Perbandingan MobileNetV2, DenseNet121, InceptionV3, dan Xception pada Klasifikasi Citra Panel Surya Bersih dan Berdebu

Aswin Mulyo Nugroho; Hindayati Mustafidah; Maulida Ayu Fitriani; Supriyono Supriyono

doi:10.30865/jurikom.v12i4.8688

Authors

Aswin Mulyo Nugroho Universitas Muhammadiyah Purwokerto, Purwokerto
Hindayati Mustafidah Universitas Muhammadiyah Purwokerto, Purwokerto
Maulida Ayu Fitriani Universitas Muhammadiyah Purwokerto, Purwokerto
Supriyono Supriyono Universitas Muhammadiyah Purwokerto, Purwokerto

DOI:

https://doi.org/10.30865/jurikom.v12i4.8688

Keywords:

Solar Panel, CNN, Image Classification, Pre-Trained Models, Transfer Learning

Abstract

The buildup of dust on solar panels can greatly diminish energy output, lower system efficiency, and raise operational expenses. A productive way to tackle this problem is to utilize image classification through Convolutional Neural Network (CNN) techniques. This study examines the classification capabilities of four CNN models, namely MobileNetV2, DenseNet121, InceptionV3, and Xception, using transfer learning. These models leverage pre-trained weights from large datasets such as ImageNet to accelerate convergence and improve generalization. The dataset of images utilized in this research is obtained from Kaggle and includes pictures of both clean and dusty solar panels. The dataset was divided into training, validation, and testing subsets using a stratified approach to ensure balanced class distribution across all subsets. During training, class weighting was used to address potential class imbalance. The models were developed using TensorFlow with multi-GPU support, optimized using the AdamW optimizer, and fine-tuned to enhance performance. Model evaluation was conducted using accuracy, precision, recall, and F1-score metrics. Among all the architectures evaluated, the Xception model achieved the best performance with an accuracy of 90.52%, outperforming MobileNetV2 with an accuracy of 87.92%, DenseNet121 with 89.78%, and InceptionV3 which achieved 87.73%. These results indicate that modern CNN-based models can effectively recognize relevant visual patterns to detect dust on solar panels.

References

F. Ferdyson, and J. Windarta, "Overview Pemanfaatan dan Perkembangan Sumber Daya Energi Surya Sebagai Energi Terbarukan di Indonesia," Jurnal Energi Baru dan Terbarukan, vol. 4, no. 1, pp. 1-6, 2023, doi: 10.14710/jebt.2023.15714.

J. Tanesab, D. Parlevliet, J. Whale, and T. Urmee, “Dust Effect and its Economic Analysis on PV Modules Deployed in a Temperate Climate Zone,” Energy Procedia, vol. 100, pp. 65–68, 2016, doi: 10.1016/j.egypro.2016.10.154.

E. A. Pratama and R. N. Ahmad, “Analisis Pengaruh Polutan Debu Terhadap Pembangkitan Daya Pada Sistem Pembangkit Listrik Tenaga Surya,” vol. 3, pp. 1–19, 2024. [Online]. Available: https://eprints.ums.ac.id/id/eprint/121708.

I. G. Perwati, N. Suarna, and T. Suprapti, “Analisis Klasifikasi Gambar Bunga Lily Menggunakan Metode Convolutional Neural Network (CNN) Dalam Pengolahan Citra,” JATI (Jurnal Mahasiswa Teknik Informatika), vol. 8, no. 3, pp. 2908–2915, 2024, doi: 10.36040/jati.v8i3.9193.

D. C. Febrianto, G. Wibisono, M. A. Fitriani, and M. Afrad, “Transfer Learning model Convolutional Neural Network menggunakan VGG-16 untuk Klasifikasi Tumor Otak pada Citra Hasil MRI,” LEDGER: Journal Informatic and Information Technology, vol. 3, no. 1, pp. 11–18, 2024, doi: 10.20895/ledger.v3i1.1387.

A. Andreansyah and J. Supardi, “VGG-16 Accuracy Optimization for Fingerprint Pattern Imager Classification,” Jurnal Sisfokom (Sistem Informasi dan Komputer), vol. 14, no. 1, pp. 42–48, 2025, doi: 10.32736/sisfokom.v14i1.2317.

A. E. Putra, M. F. Naufal, and V. R. Prasetyo, “Klasifikasi Jenis Rempah Menggunakan Convolutional Neural Network dan Transfer Learning,” Jurnal Edukasi dan Penelitian Informatika, vol. 9, no. 1, p. 12, 2023, doi: 10.26418/jp.v9i1.58186.

N. W. Kencana, R. Umar, and Murinto, “Implementasi Transfer Learning Untuk Klasifikasi Jenis Ras Ayam,” Jurnal Informatika Polinema, vol. 11, no. 2, pp. 147–154, 2025, doi: 10.33795/jip.v11i2.6469.

G. E. P. Purba, S. H. Wijoyo, and N. Y. Setiawan, “Pengaruh Transfer Learning ResNet dan DenseNet terhadap Performa Klasifikasi Ekspresi Wajah Menggunakan Dataset FER-2013,” Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer, vol. 8, no. 7, 2024. [Online]. Available: https://j-ptiik.ub.ac.id/index.php/j-ptiik/article/view/13931.

A. W. Kosman, Y. Wahyuningsih, and F. Mahendrasusila, “Pengujian Metode Inception V3 dalam Mengidentifikasi Penyakit Kanker Kulit,” Jurnal Teknlogi Informatika dan Komputer, vol. 10, no. 1, pp. 132–142, 2024, doi: 10.37012/jtik.v10i1.1940.

Darmatasia, “Deteksi Penggunaan Masker Menggunakan Xception Transfer Learning,” Jurnal Informatika Sains dan Teknologi, vol. 5, no. 2, pp. 279–288, 2020, doi: 10.24252/instek.v5i2.20132.

G. Cipriani, A. D’Amico, S. Guarino, D. Manno, M. Traverso, and V. Di Dio, “Convolutional neural network for dust and hotspot classification in PV modules,” Energies, vol. 13, no. 23, 2020, doi: 10.3390/en13236357.

Y. Shao, C. Zhang, L. Xing, H. Sun, Q. Zhao, and L. Zhang, “A new dust detection method for photovoltaic panel surface based on Pytorch and its economic benefit analysis,” Energy and AI, vol. 16, p. 100349, 2024, doi: 10.1016/j.egyai.2024.100349.

A. M. Alatwi, H. Albalawi, A. Wadood, H. Anwar, and H. M. El-Hageen, “Deep Learning-Based Dust Detection on Solar Panels: A Low-Cost Sustainable Solution for Increased Solar Power Generation,” Sustainability, vol. 16, no. 19, pp. 1–16, 2024, doi: 10.3390/su16198664.

F. UYSAL, “Dust Detection on Solar Photovoltaic Panels Used in Optoelectronics with Convolutional Neural Network-Based Deep Learning Models”, IJCESEN, vol. 11, no. 1, 2025.

M. S. H. Onim, Z. M. M. Sakif, A. Ahnaf, A. Kabir, A. K. Azad, A. M. T. Oo, R. Afreen, S. T. Hridy, M. Hossain, T. Jabid, and M. S. Ali, "SolNet: A convolutional neural network for detecting dust on solar panels," Energies, vol. 16, no. 1, p. 155, 2023, doi: 10.3390/en16010155.

M. Mamdouh and Y. A. Zaghloul, “Fusion Between Image Processing and Machine Learning for Dust Detection on Solar Panels,” in 2024 Fifteenth International Conference on Ubiquitous and Future Networks (ICUFN), IEEE, 2024, pp. 169–174. doi: 10.1109/ICUFN61752.2024.10625232.

H. M. A. Mohammed and A. E. Ba Alawi, "CASolarNet: Channel Attention EfficientNet-based model for solar panel dust detection,” in 2024 4th International Conference on Emerging Smart Technologies and Applications (eSmarTA), IEEE, 2024, pp. 1–4. doi: 10.1109/eSmarTA62850.2024.10638876.

M. Shah, M. Joshi, P. Patel, N. Mevada, R. Baria, and M. Chauhan, "Improving solar power generation with InceptionV3 dust detection on the solar panel energy systems," in 2023 Third International Conference on Ubiquitous Computing and Intelligent Information Systems (ICUIS), IEEE, 2023, pp. 448–456. doi: 10.1109/ICUIS60567.2023.00080.

B. Wijayanto, R. M. Mahendra, M. I. Salam, and T. Informatika, “Identifikasi Jenis Ikan Cupang Menggunakan Metode CNN Dengan Arsitektur MobileNetV2 Berbasis Mobile,” Seminar Nasional Teknologi & Sains, vol. 4, pp. 519–525, 2025, doi: 10.29407/8k2m1068.

R. J. H. Butar-Butar and N. L. Marpaung, “Deep Learning untuk Identifikasi Daun Tanaman Obat Menggunakan Transfer Learning MobileNetV2,” Jurnal Informatika: Jurnal Pengembangan IT, vol. 8, no. 2, pp. 142–148, 2023, doi: 10.30591/jpit.v8i2.5217.

T. S. Arulananth, S. W. Prakash, R. K. Ayyasamy, V. P. Kavitha, P. G. Kuppusamy, and P. Chinnasamy, “Classification of Paediatric Pneumonia Using Modified DenseNet-121 Deep-Learning Model,” IEEE Access, vol. 12, pp. 35716–35727, 2024, doi: 10.1109/ACCESS.2024.3371151.

C. Nisa’, E. Y. Puspaningrum, and H. Maulana, “Penerapan Metode Convolutional Neural Network untuk Klasifikasi Penyakit Daun Apel pada Imbalanced Data,” Seminar Nasional Informatika Bela Negara, vol. 1, pp. 169–175, 2020, doi: 10.33005/santika.v1i0.46.

D. P. Pamungkas and M. F. Amrulloh, “Analisis Hasil Klasifikasi Penyakit Daun Bawang Merah Menggunakan CNN Arsitektur Exception,” Jurnal Ilmiah Penelitian dan Pembelajaran Informatika, vol. 10, no. 1, pp. 359–366, 2025, doi: 10.29100/jipi.v10i1.5875.

F. Hafifah, S. Rahman, and S. Asih, “Klasifikasi Jenis Kendaraan Pada Jalan Raya Menggunakan Metode Convolutional Neural Networks (CNN),” TIN (Terapan Informatika Nusantara), vol. 2, no. 5, pp. 292–301, 2021. [Online]. Available: https://ejurnal.seminar-id.com/index.php/tin/ article/view/915.

R. Triginandri and E. R. Subhiyakto, “Deteksi Dini Cacar Monyet menggunakan Convolutional Neural Network (CNN) dalam Aplikasi Mobile,” Edumatic: Jurnal Pendidikan Informatika, vol. 8, no. 2, pp. 516–525, 2024, doi: 10.29408/edumatic.v8i2.27625.

Sutarti and F. Syaqialloh, “Klasifikasi dan Pengenalan Emosi dari Ekspresi Wajah Menggunakan CNN-BiLSTM dengan Teknik Data Augmentation”, Decode :Jurnal Pendidikan Teknologi Informasi, vol. 5, no. 1, pp. 79–91, 2025, doi: doi.org/10.51454/decode.v5i1.1038.

M. A. Leonardi and A. Y. Chandra, “Analisis Perbandingan CNN dan Vision Transformer untuk Klasifikasi Biji Kopi Hasil Sangrai,” Jurnal Media Informatika Budidarma, vol. 8, no. 3, p. 1398, 2024, doi: 10.30865/mib.v8i3.7732.