Classification of Agricultural Pests Through Digital Images Using Deep Learning

Pedro Lucas de Oliveira Costa; Thiago Matheus de Oliveira Costa; Larissa Ferreira Rodrigues Moreira; Leandro Henrique Furtado Pinto Silva; João Fernando Mari

doi:10.22456/2175-2745.143520

Authors

Pedro Lucas de Oliveira Costa Universidade Federal de Viçosa (UFV) https://orcid.org/0009-0000-2269-8524
Thiago Matheus de Oliveira Costa Universidade Federal de Viçosa (UFV) https://orcid.org/0009-0000-2773-5670
Larissa Ferreira Rodrigues Moreira Universidade Federal de Viçosa (UFV) https://orcid.org/0000-0001-8947-9182
Leandro Henrique Furtado Pinto Silva Universidade Federal de Viçosa (UFV) https://orcid.org/0000-0002-5765-5206
João Fernando Mari Universidade Federal de Viçosa (UFV) https://orcid.org/0000-0001-8271-3802

DOI:

https://doi.org/10.22456/2175-2745.143520

Keywords:

Pest Classification, Deep Learning, Data Augmentation, Hyperparameter Optimization

Abstract

In the agricultural sector, pest detection is vital, and given the susceptibility of human analysis to errors, deep learning solutions such as Convolutional Neural Networks (CNNs) provide a promising alternative. Classifying insect pests is challenging due to the high variability among species across different regions and their various life stages. In this study, we evaluate several deep learning models and training strategies for automatic pest image classification. We analyze four CNN architectures—AlexNet, ResNet-50, EfficientNet, and Vision Transformer (ViT). Following a hyperparameter optimization step, the models were fine-tuned, and we examined the impact of four data augmentation strategies on classification performance using the Agricultural Pests dataset. ViT demonstrated superior performance, achieving an accuracy of 0.9574 without data augmentation. Although ViT did not benefit from data augmentation, these techniques proved essential for enhancing the performance of AlexNet, ResNet-50, and EfficientNet. Our findings underscore the potential of deep learning methods for pest classification, offering valuable tools to help professionals maintain crop quality and value.

Downloads

Download data is not yet available.

References

PALEI, S. et al. Precision agriculture: ML and DL-based detection and classification of agricultural pests. In: IEEE. 2023 2nd International Conference on Ambient Intelligence in Health Care (ICAIHC). 2023. p. 1–6. Disponível em: https://doi.org/10.1109/ICAIHC59020.2023.10431427.

ZHANG, W. et al. Research on the identification and detection of field pests in the complex background based on the rotation detection algorithm. Frontiers in Plant Science, Frontiers Media SA, v. 13, p. 1011499, 2022. Disponível em: https://doi.org/10.3389/fpls.2022.1011499.

NETO, R. G. P. et al. Enhancing green coffee quality assessment through deep learning. In: SBC. Anais do XVIII Workshop de Visão Computacional. 2023. p. 84–89. Disponível em: https://doi.org/10.5753/wvc.2023.27537.

ESGARIO, J. G. et al. An app to assist farmers in the identification of diseases and pests of coffee leaves using deep learning. Information Processing in Agriculture, Elsevier, v. 9, n. 1, p. 38–47, 2022. Disponível em: https://doi.org/10.1016/j.inpa.2021.01.004.

MOTA, G. S. de L. et al. Classifying pests in crop images using deep learning. In: SBC. Anais do XVIII Workshop de Visão Computacional. 2023. p. 42–47. Disponível em: https://doi.org/10.5753/wvc.2023.27530.

ULLAH, N. et al. An efficient approach for crops pests recognition and classification based on novel DeepPestNet deep learning model. IEEE Access, IEEE, v. 10, p. 73019–73032, 2022. Disponível em: https://doi.org/10.1109/ACCESS.2022.3189676.

KHANRAMAKI, M.; ASLI-ARDEH, E. A.; KOZEGAR, E. Citrus pests classification using an ensemble of deep learning models. Computers and Electronics in Agriculture, Elsevier, v. 186, p. 106192, 2021. Disponível em: https://doi.org/10.1016/j.compag.2021.106192.

WANG, Q. et al. InsectMamba: Insect pest classification with state space model. arXiv preprint arXiv:2404.03611, 2024. Disponível em: https://doi.org/10.48550/arXiv.2404.03611.

UTTARWAR, M. et al. A novel deep learning model for detection of agricultural pests and plant leaf diseases. In: IEEE. 2023 IEEE Asia-Pacific Conference on Computer Science and Data Engineering (CSDE). 2023. p. 1–6. Disponível em: https://doi.org/10.1109/CSDE59766.2023.10487697.

HUANG, Q. Comparison of deep transfer learning models for pest image classification in agriculture. Highlights in Science, Engineering and Technology, v. 94, p. 9–16, 2024. Disponível em: https://doi.org/10.54097/kxbxjn03.

BHATT, A. et al. Deep transfer learning based light-weight agricultural pest classification model using convolutional neural networks. In: IEEE. 2024 International Conference on Inventive Computation Technologies (ICICT). 2024. p. 175–180. Disponível em: https://doi.org/10.1109/ICICT60155.2024.10544420.

ABDULRAZZAQ, A. R.; TÜRKBEN, A. K.; KURNAZ, S. Detection and classification of agricultural pest and vermin using full-convolutional neural network. In: IEEE. 2023 7th International Symposium on Innovative Approaches in Smart Technologies (ISAS). 2023. p. 1–6. Disponível em: https://doi.org/10.1109/ISAS60782.2023.10391393.

KRIZHEVSKY, A.; SUTSKEVER, I.; HINTON, G. E. ImageNet classification with deep convolutional neural networks. Advances in Neural Information Processing Systems, v. 25, 2012. Disponível em: https://proceedings.neurips.cc/paper_files/paper/2012.

HE, K. et al. Deep residual learning for image recognition. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. [s.n.], 2016. p. 770–778. Disponível em: https://doi.org/10.48550/arXiv.1512.03385.

TAN, M.; LE, Q. EfficientNet: Rethinking model scaling for convolutional neural networks. In: PMLR. International Conference on Machine Learning. 2019. p. 6105–6114. Disponível em: https://doi.org/10.48550/arXiv.1905.11946.

DOSOVITSKIY, A. et al. An image is worth 16x16 words: Transformers for image recognition at scale. arXiv preprint arXiv:2010.11929, 2020. Disponível em: https://doi.org/10.48550/arXiv.2010.11929.