Perbandingan Arsitektur MobileNetV2 dan EfficientNet-B0 untuk Klasifikasi Tingkat Kematangan Buah Pepaya berdasarkan Citra RGB
Abstract
Manual assessment of papaya ripeness was prone to inconsistency because visual differences between ripeness stages could be affected by color similarity, lighting, and transitional fruit characteristics. This study compared MobileNetV2 and EfficientNet-B0 for classifying papaya ripeness from RGB images by integrating dataset deduplication and model validation. The initial dataset contained 500 images in three classes: mature, partially mature, and immature. After duplicate images were removed using SHA-256, 246 unique images were obtained. The models were trained using transfer learning with ImageNet weights, stratified data splitting, and five-fold stratified cross-validation. Testing on the deduplicated dataset showed that MobileNetV2 achieved 95.92% accuracy, while EfficientNet-B0 achieved 93.88%. MobileNetV2 was more efficient in terms of model size and inference time, whereas EfficientNet-B0 was more stable in cross-validation. These findings showed that dataset deduplication, efficiency evaluation, and visual interpretation produced a more representative model assessment.
Keywords: Papaya ripeness classification; Dataset deduplication; Transfer learning; MobileNetV2; EfficientNet-B0
Abstrak
Penilaian manual terhadap kematangan buah pepaya dapat menghasilkan ketidakkonsistenan karena perbedaan visual antartingkat kematangan dipengaruhi oleh kemiripan warna, pencahayaan, dan karakter buah pada fase transisi. Penelitian ini membandingkan MobileNetV2 dan EfficientNet-B0 untuk klasifikasi tingkat kematangan buah pepaya berbasis citra RGB dengan mengintegrasikan deduplikasi dataset dan validasi model. Dataset awal terdiri dari 500 citra dalam tiga kelas, yaitu mature, partially mature, dan immature. Setelah citra duplikat dihapus menggunakan SHA-256, diperoleh 246 citra unik. Model dilatih menggunakan transfer learning dengan bobot ImageNet, stratified split, dan 5-fold stratified cross-validation. Hasil pengujian pada dataset hasil deduplikasi menunjukkan MobileNetV2 memperoleh akurasi 95,92%, sedangkan EfficientNet-B0 memperoleh 93,88%. MobileNetV2 lebih efisien dari sisi ukuran model dan waktu inferensi, sedangkan EfficientNet-B0 lebih stabil pada validasi silang. Temuan ini menunjukkan bahwa deduplikasi dataset, evaluasi efisiensi, dan interpretasi visual menghasilkan penilaian model yang lebih representatif.
Keywords
References
N. P. Sutrisna et al., “Deteksi Tingkat Kematangan Buah Pepaya menggunakan Model Convolutional Neural Network,” J. Teknol. Inf. dan Ilmu Komput., vol. 11, no. 3, pp. 569–578, Jul. 2024, doi: 10.25126/jtiik.938119.
M. Rizzo, M. Marcuzzo, A. Zangari, A. Gasparetto, and A. Albarelli, “Fruit ripeness classification: A survey,” Artif. Intell. Agric., vol. 7, pp. 44–57, Mar. 2023, doi: 10.1016/j.aiia.2023.02.004.
K. K. Thyagharajan and G. Kalaiarasi, “A Review on Near-Duplicate Detection of Images Using Computer Vision Techniques,” Arch. Comput. Methods Eng., vol. 28, no. 3, pp. 897–916, May 2021, doi: 10.1007/s11831-020-09400-w.
X. Li, L. Chang, and X. Liu, “CE-Dedup: Cost-Effective Convolutional Neural Nets Training based on Image Deduplication,” in 2021 IEEE Intl Conf on Parallel & Distributed Processing with Applications, Big Data & Cloud Computing, Sustainable Computing & Communications, Social Computing & Networking (ISPA/BDCloud/SocialCom/SustainCom), New York City: IEEE, Sep. 2021, pp. 11–18. doi: 10.1109/ISPA-BDCloud-SocialCom-SustainCom52081.2021.00017.
I. E. Tampu, A. Eklund, and N. Haj-Hosseini, “Inflation of test accuracy due to data leakage in deep learning-based classification of OCT images,” Sci. Data, vol. 9, no. 1, p. 580, Sep. 2022, doi: 10.1038/s41597-022-01618-6.
Z. K. Simbolon and B. Baniah, “Perbandingan Metode City Block Dan Canberra Distance Dalam Identifikasi Kematangan Buah Pepaya Madu (Callina Papaya) Berdasarkan Warna RGB,” J. Artif. Intell. Softw. Eng., vol. 3, no. 1, pp. 14–17, May 2023, doi: 10.30811/jaise.v3i1.4135.
W. A. Wiharja, T. Al Mudzakir, H. Y. Novita, and J. Indra, “Perbandingan Algoritma Logistic Regression dan K-Nearest Neighbor Dalam Klasifikasi Kematangan Buah Pepaya,” Bull. Comput. Sci. Res., vol. 5, no. 4, pp. 353–360, Jun. 2025, doi: 10.47065/bulletincsr.v5i4.550.
A. R. R. Ivani, A. Z. Kurniadi, A. B. A. Andira, and I. Wahyuni, “Perbandiangan VGG16 dan MobileNetV2 untuk Klasifikasi Tingkat Kematangan Buah Apel,” J. Sist. Komput. Asia, vol. 3, no. 1, pp. 30–39, 2025, doi: 10.32815/jiskomsia.v3i1.136.
R. R. Santosa, A. Sasikirana, Z. L. Ramadhan, Y. P. B. Satria, and A. Nugroho, “Perbandingan Kinerja MobileNetV2 dan ResNet50V2 dalam Klasifikasi Tingkat Kematangan Tomat,” Indones. J. Sci. Technol. Humanit., vol. 3, no. 1, pp. 49–56, Jun. 2025, doi: 10.60076/ijstech.v3i1.1335.
M. Wildan Fitrian, W. Widayani, H. Sismoro, N. Aini, and D. Nurani, “Prediksi Perbandingan Kinerja MobileNetV2 dan EfficientNetB0 dalam Prediksi Kematangan Buah Jaboticaba,” J. Autom. Comput. Inf. Syst., vol. 6, no. 1, pp. 233–246, Apr. 2026, doi: 10.47134/jacis.v6i1.177.
M. Sandler, A. Howard, M. Zhu, A. Zhmoginov, and L.-C. Chen, “MobileNetV2: Inverted Residuals and Linear Bottlenecks,” in 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Salt Lake City: IEEE, Jun. 2018, pp. 4510–4520. doi: 10.1109/CVPR.2018.00474.
M. Tan and Q. V. Le, “EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks,” in Proceedings of the 36th International Conference on Machine Learning, Long Beach: Proceedings of Machine Learning Research, May 2019, pp. 6105–6114. doi: 10.48550/arXiv.1905.11946.
R. R. Selvaraju, M. Cogswell, A. Das, R. Vedantam, D. Parikh, and D. Batra, “Grad-CAM: Visual Explanations from Deep Networks via Gradient-Based Localization,” Int. J. Comput. Vis., vol. 128, no. 2, pp. 336–359, Feb. 2020, doi: 10.1007/s11263-019-01228-7.
R. Shah and A. K. Shrivastava, “Semantic-Aware Image Deduplication: Leveraging Object Recognition for Enhanced Accuracy,” Int. J. Intell. Eng. Syst., vol. 18, no. 7, pp. 116–130, Aug. 2025, doi: 10.22266/ijies2025.0831.09.
N. Nurmalasari, Y. A. Setiawan, W. Astuti, M. R. R. Saelan, S. Masturoh, and T. Haryanti, “Classification for Papaya Fruit Maturity Level With Convolutional Neural Network,” J. Ris. Inform., vol. 5, no. 3, pp. 331–338, Jun. 2023, doi: 10.34288/jri.v5i3.225.
N. Amanda, R. Wulanningrum, and J. Sahertian, “Penerapan MobileNetV2 untuk Klasifikasi Kondisi Buah Alpukat Mentega,” in Prosiding SEMNAS INOTEK, Kediri, 2025, pp. 575–582. doi: 10.29407/bp5fen81.
C. Nastitie and I. Handayani, “Perbandingan Metode MobileNetV2 dan EfficientNetB0 untuk Klasifikasi Jenis Pisang,” INTECOMS J. Inf. Technol. Comput. Sci., vol. 8, no. 6, pp. 1990–1999, Nov. 2025, doi: 10.31539/0240cx09.
A. C. Hartono and A. R. Muslikh, “Penerapan Transfer Learning MobileNetV2 pada Klasifikasi Citra Jenis Buah-Buahan,” J. Inf. Syst. Appl. Dev., vol. 3, no. 2, pp. 102–110, 2025, doi: 10.26905/jisad.v3i2.16187.
N. R. Fauza, A. S. K. Aprilia, and I. Wahyuni, “MobileNetV2 untuk Deteksi Kualitas Tempe di Sentra Industri Tempe Sanan,” J. Sist. Komput. Asia, vol. 3, no. 1, pp. 15–29, 2025, doi: 10.32815/jiskomsia.v3i1.134.
D. N. Ananda, S. D. Dalimajun, Y. R. Bere, G. R. Talo, and I. Wahyuni, “Perbandingan Kinerja DenseNet dan MobileNetV2 dalam Pengembangan Model Deep Learning untuk Deteksi Tanaman Kacang Tanah Sehat dan Terserang Penyakit,” J. Sist. Komput. Asia, vol. 3, no. 1, pp. 63–71, 2025, doi: 10.32815/jiskomsia.v3i1.148.
W. Febrianto, D. Kurniawan, and I. Wahyuni, “Perbandingan Kinerja Metode VGG-16 dan ResNet50 untuk Deteksi Penyakit pada Tanaman Kubis di Nongkojajar Kabupaten Pasuruan,” J. Sist. Komput. Asia, vol. 3, no. 2, pp. 1–9, 2025, doi: 10.32815/jiskomsia.v3i2.144.
I. Wahyuni, C.-C. Chang, W.-J. Wang, and D. Liang, “Input Variability Analysis of Aerial Imagery in Rice Semantic Segmentation Using UNet-VGG16,” in 2025 9th International Conference On Electrical, Electronics And Information Engineering (ICEEIE), IEEE, Sep. 2025, pp. 1–6. doi: 10.1109/ICEEIE66203.2025.11252106.
T. T. Sujaka, I. N. Switrayana, and I. M. H. Fillah, “Pengaplikasian Convolutional Neural Network (MobileNetV3) Memanfaatkan Transfer Learning Untuk Membedakan Tanaman Cabai Berasal Dari Genus Capsicum Annuum,” Build. Informatics, Technol. Sci., vol. 7, no. 3, pp. 1761–1774, 2025, doi: 10.47065/bits.v7i3.8740.
M. Fadhilatuzzahro, A. Fazrin, Z. I. Arishandy, T. A. Sukoco, and H. Maulana, “Optimalisasi Model CNN EfficientNet-B0 dengan Fine Tuning untuk Klasifikasi Penyakit Buah Jeruk,” J. Ilm. Teknol. Inf. dan Robot., vol. 7, no. 1, pp. 23–30, 2025, doi: 10.33005/jifti.v7i1.162.
How To Cite This :
Refbacks
- There are currently no refbacks.
