A Depth-wise Separable VGG19-Capsule Network for Enhanced Bell Pepper and Grape Leaf Disease Classification with Ensemble Activation
Abstract
Crop disease is a significant problem in the agricultural sector, leading to decreased food production and causing substantial economic losses for farmers in farming regions. Nowadays, computer vision and deep learning models can detect and diagnose leaf diseases in their early stages, which may assist farmers and contribute to ensuring food security. This research introduces a hybrid Depth-wise Separable VGG19 and Capsule Network (VGG19-CapsNet) architecture for automated leaf disease detection and classification in bell pepper and grape plants. The novel contribution lies in the enhanced VGG19 architecture, incorporating depth-wise separable convolution, batch normalization, and a 40% dropout by introducing convolutional layers before the primary capsule layer. The process involves extracting features from VGG19, flattening them into vectors, and utilizing them as input for the capsule layer. This ensures the capsule network effectively captures spatial information and preserves the hierarchical relationships between features. A noteworthy aspect of this research work is introducing an ensemble activation function, fusing Leaky Rectified Linear Unit (Leaky ReLU) and Gaussian Error Linear Unit (GELU). A hybrid architecture combining VGG19 and CapsNet, using DWSC and batch renormalization with a dropout rate of 0.4, a learning rate of 0.001, and a batch size of 9, successfully captures complex patterns for categorizing diseases in bell pepper and grape plants. The performance of the plant disease classification model is enhanced by using Leaky ReLU activation functions and GELU, which increase the non-linearity and ensemble learning of the VGG19 model. The proposed VGG19-CapsNet framework is developed and deployed in a 128-core Jetson Nano single-board computer with graphics processing support. The research outcomes set a benchmark for accuracy and present a paradigm shift in automated leaf disease classification. The benchmark datasets PlantifyDr, Plant village and custom dataset are used to train and develop the proposed VGG19-CapNet deep learning model. Through extensive comparative analyses on various datasets and field tests, the proposed architecture has demonstrated superior performance in terms of accuracy (99.81%, 99.84%), precision (99.84%, 99.84 %), recall (99.79%, 99.84%), sensitivity (99.94%, 99.84%), F1-score (99.81%, 99.84%), and AUC (1.0, 1.0) for bell pepper, and grape leaves across different datasets. It demonstrates the potential to transform agriculture with innovative methodologies tailored for bell pepper and grape diseases.
Keywords
Computer Vision, Image Analysis, Pattern Recognition, 3D Reconstruction, Image segmentation, Video and Image Sequence Analysis, Active Vision TrackingReferences
[1] S.F. Ahmed, M.S.B. Alam, M. Hassan et al., "Deep learning modelling techniques: current progress,applications, advantages, and challenges," Artificial Intelligence Review, vol. 56, pp. 13521-13617, 2023.https://doi.org/10.1007/s10462-023-10466-8.193Name1 et al. / Electronic Letters on Computer Vision and Image Analysis 24(1):179-194, 202https://doi.org/10.1007/s10462-023-10466-
[2] D. Koldasbayeva, P. Tregubova, M. Gasanov et al., "Challenges in data-driven geospatial model-ing for environmental research and practice," Nature Communications, vol. 15, p. 10700, 2024.https://doi.org/10.1038/s41467-024-55240-8.https://doi.org/10.1038/s41467-024-55240-
[3] https://www.kaggle.com/datasets/lavaman151/plantifydr-dataset,on January 1st 2024.Accesse
[4] https://github.com/attaullah/downsampled-plant-disease-dataset, Accessedon January 1st 2024
[5] https://www.kaggle.com/datasets/sravanneeli/plant-leaf-diseases-dataset-with-augmentation, Accessed on January 1st 2024
[6] M. Bhagat, D. Kumar, R. Mahmood, B. Pati, and M. Kumar, "Bell Pepper Leaf Disease ClassificationUsing CNN," Proc. of 2nd Int. Conf. on Data, Engineering, and Applications (IDEA), Bhopal, India,2020, pp. 1-5. doi: 10.1109/IDEA49133.2020.9170728.https://doi.org/10.1109/IDEA49133.2020.917072
[7] M. Ji, L. Zhang, and Q. Wu, "Automatic grape leaf disease identification via UnitedModel based on mul-tiple convolutional neural networks," Information Processing in Agriculture, vol. 7, no. 3, pp. 418-426,2020. https://doi.org/10.1016/j.inpa.2019.10.003.https://doi.org/10.1016/j.inpa.2019.10.00
[8] A. Diana Andrushia, T. Mary Neebha, A. Trephena Patricia et al., "Image-based disease classificationin grape leaves using convolutional capsule network," Soft Computing, vol. 27, pp. 1457-1470, 2023.https://doi.org/10.1007/s00500-022-07446-5.https://doi.org/10.1007/s00500-022-07446-
[9] H.P. Wei, Y.Y. Deng, F. Tang et al., "A Comparative Study of CNN- and Transformer-Based Vi-sual Style Transfer," Journal of Computer Science and Technology, vol. 37, pp. 601-614, 2022.https://doi.org/10.1007/s11390-022-2140-7.https://doi.org/10.1007/s11390-022-2140-
[10] H. Gao, X. Liu, M. Qu, and S. Huang, "PDANet: Self-Supervised Monocular Depth Estimation Us-ing Perceptual and Data Augmentation Consistency," Applied Sciences, vol. 11, no. 12, p. 5383, 2020.https://doi.org/10.3390/app11125383.https://doi.org/10.3390/app1112538
[11] P.K. Chundi, P. Liu, S. Park, S. Lee, and M. Seok, "FPGA-based Acceleration of Binary Neural NetworkTraining with Minimized Off-Chip Memory Access," Proc. of IEEE/ACM ISLPED, Lausanne, Switzer-land, 2019, pp. 1-6. doi: 10.1109/ISLPED.2019.8824805.https://doi.org/10.1109/ISLPED.2019.882480
[12] P. Alirezazadeh, M. Schirrmann, and F. Stolzenburg, "Improving Deep Learning-based Plant Dis-ease Classification with Attention Mechanism," Gesunde Pflanzen, vol. 75, pp. 49-59, 2023.https://doi.org/10.1007/s10343-022-00796-y.https://doi.org/10.1007/s10343-022-00796-
[13] M. Nawaz, T. Nazir, M.A. Khan, V. Rajinikanth, S. Kadry, "Plant Disease Classification Using VGG-19Based Faster-RCNN," in Advances in Computing and Data Sciences, ICACDS 2023, Springer CCIS, vol.1848. https://doi.org/10.1007/978-3-031-37940-6 23.https://doi.org/10.1007/978-3-031-37940-
[14] S. Parakh, M.S. Ashraf, N. Tripathi, K. Pant, M.S. Ansari, and P. Negi, "Detection of Bell PepperCrop Diseases Using Convolution Neural Network," Proc. of 2nd Int. Conf. on Technological Advance-ments in Computational Sciences (ICTACS), Tashkent, Uzbekistan, 2022, pp. 1-1. doi: 10.1109/IC-TACS56270.2022.10705132.https://doi.org/10.1109/ICTACS56270.2022.998806
[15] M. Bhagat, D. Kumar, and S. Kumar, "Bell pepper leaf disease classification with LBP and VGG-16 basedfused features and RF classifier," International Journal of Information Technology, vol. 15, pp. 465-475,2023. https://doi.org/10.1007/s41870-022-01136-z.194Name1 et al. / Electronic Letters on Computer Vision and Image Analysis 24(1):179-194, 2024https://doi.org/10.1007/s41870-022-01136-
[16] N. Kundu, G. Rani, and V.S. Dhaka, "A Comparative Analysis of Deep Learning Models Applied forDisease Classification in Bell Pepper," Proc. of 6th Int. Conf. on PDGC, Waknaghat, India, 2020, pp.243-247. doi: 10.1109/PDGC50313.2020.9315821.https://doi.org/10.1109/PDGC50313.2020.931582
[17] F. Jiang, Y. Lu, Y. Chen, D. Cai, and G. Li, "Image recognition of four rice leaf diseases based ondeep learning and support vector machine," Computers and Electronics in Agriculture, vol. 179, 2020,p. 105824. https://doi.org/10.1016/j.compag.2020.105824.https://doi.org/10.1016/j.compag.2020.10582
[18] Y. Kurmi, S. Gangwar, D. Agrawal et al., "Leaf image analysis-based crop diseases classification," Signal,Image and Video Processing (SIViP), vol. 15, pp. 589-597, 2021. https://doi.org/10.1007/s11760-020-01780-7.https://doi.org/10.1007/s11760-020-01780-
[19] P. Thakur, A. Chug, and A.P. Singh, "Plant Disease detection of Bell Pepper Plant Using Transfer Learn-ing over different Models," Proc. of 8th Int. Conf. on SPIN, Noida, India, 2021, pp. 384-389. doi:10.1109/SPIN52536.2021.9565945.https://doi.org/10.1109/SPIN52536.2021.956594
[20] G. Altan, "Performance Evaluation of Capsule Networks for Classification of Plant Leaf Diseases," In-ternational Journal of Applied Mathematics Electronics and Computers, vol. 8, no. 3, pp. 57-63, 2020.doi:10.18100/ijamec.797392.https://doi.org/10.18100/ijamec.79739
[21] M. Ye et al., "A Lightweight Model of VGG-16 for Remote Sensing Image Classification," IEEE J.Selected Topics in Applied Earth Observations and Remote Sensing, vol. 14, pp. 6916-6922, 2021. doi:10.1109/JSTARS.2021.3090085.https://doi.org/10.1109/JSTARS.2021.309008
[22] K. Deeba, A. Balakrishnan, M. Kumar et al., "A disease monitoring system using multi-class cap-sule network for agricultural enhancement in muskmelon," Multimedia Tools and Applications, 2024.https://doi.org/10.1007/s11042-024-18717-8.https://doi.org/10.1007/s11042-024-18717-
[23] M. Dai et al., "Pepper leaf disease recognition based on enhanced lightweight convolutional neural net-works," Frontiers in Plant Science, vol. 14, 2023. doi: 10.3389/fpls.2023.1230886.https://doi.org/10.3389/fpls.2023.123088
[24] M.P. Mathew, S. Elayidom, and V. Jagathyraj, "Disease Classification in Bell Pepper Plants Based onDeep Learning Network Architecture," Proc. of 2nd Int. Conf. for Innovation in Technology (INOCON),Bangalore, India, 2023, pp. 1-6. doi: 10.1109/INOCON57975.2023.10101269.https://doi.org/10.1109/INOCON57975.2023.1010126
[25] Y.A. Bezabih, A.O. Salau, B.M. Abuhayi et al., "CPD-CCNN: classification of pepper disease us-ing a concatenation of convolutional neural network models," Scientific Reports, vol. 13, 15581, 2023.https://doi.org/10.1038/s41598-023-42843-2.https://doi.org/10.1038/s41598-023-42843-2
Published
How to Cite
Downloads
Copyright (c) 2025 Abubeker K M, Midhun P Mathew , Sudheep Elayidom , Jagathy Raj V P

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