Home / Current Issue / Paper 1716969
Transfer Learning Based Plant Species Classification Using MobileNetV3 and PlantVillage Dataset
Subject area: Science,Engineering and Technology · Area of research: Artificial Intelligence and Machine Learning
DOI: https://doi.org/10.64388/IREV9I10-1716969
Abstract
Identification of plant species is vital for the advancement of various applications ranging from precision agriculture to crop diseases surveillance and biodiversity recording. While the task has historically been accomplished through manual assessment of plant characteristics by botanists or agronomists, this traditional approach poses considerable scalability issues due to reliance on experts’ knowledge and vulnerability to inter-observer variance when dealing with high-volume plant specimen analysis. Over the past decade, however, the use of data-driven techniques has increased substantially, leading to impressive improvements in plant image classification, which can be mainly attributed to convolutional neural networks’ ability to learn distinctive visual features from raw imagery. Motivated by the promising results of recent advancements in this area, the current study presents a plant species classification framework based on transfer learning with MobileNetV3 as the main neural network architecture and PlantVillage dataset as the training data. The chosen model was known for the ability to offer competitive accuracy with significantly reduced computation costs, which made it possible to use the model with low-performance computing resources without compromising results’ quality. To ensure proper functioning of the learning algorithm, the dataset was preprocessed in multiple ways, including resizing images to the necessary dimensions, normalising pixels’ values within the desired numerical range, and assessing classes distribution through visual inspection. Finally, the learning procedure was implemented on Google Colab, where model accuracy and crossentropy loss were calculated on a validation set after every epoch. With an impressive accuracy of 97.87
Keywords
Plant Species Classification, Transfer Learning, MobileNetV3, Convolutional Neural Networks, Deep Learning, PlantVillage Dataset, Image Classification, Precision Agriculture.
References
[1] S. Sladojevic, M. Arsenovic, A. Anderla, D. Culibrk, and D. Stefanovic, ”Deep Neural Networks Based Recognition of Plant Diseases by Leaf Image Classification,” Computational Intelligence and Neuroscience, vol. 2016, pp. 1–11, 2016.
[2] K. P. Ferentinos, ”Deep Learning Models for Plant Disease Detection and Diagnosis,” Computers and Electronics in Agriculture, vol. 145, pp. 311–318, 2018.
[3] M. Too, L. Yujian, S. Njuki, and L. Yingchun, ”A Comparative Study of Fine-Tuning Deep Learning Models for Plant Disease Identification,” Computers and Electronics in Agriculture, vol. 161, pp. 272–279, 2019. [4] A. Howard et al., ”Searching for MobileNetV3,” in Proc. IEEE Int. Conf. Computer Vision (ICCV), 2019, pp. 1314–1324.
[4] J. Deng, W. Dong, R. Socher, L. Li, and L. Fei-Fei, ”ImageNet: A LargeScale Hierarchical Image Database,” in Proc. IEEE Conf. Computer Vision and Pattern Recognition (CVPR), 2009.
[5] D. P. Hughes and M. Salathe, ”An Open Access Repository of Images on Plant Health to Enable the Development of Mobile Disease Diagnostics,” PlantVillage Dataset, 2015.
[6] I. Goodfellow, Y. Bengio, and A. Courville, Deep Learning, MIT Press, 2016.
[7] K. He, X. Zhang, S. Ren, and J. Sun, ”Deep Residual Learning for Image Recognition,” in Proc. IEEE Conf. Computer Vision and Pattern Recognition (CVPR), 2016.
[8] K. Simonyan and A. Zisserman, ”Very Deep Convolutional Networks for Large-Scale Image Recognition,” in Proc. Int. Conf. Learning Representations (ICLR), 2015.
[9] C. Szegedy et al., ”Going Deeper with Convolutions,” in Proc. IEEE Conf. Computer Vision and Pattern Recognition (CVPR), 2015.
[10] M. Sandler, A. Howard, M. Zhu, A. Zhmoginov, and L. Chen, ”MobileNetV2: Inverted Residuals and Linear Bottlenecks,” in Proc. IEEE Conf. Computer Vision and Pattern Recognition (CVPR), 2018.
[11] T. Tan and Q. Le, ”EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks,” in Proc. Int. Conf. Machine Learning (ICML), 2019.
[12] A. Krizhevsky, I. Sutskever, and G. Hinton, ”ImageNet Classification with Deep Convolutional Neural Networks,” in Advances in Neural Information Processing Systems (NIPS), 2012.
[13] P. Mohanty, D. Hughes, and M. Salathe, ”Using Deep Learning for Image-Based Plant Disease Detection,” Frontiers in Plant Science, vol. 7, p. 1419, 2016.
[14] R. Girshick, ”Fast R-CNN,” in Proc. IEEE Int. Conf. Computer Vision (ICCV), 2015.
How to cite this paper
@article{1716969,
author = {Nithya P},
title = {Transfer Learning Based Plant Species Classification Using MobileNetV3 and PlantVillage Dataset},
journal = {Iconic Research And Engineering Journals},
year = {2026},
volume = {9},
number = {10},
pages = {3570-3576},
issn = {2456-8880},
url = {https://www.irejournals.com/formatedpaper/1716969.pdf},
abstract = {Identification of plant species is vital for the advancement of various applications ranging from precision agriculture to crop diseases surveillance and biodiversity recording. While the task has historically been accomplished through manual assessment of plant characteristics by botanists or agronomists, this traditional approach poses considerable scalability issues due to reliance on experts’ knowledge and vulnerability to inter-observer variance when dealing with high-volume plant specimen analysis. Over the past decade, however, the use of data-driven techniques has increased substantially, leading to impressive improvements in plant image classification, which can be mainly attributed to convolutional neural networks’ ability to learn distinctive visual features from raw imagery. Motivated by the promising results of recent advancements in this area, the current study presents a plant species classification framework based on transfer learning with MobileNetV3 as the main neural network architecture and PlantVillage dataset as the training data. The chosen model was known for the ability to offer competitive accuracy with significantly reduced computation costs, which made it possible to use the model with low-performance computing resources without compromising results’ quality. To ensure proper functioning of the learning algorithm, the dataset was preprocessed in multiple ways, including resizing images to the necessary dimensions, normalising pixels’ values within the desired numerical range, and assessing classes distribution through visual inspection. Finally, the learning procedure was implemented on Google Colab, where model accuracy and crossentropy loss were calculated on a validation set after every epoch.
With an impressive accuracy of 97.87},
keywords = {Plant Species Classification, Transfer Learning, MobileNetV3, Convolutional Neural Networks, Deep Learning, PlantVillage Dataset, Image Classification, Precision Agriculture.},
month = {April},
doi = {https://doi.org/10.64388/IREV9I10-1716969}
}