PERBANDINGAN TEKNIK OPTIMASI CNN UNTUK KLASIFIKASI GAMBAR SAMPAH BERBASIS DATASET KECIL DAN TIDAK SEIMBANG

Siti Huzzaemah Andzani; Martanto; Yudhistira Arie Wijaya; Ahmad Faqih

Authors

Siti Huzzaemah Andzani STMIK IKMI, Indonesia
Martanto STMIK IKMI Cirebon, Indonesia
Yudhistira Arie Wijaya STMIK IKMI Cirebon, Indonesia
Ahmad Faqih STMIK IKMI Cirebon, Indonesia

Abstract

Penanganan sampah rumah tangga merupakan salah satu tantangan penting dalam mendukung keberlanjutan lingkungan. Salah satu pendekatan modern untuk mengatasinya adalah pemanfaatan teknologi image classification dengan model Convolutional Neural Network (CNN). Namun, dataset sampah sering kali memiliki ukuran terbatas dan ketidakseimbangan kelas (class imbalance), sehingga berdampak pada akurasi model. Penelitian ini bertujuan untuk mengevaluasi performa model CNN dalam mengklasifikasikan gambar sampah rumah tangga dengan empat skenario pelatihan: baseline, data augmentation, weighted loss, dan kombinasi keduanya. Hasil eksperimen menunjukkan bahwa penerapan data augmentation memberikan peningkatan performa paling signifikan, dengan accuracy 82,81% dan F1-score 82,59%. Sementara itu, weighted loss terbukti efektif dalam meningkatkan recall, tetapi tidak menghasilkan performa keseluruhan yang lebih baik dibandingkan augmentation. Temuan ini menegaskan bahwa teknik augmentation merupakan strategi yang lebih optimal dalam kondisi dataset kecil dan tidak seimbang. Penelitian ini memberikan kontribusi dalam pemanfaatan metode optimasi pelatihan CNN untuk aplikasi klasifikasi sampah rumah tangga.

References

Abubakar, I. R., & al., et. (2022). Environmental Sustainability Impacts of Solid Waste Management Practices in the Global South. International Journal of Environmental Research and Public Health, 19(19), 12717.

Alom, M. Z., Taha, T. M., Yakopcic, C., Westberg, S., Sidike, P., Nasrin, M. S., & Asari, V. K. (2021). Review of deep learning: concepts, CNN architectures, challenges, applications and future directions. Journal of Big Data, 8, 53. https://doi.org/10.1186/s40537-021-00444-8

Chen, W. (2024). A survey on imbalanced learning: latest research and future challenges. Artificial Intelligence Review. https://doi.org/10.1007/s10462-024-10759-6

Dablain, D., Corizzo, R., & Japkowicz, N. (2024). Understanding CNN fragility when learning with imbalanced image data. Machine Learning. https://doi.org/10.1007/s10994-023-06326-9

Ferdinan, S., & al., et. (2022). Household Waste Control Index towards Sustainable Waste Management. Sustainability, 14(21), 14403.

Hellín, C. J., & al., et. (2024). Unraveling the Impact of Class Imbalance on Deep Learning Models for Medical Image Classification. Applied Sciences, 14(8), 3419.

Krichen, M. (2023). Convolutional Neural Networks: A Survey. Computers, 12(8), 151. https://doi.org/10.3390/2073-431X/12/8/151

Malik, M. (2022). Waste classification for sustainable development using image recognition with deep learning neural network models. Sustainability, 14(12), 7222. https://doi.org/10.3390/su14127222

Martianto, D., & al., et. (2024). The Quantity and Composition of Household Food Waste: Implications for Policy. PLOS ONE, 19(6), e0305087.

Nnamoko, N., & Barrowclough, J. (2022). Solid Waste Image Classification Using Deep Convolutional Neural Network. Infrastructures, 7(4), 47. https://doi.org/10.3390/infrastructures7040047

Purwono, P. (2023). Understanding of Convolutional Neural Network (CNN): A Review. International Journal of Robotics and Control Systems, 2(4), 888. https://doi.org/10.31763/ijrcs.v2i4.888

Resolute, P. (2024). Applying Cultural Perspective in Indonesia Municipal Solid Waste Management Process. Waste Management & Research, 42(10), 873–881.

Reza, A., & Ma, Y. (2024). A Data Augmentation Methodology to Reduce the Class Imbalance in Histopathology Images. Journal of Imaging Informatics in Medicine, 37, 1767–1782.

Scholz, D., & al., et. (2024). Imbalance-Aware Loss Functions Improve Medical Image Classification. Proceedings of The 7th International Conference on Medical Imaging with Deep Learning, 1341–1356.

Song, Z. (2024). An improved weighted cross-entropy-based convolutional neural network for auxiliary diagnosis of pneumonia. Electronics, 13(15), 2929. https://doi.org/10.3390/electronics13152929

Turn, J., Cordova-Esparza, D.-M., & Romero-González, J.-A. (2025). A comprehensive survey of loss functions and metrics in deep learning. Artificial Intelligence Review, 58, 195. https://doi.org/10.1007/s10462-025-11198-7

Wei, S., & al., et. (2023). An Efficient Data Augmentation Method for Automatic Modulation Recognition from Low-Data Imbalanced-Class Regime. Applied Sciences, 13(5), 3177.

Zhao, X. (2024). A review of convolutional neural networks in computer vision. Artificial Intelligence Review. https://doi.org/10.1007/s10462-024-10721-6

Zheng, Y., & Gu, J. (2021). A class-weighted convolutional neural network for garbage image classification under data imbalance conditions. Waste Management, 120, 680–688.

Zhong, Y., Zhou, W., & Wang, Z. (2025). A survey of data augmentation in domain generalization. Neural Processing Letters, 57, 34. https://doi.org/10.1007/s11063-025-11747-9

Zhong, Z., Zheng, L., Kang, G., Li, S., & Yang, Y. (2025). Random erasing data augmentation revisited: A survey and new insights. Knowledge-Based Systems.

Zou, R., & Wang, N. (2024). Imbalanced data parameter optimization of convolutional neural networks based on analysis of variance. Applied Sciences, 14(19), 9071. https://doi.org/10.3390/app14199071