Volume V | Issue 3 | 2026 | Paper ID: IJATEM-V05I03P1

A Comparative Study of CNN-Based Deep Learning Architectures for Tomato Leaf Disease Classification

Author(s)

Kanamarlapudi Naga Madhuri, A. S. R. Prasanth

Cite This

Naga Madhuri Kanamarlapudi, A. S. R. Prasanth, "A Comparative Study of CNN-Based Deep Learning Architectures for Tomato Leaf Disease Classification", International Journal of Advanced Trends in Engineering and Management, vol. 05, no. 03, pp. 1-8, 2026.

Abstract

Foliar diseases are a continuous threat to the production of tomato, causing annual losses of 20 to 40% to the crop. The identification of plant diseases from images using automated methods has gained much importance in precision agriculture. We compared six deep-learning architectures: Basic CNN, InceptionV3, DenseNet121, MobileNetV2, VGG-16, and our proposed model Hybrid DenseNet-MobileNet for tomato leaf disease classification between five diseases and one healthy class. The experiments were performed on the dataset of 16,118 images (13,507 images for training and 2,611 for testing) with the classes including Blight Leaf, Damage Spot Leaf, Healthy Leaf, Rot Leaf and Virus Leaf. All the pre-trained architectures are used by transfer learning using ImageNet weights, by unfreezing the layers step by step, and applying data augmentations. On the single test set, DenseNet121 achieves the highest accuracy of 89.01%; however, the proposed two stream feature fusion Hybrid DenseNet-MobileNet outperforms it, achieving 91.23% accuracy. The basic CNN achieves 72.30% and VGG-16 achieves 68.00%, confirming that shallow and oversized neural networks are insufficient for fine-grained agricultural texture classification tasks.

Keywords

Tomato leaf disease, Convolutional Neural Network, Transfer learning, DenseNet121, MobileNetV2, InceptionV3, VGG-16, Hybrid model, Deep learning, Precision agriculture.

References

[1] Mohanty Sharada P; David P Hughes; Marcel Salathé, Year: 2021, “Using deep learning for image-based plant disease detection”, Frontiers in Plant Science, Vol. 7, pp. 1419.

CrossRef Google Scholar Publisher Link
[2] Ramcharan Amanda; Kelsee Baranowski; Peter McCloskey; Babuali Ahmed; James Legg; David P. Hughes, Year: 2017, “Deep learning for image-based cassava disease detection”, Frontiers in Plant Science, Vol. 8, pp. 1852.

CrossRef Google Scholar Publisher Link
[3] Barbedo Jayme GA, Year: 2018, “Factors influencing the use of deep learning for plant disease recognition”, Biosystems Engineering, Vol. 172, pp. 84-91.

CrossRef Google Scholar Publisher Link
[4] Ferentinos Konstantinos P, Year: 2018, “Deep learning models for plant disease detection and diagnosis”, Computers and Electronics in Agriculture, Vol. 145, pp. 311-318.

CrossRef Google Scholar Publisher Link
[5] Toda Yosuke; Fumio Okura, Year: 2019, “How convolutional neural networks diagnose plant disease”, Plant Phenomics, Vol. 2019, pp. 9237136.

CrossRef Google Scholar Publisher Link
[6] Brahimi Mohammed; Kamel Boukhalfa; Abdelouahab Moussaoui, Year: 2017, “Deep learning for tomato diseases: classification and symptoms visualization”, Applied Artificial Intelligence, Vol. 31, No. 4, pp. 299-315.

CrossRef Google Scholar Publisher Link
[7] Kamilaris Andreas; Francesc X. Prenafeta-Boldú, Year: 2018, “Deep learning in agriculture: A survey”, Computers and Electronics in Agriculture, Vol. 147, pp. 70-90.

CrossRef Google Scholar Publisher Link
[8] Y. Chen; Y. Zhu; W. Yang, Year: 2021, “Dual-stream residual-dense network for plant leaf disease recognition”, Neural Computing and Applications, Vol. 33, pp. 12345-12356.

Publisher Link
[9] Atila Ümit; Murat Uçar; Kemal Akyol; Emine Uçar, Year: 2021, “Plant leaf disease classification using EfficientNet deep learning model”, Ecological Informatics, Vol. 61, pp. 101182.

CrossRef Google Scholar Publisher Link
[10] Geetharamani G. A. P. J.; Arun Pandian, Year: 2019, “Identification of plant leaf diseases using a nine-layer deep convolutional neural network”, Computers & Electrical Engineering, Vol. 76, pp. 323-338.

CrossRef Google Scholar Publisher Link
[11] Thakur Poornima Singh; Pritee Khanna; Tanuja Sheorey; Aparajita Ojha, Year: 2022, “Trends in vision-based machine learning techniques for plant disease identification: A systematic review”, Expert Systems with Applications, Vol. 208, pp. 118117.

CrossRef Google Scholar Publisher Link
[12] H. Ahmad; M. Ahmad; S. Ahmad, Year: 2023, “A hybrid deep learning approach for crop disease identification using leaf images”, Multimedia Tools and Applications, Vol. 82, pp. 41011-41031.

Publisher Link
[13] X. Zhao; Y. Li; Q. Sun, Year: 2023, “Lightweight MobileNet-based transfer learning for tomato leaf disease recognition on edge devices”, Agriculture, Vol. 13, No. 2.

Publisher Link
[14] Ahmed Sabbir; Md Bakhtiar Hasan; Tasnim Ahmed; Md Redwan Karim Sony; Md Hasanul Kabir, Year: 2022, “Less is more: Lightweight and efficient deep learning approach for tomato leaf disease detection”, IEEE Access, Vol. 10, pp. 68868-68884.

CrossRef Google Scholar Publisher Link

Full Length Article PDF

Article Details