Explainable Multi-Scale Attention-Enhanced Vision Transformer for Robust Detection of Pest-Related Food Safety Hazards
International Journal of Medical Toxicology and Forensic Medicine,
Vol. 16 (2026),
1 Dey 2026
,
Page 1-7
https://doi.org/10.22037/ijmtfm.v16.53178
Abstract
Background: Agricultural diseases and insect pests pose a major threat to crop production. Initially, identifying and detecting pests could effectively reduce the economic impacts they cause. Some agricultural experts are concerned with deep learning techniques, as they have exhibited critical accuracy in image classification.
Methods: This study introduces a novel Multi-Scale Attention Enhanced Transformer Network for Pest Detection in Precision Agriculture (MSAETN-PDPA). The primary objective of the proposed framework is to effectively identify and classify categories of agricultural pests in agricultural field images. The proposed model employs a multi-scale attention-enhanced multi-axis vision transformer network to extract robust features from the input image. GradCAM++ is utilized to interpret the decision-making process of the deep learning technique employed for agricultural pest classification. Extensive simulation analysis is performed to ensure the improved proficiency of the MSAETN-PDPA approach.
Results: The comparative evaluation showed that the MSAETN-PDPA technique achieved 98.73% accuracy, outperforming existing methods.
Conclusion: The proposed model is useful for identifying food-related safety hazards.
- Agricultural pests, Image analysis, Deep learning, Multi-Axis vision transformer, Hyperparameter optimization
How to Cite
References
[1] KS G, P SK, A R, K G. Crop pest identification using deep network-based extracted features and MobileENet in smart agriculture. Land Degrad Dev. 2024;35(11):3642-52. [DOI: 10.1002/ldr.5157]
[2] Amulothu DVRT, Rodge RR, Hasan W, Gupta S. Machine learning for pest and disease detection in crops. In: Agriculture 4.0. CRC Press; 2024. p. 111-32. [DOI: 10.1201/9781003570219-6]
[3] Venkateswara SM, Padmanabhan J. Deep learning-based agricultural pest monitoring and classification. Sci Rep. 2025;15(1):8684. [DOI: 10.1038/s41598-025-92659-5]
[4] Amrani A, Diepeveen D, Murray D, Jones MG, Sohel F. Multi-task learning model for agricultural pest detection from crop-plant imagery: a Bayesian approach. Comput Electron Agric. 2024;218:108719. [DOI: 10.1016/j.compag.2024.108719]
[5] Li S, Yuan Z, Peng R, Leybourne D, Xue Q, Li Y, et al. An effective farmer-centered mobile intelligence solution using lightweight deep learning for integrated wheat pest management. J Ind Inf Integr. 2024;42:100705. [DOI: 10.1016/j.jii.2024.100705]
[6] Zhang M, Liu C, Li Z, Yin B. From convolutional networks to vision transformers: evolution of deep learning in agricultural pest and disease identification. Agronomy. 2025;15(5):1079. [DOI: 10.3390/agronomy15051079]
[7] Abinaya M, Daniel JA. Deploying YOLOv8m and deep Q-network for intelligent prediction of pests and soil-borne illness in sugarcane fields. Int J Comput Intell Syst. 2026. [DOI: 10.1007/s44196-025-01050-9]
[8] Yin S, Li H, Ivanović M, Chen T, Teng L. FNNMFF: crop pests and diseases detection based on fuzzy neural network and multilevel feature fusion in remote sensing images. Comput Sci Inf Syst. 2026;(00):12-12. [DOI: 10.2298/csis251109012y]
[9] Butt M. Deep learning for plant disease detection using image analysis [master’s thesis]. University of Eastern Finland; 2026. [Link]
[10] Baeza-Moreno L, Blanco-Carmona P, Hidalgo-Fort E, Martín-Clemente R, González-Carvajal R. AI-driven detection of tiny pests in foliage: integrating image processing and deep learning. Mach Learn Appl. 2025:100834. [DOI: 10.1016/j.mlwa.2025.100834]
[11] Wu W, Huang J, Zhang M, Li Y, Yu Q, Zhao Q. MSA-MaxNet: multi-scale attention enhanced multi-axis vision transformer network for medical image segmentation. J Cell Mol Med. 2024;28(24):e70315. [DOI: 10.1111/jcmm.70315]
[12] Saha N, Dutta A. WCM-GCN: identifying influential spreaders using graph convolutional networks on weighted directed networks based on spreading properties. Knowl Inf Syst. 2026;68(1):33. [DOI: 10.1007/s10115-025-02675-0]
[13] Agricultural pests image dataset [dataset on the Internet]. Kaggle; [date unknown] [cited 2026 Sep 21]. [Link]
[14] de Oliveira Costa PL, de Oliveira Costa TM, Moreira LFR, Silva LHFP, Mari JF. Classification of agricultural pests through digital images using deep learning. Rev Inform Teor Apl. 2025;32(1):18-25. [DOI: 10.22456/2175-2745.143520]
[15] Cen X, Lu S, Qian T. YOLO-LCE: a lightweight YOLOv8 model for agricultural pest detection. Agronomy. 2025;15(9):2022. [DOI: 10.3390/agronomy15092022]
[16] Peng H, Xu H, Shen G, Liu H, Guan X, Li M. A lightweight crop pest classification method based on improved MobileNet-V2 model. Agronomy. 2024;14(6):1334. [DOI: 10.3390/agronomy14061334]
- Abstract Viewed: 9 times
- 53178 Downloaded: 9 times