperlihatkan potensi penerapan nyata dalam pengelolaan sampah berbasis teknologi. Hasil penelitian membuktikan bahwa transfer learning dengan ResNet-50 efektif untuk klasifikasi citra sampah dan berpotensi mendukung solusi lingkungan berbasis kecerdasan buatan
References
[1] H. Ahmed Khan, S. S. Naqvi, A. A. K. Alharbi, S. Alotaibi, and M. Alkhathami, “Enhancing trash classification in smart cities using federated deep learning,” Scientific Reports, vol. 14, no. 1, May 2024, doi: 10.1038/s41598-024-62003-4.
[2] A. Bustamin, B. Zaman, and F. K. Hakim, “Sistem Multi Klasifikasi Sampah Anorganik Dengan Menggunakan Transfer Learning,” Jurnal INSYPRO (Information System and Processing), vol. 8, no. 2, 2023, doi: 10.24252/insypro.v8i2.42385.
[3] C. E. Nainggolan, M. Nasir, Fatoni, and D. Udariansyah, “Perbandingan Klasifikasi Jenis Sampah Menggunakan Convolutional Neural Network Dengan Arsitektur ResNet18 dan ResNet50,” CSRID (Computer Science Research and Its Development Journal), vol. 16, no. 1, pp. 76–90, Mar. 2024, doi: 10.22303/csrid-.16.1.2024.76-90.
[4] J. N. Nuariputri, M. Maimunah, and P. S. Sukmasetya, “Klasifikasi Jenis Sampah Menggunakan Base ResNet-50”, jikstik, vol. 22, no. 3, pp. 379–386, Oct. 2023, doi: 10.32409/jikstik.22.3.3380
[5] L. Li, R. Wang, M. Zou, F. Guo, and Y. Ren, “Enhanced ResNet-50 for garbage classification: Feature fusion and depth-separable convolutions,” PLOS ONE, vol. 20, no. 1, p. e0317999, 2025, doi: 10.1371/journal.pone.0317999.
[6] L. Z. Nurmia, A. Andriayatno, F. Fitriani, and R. Munandar, “Meningkatkan Hasil Belajar Matematika Siswa Menggunakan Media Interaktif Powtoon Kelas V SDN 30 Ampenan,” Lambda: Jurnal Ilmiah Pendidikan MIPA dan Aplikasinya, vol. 5, no. 1, pp. 1–8, Feb. 2025, doi: 10.58218/lambda.v5i1.1104.
[7] W. Muhammad Ardana and Kusrini Kusrini, “Optimasi Algoritma Convolutional Neural Network dengan Arsitektur Efficientnet-B0 dan Resnet-50 untuk Klasifikasi Jenis Sampah: Optimization of Convolutional Neural Network Algorithm with Efficientnet-B0 and Resnet-50 Architecture for Waste Type Classification,” MALCOM: Indonesian Journal of Machine Learning and Computer Science, vol. 5, no. 4, pp. 1274–1286, 2025, doi: 10.57152/malcom.v5i4.2030.
[8] X. Ma, Z. Li, and Z. Lei, “An Improved ResNet-50 for Garbage Image Classification,” Tehnički vjesnik, vol. 29, no. 5, pp. 1552–1559, Oct. 2022, doi: 10.17559/TV-20220420124810.
[9] S. Kundu, M. Sharma, and A. S. Pillai, “AI-Powered Trash Classification System for Lakes and Water Bodies Using Transfer Learning,” 2024 Third International Conference on Power, Control and Computing Technologies (ICPC2T), pp. 163–167, Jan. 2024, doi: 10.1109/icpc2t60072.2024.10474611.
[10] F. Wu and H. Lin, “Effect of transfer learning on the performance of VGGNet-16 and ResNet-50 for the classification of organic and residual waste,” Frontiers in Environmental Science, vol. 10, Oct. 2022, doi: 10.3389/fenvs.2022.1043843
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2026 Jurnal Ilmu Teknik dan Komputer
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
The copyright to this article is transferred to Universitas Mercu Buana (UMB) 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 UMB. 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 this paper has not been published in the same form elsewhere.
Furthermore, I/We hereby transfer the unlimited rights of publication of the above-mentioned paper as a whole to UMB. The copyright transfer covers the 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
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.
Our Articles are licensed under CC BY-NC
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
