Implementation of Xception method for Early Detection of Plant Diseases

Gilbertus Kristo Anugerah Adviandito; Albertus Dwiyoga Widiantoro

doi:10.24167/sisforma.v11i2.14741

Implementation of Xception method for Early Detection of Plant Diseases

Gilbertus Kristo Anugerah Adviandito, Albertus Dwiyoga Widiantoro

Abstract

Plant diseases pose a significant threat to global food security, reducing both crop quality and productivity. To address this challenge, this study proposes an enhanced deep learning approach using a modified Xception architecture for plant disease detection. The model integrates Global Average Pooling and a Dropout layer to reduce overfitting and improve generalization performance on complex agricultural data. A large-scale multi-class dataset of 54,305 images across 38 plant species was employed, with preprocessing steps including resizing, augmentation, and normalization. The model was trained using the Adam optimizer with a batch size of 32 over 5 epochs, and its performance was compared against CNN, InceptionV3, InceptionResNetV2, and XGBoost. Experimental results demonstrate that the modified Xception achieved the highest accuracy, with 97.61% on training data and 97.63% on validation data, outperforming the other models under identical experimental settings.

The novelty of this research lies in the application of a modified Xception network on a large-scale, multi-class plant disease dataset combined with a systematic comparative evaluation against multiple architectures, an approach that has been rarely explored in previous studies. The findings not only confirm the robustness of Xception in handling complex agricultural imagery but also provide a practical framework for developing early disease detection systems in precision agriculture.

Keywords

Accuracy, Detection, Disease, Plants, Xception

Full Text:

PDF

References

K. G. Panchbhai, M. G. Lanjewar, V. V. Malik, and P. Charanarur, “Small size CNN (CAS-CNN), and modified MobileNetV2 (CAS-MODMOBNET) to identify cashew nut and fruit diseases,” Multimed Tools Appl, 2024, doi: 10.1007/s11042-024-19042-w.

M. Srivastava and J. Meena, “Plant leaf disease detection and classification using modified transfer learning models,” Multimed Tools Appl, Apr. 2023, doi: 10.1007/s11042-023-16929-y.

A. Abisha and N. Bharathi, “Modified Xception-RF: A biotic stress prediction model in plant leaves using feature extraction and classification,” Multimed Tools Appl, Apr. 2024, doi: 10.1007/s11042-024-18534-z.

J. U. M. Akbar, S. F. Kamarulzaman, A. J. M. Muzahid, M. A. Rahman, and M. Uddin, “A Comprehensive Review on Deep Learning Assisted Computer Vision Techniques for Smart Greenhouse Agriculture,” IEEE Access, vol. 12, pp. 4485–4522, 2024, doi: 10.1109/ACCESS.2024.3349418.

N. Yao et al., “L2MXception: an improved Xception network for classification of peach diseases,” Plant Methods, vol. 17, no. 1, Dec. 2021, doi: 10.1186/s13007-021-00736-3.

“764-2559-1-PB”.

S. Mishra, R. Sachan, and D. Rajpal, “Deep Convolutional Neural Network based Detection System for Real-time Corn Plant Disease Recognition,” in Procedia Computer Science, Elsevier B.V., 2020, pp. 2003–2010. doi: 10.1016/j.procs.2020.03.236.

N. An, “Xception Network for Weather Image Recognition Based on Transfer Learning,” in Proceedings - 2022 International Conference on Machine Learning and Intelligent Systems Engineering, MLISE 2022, Institute of Electrical and Electronics Engineers Inc., 2022, pp. 330–333. doi: 10.1109/MLISE57402.2022.00072.

S. A. Joshi, A. M. Bongale, P. O. Olsson, S. Urolagin, D. Dharrao, and A. Bongale, “Enhanced Pre-Trained Xception Model Transfer Learned for Breast Cancer Detection,” Computation, vol. 11, no. 3, Mar. 2023, doi: 10.3390/computation11030059.

H. H. Farag, L. A. A. Said, M. R. M. Rizk, and M. A. E. Ahmed, “Hyperparameters optimization for ResNet and Xception in the purpose of diagnosing COVID-19,” Journal of Intelligent and Fuzzy Systems, vol. 41, no. 2, pp. 3555–3571, 2021, doi: 10.3233/JIFS-210925.

F. Chollet, “Xception: Deep Learning with Depthwise Separable Convolutions.”

W. Muhammad, S. Aramvith, and T. Onoye, “Multi-scale Xception based depthwise separable convolution for single image superresolution,” PLoS One, vol. 16, no. 8 August, Aug. 2021, doi: 10.1371/journal.pone.0249278.

B. S. Abunasser, M. Rasheed, J. Al-Hiealy, I. S. Zaqout, and S. S. Abu-Naser, “Breast Cancer Detection and Classification using Deep Learning Xception Algorithm.” [Online]. Available: www.ijacsa.thesai.org

Universitas Diponegoro. Faculty of Science and Mathematics. Department of Informatics, Institute of Electrical and Electronics Engineers. Indonesia Section, and Institute of Electrical and Electronics Engineers, Proceeding : the 4th International Conference on Informatics and Computational Sciences : ICICoS 2020, 10-11 November 2020.

S. Sarkar, J. Adhikari Ray, C. Mukherjee, S. Ghosh, N. Jayanthi, and K. R. Chairma Lakshmi, “Plant Leaf Disease Classification Based on SVM Based Densenets,” in 2023 International Conference on Advances in Computation, Communication and Information Technology, ICAICCIT 2023, Institute of Electrical and Electronics Engineers Inc., 2023, pp. 636–641. doi: 10.1109/ICAICCIT60255.2023.10465886.

Ö. POLAT, “Detection of Covid-19 from Chest CT Images using Xception Architecture: A Deep Transfer Learning based Approach,” Sakarya University Journal of Science, vol. 25, no. 3, pp. 800–810, Jun. 2021, doi: 10.16984/saufenbilder.903886.

X. Wu, R. Liu, H. Yang, and Z. Chen, “An Xception Based Convolutional Neural Network for Scene Image Classification with Transfer Learning,” in Proceedings - 2020 2nd International Conference on Information Technology and Computer Application, ITCA 2020, Institute of Electrical and Electronics Engineers Inc., Dec. 2020, pp. 262–267. doi: 10.1109/ITCA52113.2020.00063.

H. Madeira, J. G. Silva, and J. Carreira, “Xception: Software Fault Injection and Monitoring in Processor Functional Units Xception: Software Fault Injection and Monitoring in Processor Functional Units 1,” 2001. [Online]. Available: https://www.researchgate.net/publication/2396771

F. Salim, F. Saeed, S. Basurra, S. N. Qasem, and T. Al-Hadhrami, “DenseNet-201 and Xception Pre-Trained Deep Learning Models for Fruit Recognition,” Electronics (Switzerland), vol. 12, no. 14, Jul. 2023, doi: 10.3390/electronics12143132.

N. Yao et al., “L2MXception: an improved Xception network for classification of peach diseases,” Plant Methods, vol. 17, no. 1, Dec. 2021, doi: 10.1186/s13007-021-00736-3.

N. Jinsakul, C. F. Tsai, C. E. Tsai, and P. Wu, “Enhancement of deep learning in image classification performance using Xception with the swish activation function for colorectal polyp preliminary screening,” Mathematics, vol. 7, no. 12, Dec. 2019, doi: 10.3390/MATH7121170.

M. T. Do, H. N. Huynh, T. N. Tran, and T. L. Hoang, “Prediction of Retina Damage in Optical Coherence Tomography Image Using Xception Architecture Model,” in 2023 IEEE 5th Eurasia Conference on Biomedical Engineering, Healthcare and Sustainability, ECBIOS 2023, Institute of Electrical and Electronics Engineers Inc., 2023, pp. 58–61. doi: 10.1109/ECBIOS57802.2023.10218586.

R. K. Dubey and D. K. Choubey, “Feature selection with Optimized XGBoost model-based paddy plant leaf disease classification,” Multimed Tools Appl, 2024, doi: 10.1007/s11042-024-18524-1.

N. M. Pandiyan, “Digital Image Forgery Detection Using Pre-Trained Xception Model as Feature Extractor,” in 2023 International Conference for Advancement in Technology, ICONAT 2023, Institute of Electrical and Electronics Engineers Inc., 2023. doi: 10.1109/ICONAT57137.2023.10080127.

DOI: https://doi.org/10.24167/sisforma.v11i2.14741