Implementasi Arsitektur MobileNetV2 untuk Klasifikasi Sampah Organik dan Anorganik Berbasis Android
DOI:
https://doi.org/10.29407/hby4zx45Keywords:
android, Deep Learning, Klasifikasi Sampah, MobileNetV2Abstract
Permasalahan penumpukan sampah di Indonesia memerlukan solusi pemilahan yang efektif sejak dari sumbernya. Metode pemilahan manual yang ada saat ini dinilai kurang efisien, sehingga diperlukan pendekatan teknologi otomatis untuk membantu masyarakat mengenali jenis sampah. Penelitian ini bertujuan untuk merancang bangun aplikasi klasifikasi sampah organik dan anorganik berbasis Android menggunakan algoritma Deep Learning. Metode yang diterapkan adalah Transfer Learning menggunakan arsitektur MobileNetV2 yang dipilih karena efisiensi komputasinya pada perangkat mobile. Dataset yang digunakan adalah Waste Classification Data dari Kaggle yang terdiri dari 25.077 citra sampah rumah tangga. Model dilatih selama 10 epoch dan dikonversi menjadi format TensorFlow Lite (.tflite) untuk implementasi on-device. Hasil pengujian menunjukkan bahwa model mampu mencapai tingkat akurasi sebesar 91%, dengan nilai Precision kelas organik sebesar 96% dan Recall kelas anorganik sebesar 97%. Ukuran model yang dihasilkan sangat ringan, yaitu 8.8 MB, sehingga aplikasi dapat berjalan responsif pada perangkat Android standar tanpa membebani memori. Penelitian ini menyimpulkan bahwa MobileNetV2 sangat efektif untuk diterapkan sebagai asisten cerdas pemilahan sampah portabel
Downloads
References
[1] Kementerian Lingkungan Hidup dan Kehutanan (KLHK), "Capaian Kinerja Pengelolaan Sampah," Sistem Informasi Pengelolaan Sampah Nasional (SIPSN), 2024. https://sipsn.kemenlh.go.id/ diakses pada tanggal 22 Desmber 2025
[2] S. S. Alladwar and M. Hajare, "Deep Learning Techniques Using Garbage Classification," International Journal of Research in Engineering, Science and Management, vol. 3, no. 1, 2020.
[3] M. Sandler, A. Howard, M. Zhu, A. Zhmoginov, and L.-C. Chen, "MobileNetV2: Inverted Residuals and Linear Bottlenecks," in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2018, pp. 4510–4520.
[4] R. A. Nugraha, E. W. Hidayat, and R. N. Shofa, "Klasifikasi Jenis Buah Jambu Biji Menggunakan Algoritma Principal Component Analysis dan K-Nearest Neighbor," Generation Journal, vol. 7, no. 1, pp. 1-7, 2023.
[5] S. Sai, "Waste Classification Data," Kaggle, 2019. [Online]. Available: https://www.kaggle.com/datasets/techsash/waste-classification-data 22 Desember 2025.
[6] F. Chollet, "Xception: Deep Learning with Depthwise Separable Convolutions," in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2017, pp. 1251-1258.
[7] D. P. Kingma and J. Ba, "Adam: A Method for Stochastic Optimization," International Conference on Learning Representations (ICLR), 2015.
[8] TensorFlow, "TensorFlow Lite Guide," TensorFlow Documentation, 2024. [Online]. Available: https://www.tensorflow.org/lite/guide. [Diakses: 22-Des-2025].
[9] A. Ignatov et al., "AI Benchmark: Running Deep Neural Networks on Android Smartphones," in Proceedings of the European Conference on Computer Vision (ECCV), 2018.
[10] R. K. Singh, R. Tiwari, and N. Gupta, "Deep Learning Approaches for Garbage Detection and Classification: A Review," Journal of Ambient Intelligence and Humanized Computing, vol. 12, no. 5, pp. 1-20, 2021.
[11] K. He, X. Zhang, S. Ren, and J. Sun, "Deep Residual Learning for Image Recognition," in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2016, pp. 770-778.
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Adam Toyib Nur Wahid, Farrel Ghozy Affifudin, Mufid Aditya, Dihin Muriyatmoko
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
Authors who publish with this journal agree to the following terms:
- Copyright on any article is retained by the author(s).
- The author grants 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 acknowledgment of the work’s authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal’s published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgment of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work.
- The article and any associated published material is distributed under the Creative Commons Attribution-ShareAlike 4.0 International License
