CNN MODEL OPTIMIZATION USING MULTI-STAGE DATA AUGMENTATION FOR LOCAL PLANT LEAF DISEASE CLASSIFICATION
DOI:
https://doi.org/10.33480/jitk.v11i4.7845Keywords:
Agricultural AI, Deep Learning Optimization, Inception V3, Multi-Stage Data Augmentation, Plant Leaf Disease ClassificationAbstract
Plant leaf diseases are a major factor in reducing agricultural productivity, particularly for local commodities that often lack adequate artificial intelligence-based disease detection systems. This study aims to optimize the performance of a Convolutional Neural Network (CNN) model using the Inception V3 architecture through the application of multi-stage data augmentation to improve the classification accuracy of local plant leaf diseases. The dataset used is PlantifyDR from Kaggle, which has limited data volume and visual variation, requiring an effective augmentation strategy to improve the model's generalization ability. The proposed multi-stage augmentation approach consists of three stages—geometric, photometric, and texture-noise augmentation—that systematically enrich the diversity of training images. Evaluation results show that the proposed model provides significant performance improvements compared to the baseline model. The Inception V3 model with multi-stage augmentation achieved an accuracy of 0.762, an F1-score of 0.727, and a perfect AUC (1.00) across all classes, while the baseline model only achieved an accuracy of 0.595 and an average AUC of 0.877. Accuracy, loss, ROC curve, and confusion matrix analyses confirmed that multi-stage augmentation reduced overfitting and enhanced the model's ability to differentiate disease symptoms across leaf types. Therefore, this study concludes that multi-stage data augmentation is an effective approach for optimizing deep learning models on small and complex datasets, while also providing a significant contribution to the development of more accurate and reliable AI-based plant disease detection systems.
Downloads
References
[1] V. Rajasekhar, G. Arulselvi, and K. Suresh Babu, “Detection of diseases in rice leaf using convolutional neural network with transfer learning based on ResNeXt,” Int. J. Electr. Comput. Eng., vol. 14, no. 2, pp. 1739–1749, 2024, doi: 10.11591/ijece.v14i2.pp1739-1749.
[2] M. S. H. Shovon, S. J. Mozumder, O. K. Pal, M. F. Mridha, N. Asai, and J. Shin, “PlantDet: A Robust Multi-Model Ensemble Method Based on Deep Learning For Plant Disease Detection,” IEEE Access, vol. 11, no. March, pp. 34846–34859, 2023, doi: 10.1109/ACCESS.2023.3264835.
[3] V. Saxena, “Utilizing Support Vector Machines for Early Detection of Crop Diseases in Precision Agriculture a Data Mining Perspective,” Int. J. Intell. Syst. Appl. Eng., vol. 12, no. 16, pp. 281–288, 2024.
[4] A. Comander, B. Frates, and M. Tollefson, “PAVING the Path to Wellness for Breast Cancer Survivors: Lifestyle Medicine Education and Group Interventions,” Am. J. Lifestyle Med., vol. 15, no. 3, pp. 242–248, 2021, doi: 10.1177/1559827620986066.
[5] C. Dewi, F. Y. Bilaut, H. J. Christanto, and G. Dai, “Deep Learning for the Classification of Rice Leaf Diseases Using YOLOv8,” Math. Model. Eng. Probl., vol. 11, no. 11, pp. 3025–3034, 2024, doi: 10.18280/mmep.111115.
[6] H. Wang, “Phytotoxicity of Chemical Compounds from Cinnamomum camphora Pruning Waste in Germination and Plant Cultivation,” Int. J. Environ. Res. Public Health, vol. 19, no. 18, 2022, doi: 10.3390/ijerph191811617.
[7] R. Poojary, “Effect of data-augmentation on fine-tuned cnn model performance,” Iaes Int. J. Artif. Intell., vol. 10, no. 1, pp. 84–92, 2021, doi: 10.11591/ijai.v10.i1.pp84-92.
[8] A. García-Pérez, “CNN-based in situ tool wear detection: A study on model training and data augmentation in turning inserts,” J. Manuf. Syst., vol. 68, pp. 85–98, 2023, doi: 10.1016/j.jmsy.2023.03.005.
[9] S. Zhong, “Molecular image-convolutional neural network (CNN) assisted QSAR models for predicting contaminant reactivity toward OH radicals: Transfer learning, data augmentation and model interpretation,” Chem. Eng. J., vol. 408, 2021, doi: 10.1016/j.cej.2020.127998.
[10] X. Xu, “Sleep Stage Classification With Multi-Modal Fusion and Denoising Diffusion Model,” IEEE J. Biomed. Heal. Informatics, pp. 1–12, 2024, doi: 10.1109/JBHI.2024.3422472.
[11] R. Josphineleela, “A Multi-Stage Faster RCNN-Based iSPLInception for Skin Disease Classification Using Novel Optimization,” J. Digit. Imaging, vol. 36, no. 5, pp. 2210–2226, 2023, doi: 10.1007/s10278-023-00848-3.
[12] J. Y. Wu, “A joint classification-regression method for multi-stage remaining useful life prediction,” J. Manuf. Syst., vol. 58, pp. 109–119, 2021, doi: 10.1016/j.jmsy.2020.11.016.
[13] K. Alomar, “Data Augmentation in Classification and Segmentation: A Survey and New Strategies,” J. Imaging, vol. 9, no. 2, 2023, doi: 10.3390/jimaging9020046.
[14] J. J. Bird, “Chatbot Interaction with Artificial Intelligence: human data augmentation with T5 and language transformer ensemble for text classification,” J. Ambient Intell. Humaniz. Comput., vol. 14, no. 4, pp. 3129–3144, 2023, doi: 10.1007/s12652-021-03439-8.
[15] S. N. Yousafzai, I. M. Nasir, I. Keshta, and N. L. Fitriyani, “Multi-Stage Neural Network-Based Ensemble Learning Approach for Wheat Leaf Disease Classification,” IEEE Access, vol. 13, no. February, pp. 30101–30116, 2025, doi: 10.1109/ACCESS.2025.3541347.
[16] W. H. Alwan and S. M. Alturfi, “Multi-Stage Vision Transformer and Knowledge Graph Fusion for Enhanced Plant Disease Classification,” 2025, doi: 10.32604/csse.2025.064195.
[17] Q. Sun, “Multi-stage Co-planning Model for Power Distribution System and Hydrogen Energy System Under Uncertainties,” J. Mod. Power Syst. Clean Energy, vol. 11, no. 1, pp. 80–93, 2023, doi: 10.35833/MPCE.2022.000337.
[18] H. Nozari, “A multi-stage stochastic inventory management model for transport companies including several different transport modes,” Int. J. Manag. Sci. Eng. Manag., vol. 18, no. 2, pp. 134–144, 2023, doi: 10.1080/17509653.2022.2042747.
[19] K. Wang, “Two-stage stochastic optimal scheduling for multi-microgrid networks with natural gas blending with hydrogen and low carbon incentive under uncertain envinronments,” J. Energy Storage, vol. 72, 2023, doi: 10.1016/j.est.2023.108319.
[20] J. Wang, “Sensor Data Augmentation by Resampling in Contrastive Learning for Human Activity Recognition,” IEEE Sens. J., vol. 22, no. 23, pp. 22994–23008, 2022, doi: 10.1109/JSEN.2022.3214198.
[21] J. Zhang, “Data Augmentation and Dense-LSTM for Human Activity Recognition Using WiFi Signal,” IEEE Internet Things J., vol. 8, no. 6, pp. 4628–4641, 2021, doi: 10.1109/JIOT.2020.3026732.
[22] R. Bai, “Rolling bearing fault diagnosis based on multi-channel convolution neural network and multi-scale clipping fusion data augmentation,” Meas. J. Int. Meas. Confed., vol. 184, 2021, doi: 10.1016/j.measurement.2021.109885.
[23] Q. Guo, “Data Augmentation for Intelligent Mechanical Fault Diagnosis Based on Local Shared Multiple-Generator GAN,” IEEE Sens. J., vol. 22, no. 10, pp. 9598–9609, 2022, doi: 10.1109/JSEN.2022.3163658.
[24] R. Li, “Sample-Based Data Augmentation Based on Electroencephalogram Intrinsic Characteristics,” IEEE J. Biomed. Heal. Informatics, vol. 26, no. 10, pp. 4996–5003, 2022, doi: 10.1109/JBHI.2022.3185587.
[25] D. Wu, “Operando investigation of aqueous zinc manganese oxide batteries: multi-stage reaction mechanism revealed,” J. Mater. Chem. A, vol. 11, no. 30, pp. 16279–16292, 2023, doi: 10.1039/d3ta01549a.
[26] F. Guo, “A machine learning method for prediction of remaining useful life of supercapacitors with multi-stage modification,” J. Energy Storage, vol. 73, 2023, doi: 10.1016/j.est.2023.109160.
[27] J. Geng, “Investigation of dynamic response of drilling parameters and deformation characteristics of coal around borehole during multi-stage reaming in tectonic coal,” Int. J. Rock Mech. Min. Sci., vol. 170, 2023, doi: 10.1016/j.ijrmms.2023.105540.
[28] T. R. Mahesh, “Early Predictive Model for Detection of Plant Leaf Diseases Using MobileNetV2 Architecture,” Int. J. Intell. Syst. Appl. Eng., vol. 11, no. 2, pp. 46–54, 2023.
[29] N. U. A. Asan, “Antibacterial activity of Sireh (Piper betle L.) leaf extracts for controlling bacterial leaf blight diseases in rice plant,” Malays. J. Microbiol., vol. 18, no. 3, pp. 291–300, 2022, doi: 10.21161/mjm.221395.
[30] F. K. Sahlan, “The role of bacterial consortium as bioactivator to stimulate production and suppress grain rot disease and bacterial leaf blight in rice,” J. Trop. Plant Pests Dis., vol. 23, no. 2, pp. 65–70, 2023, doi: 10.23960/jhptt.22365-70.
[31] R. Parsibenehkohal, “A multi-stage framework for coordinated scheduling of networked microgrids in active distribution systems with hydrogen refueling and charging stations,” Int. J. Hydrogen Energy, vol. 71, pp. 1442–1455, 2024, doi: 10.1016/j.ijhydene.2024.05.364.
[32] M. U. Tahir, “Overview of multi-stage charging strategies for Li-ion batteries,” J. Energy Chem., vol. 84, pp. 228–241, 2023, doi: 10.1016/j.jechem.2023.05.023.
[33] W. Wang, “Designing an energy-efficient multi-stage selective electrodialysis process based on high-performance materials for lithium extraction,” J. Memb. Sci., vol. 675, 2023, doi: 10.1016/j.memsci.2023.121534.
[34] H. Li, “Acoustic emission characteristics of rock salt under multi-stage cyclic loading,” Int. J. Fatigue, vol. 176, 2023, doi: 10.1016/j.ijfatigue.2023.107911.
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Verdi Yasin, Timbo Faritcan P. Siallagan
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
-a.jpg)
-b.jpg)
