Attention Augmented Deep Learning Model for Enhanced Feature Extraction in Cacao Disease Recognition

Authors

  • Robet Department of Informatics, STMIK Time, Medan, Indonesia
  • Johanes Terang Kita Perangin Angin Department of Information Systems, STMIK Time, Medan, Indonesia
  • Tarq Hilmar Siregar Department of Informatics, STMIK Time, Medan, Indonesia

DOI:

10.33395/sinkron.v9i4.15249

Keywords:

Cacao Disease, ResNeXt50, SE, CBAM, Attention Mechanisms

Abstract

Accurate cacao disease recognition is critical for safeguarding yields and reducing losses. Prior cacao studies primarily rely on handcrafted descriptors (eg, Color Histogram, LBP, GLCM) or standard CNN/transfer-learning pipelines, often limited to ≤ 3 classes and a single plant organ; explicit channel-spatial attention and comprehensive multiclass evaluation remain uncommon. To the best of our knowledge, no prior work integrates Squeeze-and-Excitation (SE) and the Convolutional Block Attention Module (CBAM) on a ResNeXt50 backbone for six-class cacao disease classification, accompanied by a standardized ablation study and t-SNE-based interpretability. We propose a six-class classifier (five diseases + healthy) built on ResNeXt-50 enhanced with SE (channel recalibration) and CBAM (channel-spatial emphasis) to highlight lesion-relevant cues. The dataset comprises labeled leaf and pod images from public sources collected under field-like conditions; preprocessing includes resizing to 224x224, normalization, and augmentation (flips, small rotations, color jitter, random resized crops). Trained with Adam and early stopping, ResNeXt50+SE+CBAM attains 97% test accuracy and 0.97 macro-F1, surpassing a ResNeXt50 baseline of 94% and 0.95 and SE-only/CBAM-only variants. Confusion matrix and t-SNE analyses show fewer mix-ups among visual classes and clearer separability, while the ablation validates complementary benefits of SE and CBAM. On a desktop-hosted, web-based setup, batch-1 inference at 224x224 is 7.46 ms/image (134 FPS), demonstrating real-time capability. The findings support deployment as browser-based decision-support tools for farmers and integration into continuous field-monitoring systems.

GS Cited Analysis

Downloads

Download data is not yet available.

References

Baculio, N. G., & Barbosa, J. B. (2022). An Objective Classification Approach of Cacao Pods using Local Binary Pattern Features and Artificial Neural Network Architecture (ANN). Indian Journal of Science and Technology, 15(11), 495–504. https://doi.org/10.17485/ijst/v15i11.60

Bhuiyan, M. A. B., Abdullah, H. M., Arman, S. E., Saminur Rahman, S., & Al Mahmud, K. (2023). BananaSqueezeNet: A very fast, lightweight convolutional neural network for the diagnosis of three prominent banana leaf diseases. Smart Agricultural Technology, 4(March), 100214. https://doi.org/10.1016/j.atech.2023.100214

Brosas, D. G., Villafuerte, R. S., & Obediencia, D. C. (2020). Adaptive Neuro-Fuzzy Approach for Cacao Bean Grading Classification Process. Proceeding - 2020 3rd International Conference on Vocational Education and Electrical Engineering: Strengthening the Framework of Society 5.0 through Innovations in Education, Electrical, Engineering and Informatics Engineering, ICVEE 2020. https://doi.org/10.1109/ICVEE50212.2020.9243281

Bueno, G. E., Valenzuela, K. A., & Arboleda, E. R. (2020). Maturity classification of cacao through spectrogram and convolutional neural network. Jurnal Teknologi Dan Sistem Komputer, 8(3), 228–233. https://doi.org/10.14710/jtsiskom.2020.13733

Cagadas, D. O., Labajan, R. A., Cagadas, D. O., April, R., Cacao, L. L., Classification, D., Image, U., Cagadas, D. O., & Labajan, R. A. A. (2024). Leaf-Based Cacao Diseases Classification Using Image Processing To cite this version : HAL Id : hal-04483097 Leaf-Based Cacao Diseases Classification Using Image. HAL Open Science.

Clarence, E., Diego, S. S., Rodrin, S. G. C., & Arboleda, E. R. (2025). Multi-Feature Visual Analysis And K-Nearest. 28–34.

Godmalin, R. A., Aliac, C. J., & Feliscuzo, L. (2022). Classification of Cacao Pod if Healthy or Attack by Pest or Black Pod Disease Using Deep Learning Algorithm. 4th IEEE International Conference on Artificial Intelligence in Engineering and Technology, IICAIET 2022. https://doi.org/10.1109/IICAIET55139.2022.9936817

Godmalin, R. A., Aliac, C. J., & Feliscuzo, L. (2023). Cacao Pod Infection Level Classification Using Transfer Learning. 2023 IEEE Open Conference of Electrical, Electronic and Information Sciences, EStream 2023 - Proceedings. https://doi.org/10.1109/eStream59056.2023.10135062

Hamida, H. (2024). 10 Negara Produsen Kakao Terbesar Di Dunia. Tempo.Co. https://www.tempo.co/ekonomi/10-negara-produsen-kakao-terbesar-di-dunia-39695

Harvyanti, A. F. M., Baihaki, R. I., Dafik, Ridlo, Z. R., & Agustin, I. H. (2023). Application of Convolutional Neural Network for Identifying Cocoa Leaf Disease (Vol. 2). Atlantis Press International BV. https://doi.org/10.2991/978-94-6463-174-6_21

Hortinela, C. C., & Tupas, K. J. R. (2022). Classification of Cacao Beans Based on their External Physical Features Using Convolutional Neural Network. 2022 IEEE Region 10 Symposium, TENSYMP 2022. https://doi.org/10.1109/TENSYMP54529.2022.9864337

Indotama, P. F. (2022). Indonesia Masuk Daftar 7 Negara Penghasil Kakao Terbesar di Dunia. Freyabadi.Com. https://www.freyabadi.com/id/blog/indonesia-masuk-daftar-7-negara-penghasil-kakao-terbesar-di-dunia

Jesse, A., Douha, N. Y.-R., & Lenka, P. (2024). Image Classification for CSSVD Detection in Cacao Plants. ArXiv, 1–5. http://arxiv.org/abs/2405.04535

Jeswani, J., Ahmed, A. S., Zaid, K., & Fernandes, R. (2023). Leaf Disease Detection Using Deep Learning. Atlantis Press International BV. https://doi.org/10.2991/978-94-6463-136-4_87

Kouassi, K. S., Diarra, M., Edi, K. H., & Jean-Claude, K. B. (2025). Detection of cocoa pod diseases using a hybrid feature extractor combining CNN and vision transformer with dual classifier. Edelweiss Applied Science and Technology, 9(1), 668–681. https://doi.org/10.55214/25768484.V9I1.4209

Mahadevan, K., Punitha, A., & Suresh, J. (2024). Automatic recognition of Rice Plant leaf diseases detection using deep neural network with improved threshold neural network. E-Prime - Advances in Electrical Engineering, Electronics and Energy, 8(February), 100534. https://doi.org/10.1016/j.prime.2024.100534

Mamadou, D., Ayikpa, K. J., Ballo, A. B., & Kouassi, B. M. (2023). Cocoa Pods Diseases Detection by MobileNet Confluence and Classification Algorithms. International Journal of Advanced Computer Science and Applications, 14(9), 344–352. https://doi.org/10.14569/IJACSA.2023.0140937

Miracle, A. (2024). Enhancing Cocoa Crop Resilience in Ghana: The Application of Convolutional Neural Networks for Early Detection of Disease and Pest Infestations. Qeios, 1–14. https://doi.org/10.32388/dps5zh

Moore, S. E., & Modupe, A. (2025). Cacao Plant Disease Detection and Classification. Research Square, January. https://doi.org/10.21203/rs.3.rs-5763786/v1

Nikith, B. V., Keerthan, N. K. S., Praneeth, M. S., & Amrita, D. T. (2022). Leaf Disease Detection and Classification. Procedia Computer Science, 218, 291–300. https://doi.org/10.1016/j.procs.2023.01.011

Robet, Juliandy, C., Andi, Hendri, Hendrik, J., & Tarigan, F. A. (2022). Image Road Surface Classification Based on GLCM Feature Using LGBM Classifier. IOP Conference Series: Earth and Environmental Science, 1083(1). https://doi.org/10.1088/1755-1315/1083/1/012006

Robet, R., Terang, J., Perangin, K., & Pribadi, O. (2024). Implementation of Deep Learning Model for Classification of Household Trash Image. Sinkron, 8(October), 2575–2583.

Robet, R., Terang, J., Perangin, K., & Wijaya, E. (2025). Improving Resnet Model in Safety Gear Classification using Finest Optimizer. Journal of Artificial Intelligence and Engineering Applications, 4(2).

Rola, J. B., Barrera, J. J. A., Calhoun, M. V., Maaghop, J. F. O., Unajan, M. C., Boncalon, J. M., Sebios, E. T., & Espinosa, J. S. (2024). Convolutional Neural Network Model for Cacao Phytophthora Palmivora Disease Recognition. International Journal of Advanced Computer Science and Applications, 15(8), 986–990. https://doi.org/10.14569/IJACSA.2024.0150897

Sing Soh, K., Gubin Moung, E., John Julius Danker, K., Dargham, J. A., & Farzamnia, A. (2024). Cocoa Diseases Classification using Deep Learning Algorithm. ITM Web of Conferences, 63, 01014. https://doi.org/10.1051/itmconf/20246301014

Varma, T., Mate, P., Azeem, N. A., Sharma, S., & Singh, B. (2025). Automatic mango leaf disease detection using different transfer learning models. Multimedia Tools and Applications, 84(11), 9185–9218. https://doi.org/10.1007/s11042-024-19265-x

Vera, D. B., Oviedo, B., Casanova, W. C., & Zambrano-Vega, C. (2024). Deep Learning-Based Computational Model for Disease Identification in Cocoa Pods (Theobroma cacao L.). ArXiv, 1–16. https://doi.org/10.1109/PIC62406.2024.10892748

Yarak, K., Witayangkurn, A., Kritiyutanont, K., Arunplod, C., & Shibasaki, R. (2021). Oil palm tree detection and health classification on high‐resolution imagery using deep learning. Agriculture (Switzerland), 11(2), 1–17. https://doi.org/10.3390/agriculture11020183

Downloads


Crossmark Updates

How to Cite

Robet, R., Perangin Angin, J. T. K. ., & Siregar, T. H. (2025). Attention Augmented Deep Learning Model for Enhanced Feature Extraction in Cacao Disease Recognition. Sinkron : Jurnal Dan Penelitian Teknik Informatika, 9(4), 1965-1977. https://doi.org/10.33395/sinkron.v9i4.15249

Issue

Vol. 9 No. 4 (2025): Articles Research October 2025

Section

Articles

License

Copyright (c) 2025 Robet, Johanes Terang Kita Perangin Angin, Tarq Hilmar Siregar

Creative Commons License

This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.

Crossref
Scopus
Google Scholar
Europe PMC