AI-Based Smart Farming Implementation: A Comparative Analysis of EfficientNet-B0-SVM and ResNet-50-SVM for Cacao Leaf Disease Detection

Nurhikma Arifin; Milasari

doi:10.24252/instek.v11i2.71678

Authors

Nurhikma Arifin Universitas Sulawesi Barat
Milasari Universitas Sulawesi Barat

DOI:

https://doi.org/10.24252/instek.v11i2.71678

Keywords:

EfficientNet-B0, hybrid method, PCA, ResNet-50, support vector machine

Abstract

Cacao productivity in West Sulawesi remains suboptimal, primarily due to leaf diseases such as Vascular Streak Dieback (VSD) and Leaf Blight, which negatively affect plant growth. Manual disease identification performed by farmers still has several limitations, particularly in terms of subjectivity and accuracy. These limitations become more apparent when distinguishing early-stage disease symptoms that share similar visual characteristics, often leading to misidentification and reduced crop quality. Therefore, this study aimed to develop a more accurate and objective system for cacao leaf disease classification. The dataset consisted of 549 images divided into three classes: Leaf Blight, Healthy, and Vascular Streak Dieback (VSD). The training process was conducted in two stages, namely transfer learning and fine-tuning, while Grid Search was employed to optimize the SVM hyperparameters. The results showed that the EfficientNet-B0-SVM model achieved a validation accuracy of 95.45% and a final test accuracy of 91.82%, whereas the ResNet-50-SVM model achieved a validation accuracy of 93.18% and a final test accuracy of 92.73%. These findings indicate that both models were able to classify cacao leaf diseases with a high level of accuracy. In addition, the application of Principal Component Analysis (PCA) significantly reduced the number of extracted features while preserving more than 95% of the original information. Therefore, the proposed CNN-PCA-SVM approach can be considered an effective method for supporting image-based cacao leaf disease identification.

Downloads

Download data is not yet available.

References

[1] F. Herman, “Performance Analysis of Cocoa Certification Program in Polewali Mandar Regency,” anjoro, vol. 1, no. 1, hlm. 1–6, Mar 2020, doi: 10.31605/anjoro.v1i1.641

[2] “Badan Pusat Statistik BPS-Statistics Indonesia,” 2024.

[3] G. A. S. Gado and P. N. Primandari, “Sistem Klasifikasi Berbasis Android untuk Penyakit Buah Kakao Menggunakan CNN,” Jurnal Teknologi Terpadu, vol. 11, pp. 27–35, 2025.

[4] R. Robet, J. T. K. Perangin Angin, and T. H. Siregar, “Attention Augmented Deep Learning Model for Enhanced Feature Extraction in cacao Disease Recognition,” Sinkron : Jurnal dan Penelitian Teknik Informatika, vol. 9, no. 4, pp. 1965–1977, Oct. 2025, doi: 10.33395/sinkron.v9i4.15249.

[5] S. Kumi, D. Kelly, J. Woodstuff, R. K. Lomotey, R. Orji, dan R. Deters, “cacao Companion: Deep Learning-Based Smartphone Application for cacao Disease Detection,” Procedia Comput. Sci., vol. 203, hlm. 87–94, 2022, doi: 10.1016/j.procs.2022.07.013.

[6] S. Sintia, N. Arifin, dan Chairi Nur, “Implementasi Mobilenetv2 Pada Klasifikasi Penyakit Daun Bibit Kakao Berbasis Pengolahan Citra Digital,” Insect Inform. Secur. J. Tek. Inform., vol. 12, no. 1, hlm. 75–85, 2026, doi: https://doi.org/10.33506/insect.v12i01.5396.

[7] K. Sing Soh, E. Gubin Moung, K. John Julius Danker, J. A. Dargham, dan A. Farzamnia, “cacao Diseases Classification using Deep Learning Algorithm,” ITM Web Conf., vol. 63, hlm. 01014, 2024, doi: 10.1051/itmconf/20246301014.

[8] V. V, S. M and K. P, "Early Detection of cacao Plant Diseases Using Custom Deep Learning Model," 2026 3rd International Conference on Research Methodologies in Knowledge Management, Artificial Intelligence and Telecommunication Engineering (RMKMATE), Chennai, India, 2026, pp. 1-5, doi: 10.1109/RMKMATE69073.2026.11519061.

[9] S. Kumar, A. K. Pandey, P. K. Srivastava, K. Sharma, N. Pal, and P. K. Patidar, "Smartcacao: Leveraging deep learning for cacao leaf disease identification," in Proceedings of International Conference on Computing Systems and Intelligent Applications (ComSIA 2025), A. Jaiswal, S. Anand, A. E. Hassanien, and A. T. Azar, Eds., Lecture Notes in Networks and Systems, vol. 1501. Singapore: Springer, 2026, doi: 10.1007/978-981-96-8350-5_39.

[10] G. A. S. Gado dan P. N. Primandari, “Sistem Klasifikasi Berbasis Android Untuk Penyakit Buah Kakao Menggunakan CNN Nasnet-Mobile,” J. Teknol. Terpadu, vol. 11, no. 1, hlm. 27–35, 2025, doi: https://journal.nurulfikri.ac.id/index.php/jtt.

[11] Richard K. Lomotey, Sandra Kumi, Rita Orji, dan Ralph Deters, “Automatic detection and diagnosis of cacao diseases using mobile tech and deep learning,” Int. J. Sustain. Agric. Manag. Inform., vol. 10, no. 1, Des 2023. https://www.inderscienceonline.com/doi/pdf/10.1504/IJSAMI.2024.135403? download=true.

[12] K. J. Ayikpa, A. B. Ballo, D. Mamadou, dan P. Gouton, “A Hybrid Feature Fusion and Machine Learning Approach for Cocoa Pod Disease Detection,” Hum-Cent Intell Syst, vol. 5, no. 4, hlm. 531–544, Nov 2025, doi: 10.1007/s44230-025-00117-2.

[13] F. Mohameth, C. Bingcai, dan K. A. Sada, “Plant Disease Detection with Deep Learning and Feature Extraction Using Plant Village,” J. Comput. Commun., vol. 08, no. 06, hlm. 10–22, 2020, doi: 10.4236/jcc.2020.86002.

[14] D. F. Namaemba, “Maize Crop Pests and Diseases Classification Using Hybrid Models.,” MSc in Data Analytics, CCT College Dublin, 2024. doi: 10.63227/370.723.93.

[15] D. S. Alwan and M. H. Naji, "Rice Diseases Classification by Residual Network 50 (RESNET-50) and Support Vector Machine (SVM) Modeling," Journal of Kufa for Mathematics and Computer, vol. 10, no. 1, pp. 96–101, Mar. 2023, doi: 10.31642/JoKMC/2018/100114.

[16] K. V. Reddy, L. Jai Vinita, P. Helen Vijitha, dan A. Pravin Renold, “Mango leaf disease classification using a hybrid model of ResNet-50 and quantum machine learning,” Front. Sustain. Food Syst., vol. 10, hlm. 1837661, Jun 2026, doi: 10.3389/fsufs.2026.1837661.