Explainable Hybrid Deep Learning with EfficientNetB3–DenseNet121 Feature Fusion for Banana Leaf Spot Disease Classification

Bilal Bataineh; Ghazi Shakah; Bajeszeyadaljunaeidia; Issa Alsmadi; Mohamed S. Sawah

doi:10.29284/yeveds27

Authors

Bilal Bataineh Department of Computer Science, Jadara University, Irbid, Jordan.
Ghazi Shakah Department of Software Engineering, Faculty of Information Technology, Ajloun National University, P.O.43, Ajloun-26810, Jordan
Bajeszeyadaljunaeidia Department of Computer Science, Faculty of Information Technology, Ajloun National University, P.O.43, Ajloun-26810, Jordan
Issa Alsmadi Department of Data Science and Artificial Intelligence, Faculty of Information Technology, Ajloun National University P.O.43, Ajloun-26810, Jordan
Mohamed S. Sawah Department of Data Science and Artificial Intelligence, Faculty of Information Technology, Ajloun National University, P.O.43, Ajloun-26810, Jordan

DOI:

https://doi.org/10.29284/yeveds27

Keywords:

Banana Leaf Disease Classification; Hybrid Deep Learning; Efficientnetb3; Densenet121; Transfer Learning; Feature Fusion; Explainable Ai; Grad-Cam; Plant Disease Detection; Precision Agriculture.

Abstract

Banana leaf spot diseases reduce crop quality and yield, making early and accurate diagnosis essential for precision agriculture. This study proposes an accurate and explainable hybrid deep learning framework for multiclass banana leaf spot disease classification. The model combines EfficientNetB3 and DenseNet121 in dual-branch feature-fusion architecture to capture complementary visual features. A stratified 70/15/15 split was used on the BananaLSD dataset (Cordana, Healthy, Pestalotiopsis, and Sigatoka), with class weighting, data augmentation, random erasing, label smoothing, and staged transfer learning to improve generalization. Performance was evaluated using accuracy, precision, recall, F1-score, confusion matrix, and multiclass ROC analysis, while Grad-CAM was used for explainability. The proposed hybrid model outperformed single-backbone baselines (EfficientNetB3 and DenseNet121), achieving 98.58% accuracy with the lowest loss (0.2578) and strong class-wise performance. Grad-CAM visualizations further confirmed that the model focused on disease-relevant lesion regions. These results support the proposed framework as a reliable tool for automated banana disease diagnosis.

References

[1] H. Patel, N. Patil, and D. Patel, “An Intelligent Grading System for Automated Identification and Classification of Banana Fruit Diseases Using Deep Neural Network,” International Journal of Computing and Digital Systems, vol. 15, no. 1, pp. 2210–142, doi: https://doi.org/10.12785/ijcds/150155.

[2] R. Retkute, K. S. Crew, J. E. Thomas, and C. A. Gilligan, “Detection of Banana Diseases Based on Landsat-8 Data and Machine Learning,” Remote Sensing, vol. 17, no. 13, pp. 2308–2308, Jul. 2025, doi: https://doi.org/10.3390/rs17132308.

[3] C. Ünal, “Incorporating Deep Learning into the Diagnosis of Banana Leaf Spot Diseases for the Protection of Banana Crops,” Tarım Bilimleri Dergisi, vol. 31, no. 3, pp. 780–794, Jul. 2025, doi: https://doi.org/10.15832/ankutbd.1579442.

[4] V. Chaudhari and M. Patil, "Banana leaf disease detection using K-means clustering and Feature extraction techniques," 2020 International Conference on Advances in Computing, Communication & Materials (ICACCM), Dehradun, India, 2020, pp. 126-130, doi: 10.1109/ICACCM50413.2020.9212816.

[5] C. Kanimalar and M. Karthikeyan, "Advances in Machine Learning and Deep Learning for Automated Banana Disease Detection: A Comprehensive Survey," 2024 4th International Conference on Ubiquitous Computing and Intelligent Information Systems (ICUIS), Gobichettipalayam, India, 2024, pp. 1160-1168, doi: 10.1109/ICUIS64676.2024.10866142.

[6] P. R. and V. A., “Disease diagnosis in banana leaves: a review on AI powered techniques,” PeerJ Computer Science, vol. 11, p. e3310, Dec. 2025, doi: https://doi.org/10.7717/peerj-cs.3310.

[7] M. S. Devi, B. Vamshikrishna, J. A. Pandian, P. V. Rao and S. R. Yaseen, "Eight Convolutional Layered Deep Convolutional Neural Network based Banana Leaf Disease Prediction," 2022 International Conference on Computing, Communication, and Intelligent Systems (ICCCIS), Greater Noida, India, 2022, pp. 313-317, doi: 10.1109/ICCCIS56430.2022.10037628.

[8] M. K. Haq, K. Tanvir, M. S. Kabir, N. S. B. Alam, S. K. Dey and V. Raju, "Enhancing Banana Leaf Spot Disease Classification using Dense Mobilenet V2," 2024 International Conference on Advances in Computing, Communication and Applied Informatics (ACCAI), Chennai, India, 2024, pp. 1-6, doi: 10.1109/ACCAI61061.2024.10602228.

[9] N. Jiménez, S. Orellana, B. Mazon-Olivo, W. Rivas-Asanza, and I. Ramírez-Morales, “Detection of Leaf Diseases in Banana Crops Using Deep Learning Techniques,” AI, vol. 6, no. 3, p. 61, Mar. 2025, doi: https://doi.org/10.3390/ai6030061.

[10] K. V. Gokula, D. J., R. Pinagadi Venkateswara, D. V., and K. S., “An automated segmentation and classification model for banana leaf disease detection,” Journal of Applied Biology & Biotechnology, Oct. 2022, doi: https://doi.org/10.7324/jabb.2021.100126.

[11] R. Sangeetha, J. Logeshwaran, J. Rocher, and J. Lloret, “An Improved Agro Deep Learning Model for Detection of Panama Wilts Disease in Banana Leaves,” AgriEngineering, vol. 5, no. 2, pp. 660–679, Mar. 2023, doi: https://doi.org/10.3390/agriengineering5020042.

[12] S. Samridhi, M. Kalpana, R. Parimalarangan, and N. V. Palanichamy, “Identification of Sigatoka Leaf Spot Disease in Banana Using Convolutional Neural Network (CNN),” Asian Journal of Agricultural Extension, Economics & Sociology, vol. 41, no. 9, pp. 931–936, Aug. 2023, doi: https://doi.org/10.9734/ajaees/2023/v41i92123.

[13] D. Thomas and J. M. David, "Banana Plant Disease Detection-Hybrid Machine Learning Approach," 2023 Annual International Conference on Emerging Research Areas: International Conference on Intelligent Systems (AICERA/ICIS), Kanjirapally, India, 2023, pp. 1-5, doi: 10.1109/AICERA/ICIS59538.2023.10420299.

[14] A. A and B. N, “An Hybrid Feature Extraction and Classification using Xception-RF for Multiclass Disease Classification in Plant Leaves,” Applied Artificial Intelligence, vol. 37, no. 1, Feb. 2023, doi: https://doi.org/10.1080/08839514.2023.2176614.

[15] A. Baki, Hasan Muhammad Abdullah, S. E. Arman, Sayed Saminur Rahman, and Kaies Al Mahmud, “BananaSqueezeNet: A very fast, lightweight convolutional neural network for the diagnosis of three prominent banana leaf diseases,” Smart Agricultural Technology, vol. 4, pp. 100214–100214, Aug. 2023, doi: https://doi.org/10.1016/j.atech.2023.100214.

[16] P. Nasra and S. Gupta, "ResNet50: Deep Learning Method for Automated Detection of Banana Leaf Spot Diseases," 2024 First International Conference on Innovations in Communications, Electrical and Computer Engineering (ICICEC), Davangere, India, 2024, pp. 1-6, doi: 10.1109/ICICEC62498.2024.10808847.

[17] M. Kumar and A. Kumar, "Deep Learning Meets Support Vector Machines: An Effective Hybrid Model for Banana Leaf Wilt Disease Severity Assessment," 2024 2nd International Conference on Disruptive Technologies (ICDT), Greater Noida, India, 2024, pp. 386-390, doi: 10.1109/ICDT61202.2024.10489081.

[18] Muhammad Saifuddeen, “LEAF DISEASE DETECTION IN BANANA PLANTS FOR PRECISION AGRICULTURE,” International Journal of Applied Mathematics, vol. 38, no. 7s, pp. 1251–1279, Oct. 2025, doi: https://doi.org/10.12732/ijam.v38i7s.585.

[19] J. B. Kakati and T. K. Das, "Classification of healthy and unhealthy banana leaves using deep learning approach: A comparative assessment," 2023 4th International Conference on Computing and Communication Systems (I3CS), Shillong, India, 2023, pp. 1-5, doi: 10.1109/I3CS58314.2023.10127309.

[20] C. A. Elinisa and N. Mduma, “Mobile-Based convolutional neural network model for the early identification of banana diseases,” Smart Agricultural Technology, vol. 7, p. 100423, Mar. 2024, doi: https://doi.org/10.1016/j.atech.2024.100423.

[21] N. B. Raja and P. S. Rajendran, “An efficient banana plant leaf disease classification using optimal ensemble deep transfer network,” Journal of Experimental & Theoretical Artificial Intelligence, pp. 1–24, Aug. 2023, doi: https://doi.org/10.1080/0952813x.2023.2241867.

[22] J. D. Thiagarajan et al., “Analysis of banana plant health using machine learning techniques,” Scientific Reports, vol. 14, no. 1, Jul. 2024, doi: https://doi.org/10.1038/s41598-024-63930-y.

[23] T. Hoque, N. A. Abdullah, M. A. Hasan and N. M. Sadali, "Banana Disease Recognition Using Deep Learning Based Method Combined With Attention Mechanism," 2024 27th International Conference on Computer and Information Technology (ICCIT), Cox's Bazar, Bangladesh, 2024, pp. 2852-2857, doi: 10.1109/ICCIT64611.2024.11022158.

[24] S. Singh, P. Prakash, G. Baghel, A. Singh, D. Raj and A. K. Agrawal, "Banana Crop Health: A Deep Learning-Based Model for Disease Detection and Classification," 2024 27th International Symposium on Wireless Personal Multimedia Communications (WPMC), Greater Noida, India, 2024, pp. 1-6, doi: 10.1109/WPMC63271.2024.10863138.

[25] P. Sahu, A. Chug, A. P. Singh, D. Singh, and R. P. Singh, “Classification and activation map visualization of banana diseases using deep learning models,” in International Conference on Innovative Computing and Communications, A. Khanna, D. Gupta, S. Bhattacharyya, A. E. Hassanien, S. Anand, and A. Jaiswal, Eds., Advances in Intelligent Systems and Computing, vol. 1394. Singapore: Springer, 2022, doi: 10.1007/978-981-16-3071-2_61.

[26] N. R. Rajalakshmi, S. Saravanan, J. Arunpandian, Sandeep Kumar Mathivanan, and G. Qin, “Early Detection of Banana Leaf Disease Using Novel Deep Convolutional Neural Network,” Journal of Data Science and Intelligent Systems, Aug. 2024, doi: https://doi.org/10.47852/bonviewJDSIS42021530.

[27] K. Yan, M. Kamran, and Y. Sun, “A Transfer Learning-Based Deep Convolutional Neural Network for Detection of Fusarium Wilt in Banana Crops,” AgriEngineering, vol. 5, no. 4, pp. 2381–2394, Dec. 2023, doi: https://doi.org/10.3390/agriengineering5040146.

[28] C. H R, R. G S, S. BR, and G. N. Yallappa, “Plant Disease Detection using Deep Learning in Banana and Sunflower,” Journal of Advanced Zoology, vol. 44, no. 3, pp. 93–106, Oct. 2023, doi: https://doi.org/10.17762/jaz.v44i3.369.

[29] S. E. Arman, Md. A. B. Bhuiyan, H. M. Abdullah, S. Islam, T. T. Chowdhury, and Md. A. Hossain, “BananaLSD: A banana leaf images dataset for classification of banana leaf diseases using machine learning,” Data in Brief, vol. 50, p. 109608, Oct. 2023, doi: https://doi.org/10.1016/j.dib.2023.109608.

[30] M. S. Sawah, “A novel feature selection-driven framework for sustainable and efficient rainfall classification using machine learning models,” Unconventional Resources, vol. 9, p. 100286, Jan. 2026, doi: https://doi.org/10.1016/j.uncres.2025.100286.

[31] N. Helmawati and E. Utami, “Analysis for Detecting Banana Leaf Disease Using the CNN Method,” JUITA: Jurnal Informatika, pp. 29–36, Mar. 2025, doi: https://doi.org/10.30595/juita.v13i1.24514.

[32] A. Jadhaw and H. R. Shah, “Hybrid EfficientNetB2 model for banana leaf disease classification using PCM-segmented and augmented images with stratified K-fold validation,” IAENG International Journal of Computer Science, vol. 53, no. 2, pp. 706–720, Feb. 2026.