A Data-Driven Multi-Residue Risk Classification Framework for Pesticide Contamination in Vegetables Using Hybrid Deep Learning

Chaithra Reddy; Joseph Michael Jerard

doi:10.48084/etasr.17576

Authors

Chaithra Reddy Presidency University, India
Joseph Michael Jerard Presidency University, India

Volume: 16 | Issue: 3 | Pages: 35638-35643 | June 2026 | https://doi.org/10.48084/etasr.17576

Received: 15 January 2026 | Revised: 11 March 2026 and 30 March 2026 | Accepted: 1 April 2026 | Online: 5 May 2026

Corresponding author: Joseph Michael Jerard

Abstract

The increasing presence of pesticide residues in vegetables poses a major threat to public health, and there is an urgent need to develop efficient, accurate, and scalable detection methods. Traditional analytical techniques such as Gas Chromatography (GC) and Liquid Chromatography–Mass Spectrometry (LC–MS) offer high sensitivity but are expensive, labor-intensive, and unsuitable for large-scale or real-time screening applications. Recent advances in spectroscopy, machine vision, and Machine Learning (ML) show promise; however, most existing models rely on handcrafted or shallow features and fail to capture the complex spatial, textural, and thermal variations associated with multi-residue contamination. In this direction, the present study proposes a data-driven hybrid deep learning framework for multi-residue risk classification in vegetables using thermal imaging. This framework integrates ΔT-based thermal image preprocessing, Gray-Level Co-occurrence Matrix (GLCM) texture descriptors, and deep feature embeddings extracted from InceptionV3, thereby forming a comprehensive hybrid feature vector. This fused representation is classified by a custom neural network trained with categorical focal loss to mitigate class imbalance and optimized using a cosine-decay learning rate to enhance convergence stability. Experimental evaluation on a custom thermal vegetable image dataset resulted in 84.97% validation accuracy and a loss of 0.0373, outperforming conventional Convolutional Neural Networks (CNNs) and other shallow classifiers. The model demonstrated good generalization with balanced precision and recall on contamination classes, supported by a well-converged training–validation performance and confusion matrix analysis. These results highlight the efficacy of this framework for non-destructive, real-time, and scalable pesticide contamination risk classification, pointing to its potential for deployment in automated food safety monitoring and smart agricultural inspection systems.

Keywords:

hybrid deep learning framework, pesticide residue classification, thermal image analysis, Gray-Level Co-occurrence Matrix (GLCM), InceptionV3 feature extraction, ΔT thermal normalization, cosine-decay learning rate scheduling, categorical focal loss, food safety monitoring systems, multi-residue risk assessment

References

N. Evarist, N. Deborah, G. Birungi, N. K. Caroline, and B. J. M. Kule, "A Model for Detecting the Presence of Pesticide Residues in Edible Parts of Tomatoes, Cabbages, Carrots, and Green Pepper Vegetables," Artificial Intelligence and Applications, vol. 2, no. 3, pp. 196–203, Jan. 2024.

J. P. Chmielewski, E. Wszelaczyńska, J. Pobereżny, B. Gworek, A. Walosik, and M. Florek-Łuszczki, "Effect of consumption of vegetables contaminated with pesticides on consumers’ health – risk analysis," Annals of Agricultural and Environmental Medicine, vol. 32, no. 3, pp. 346–352, Sept. 2025.

D. I. Popescu, C. M. Oprita, R. Tamaian, and V.-C. Niculescu, "Consumer Safety and Pesticide Residues: Evaluating Mitigation Protocols for Greengrocery," Journal of Xenobiotics, vol. 14, no. 4, pp. 1638–1669, Nov. 2024.

Ö. F. Özbek, T. Balkan, and K. Kara, "Pesticide Residues in Raisin and Health Risk Assessment," Yuzuncu Yıl University Journal of Agricultural Sciences, vol. 35, no. 2, pp. 248–258, June 2025.

N. Sridhar et al., "Sustainable Plant Disease Management with Real-Time Crop Optimization," Engineering, Technology & Applied Science Research, vol. 15, no. 6, pp. 30580–30585, Dec. 2025.

Z. Dashtbozorgi, M. K. Ramezani, S. W. Husain, P. Abrumand-Azar, and M. Morowati, "Validation of Matrix Matched Calibration for Analysis of Insecticide and Fungicid Residues in Cucumber and Tomato Using Quechers Sample Preparation Followed by Gas Chromatography-Mass Spectrometry," Journal of the Chilean Chemical Society, vol. 58, no. 2, pp. 1701–1705, 2013.

S. Majumder, A. Kumar, S. Debnath, Abhinay, A. N. Singh, and T. K. Behera, "Development of an advanced analytical technique for detecting multiple pesticide residues in vegetables through liquid chromatography tandem mass spectroscopy (LC-MS/MS)," Journal of Environmental Science and Health, Part B, vol. 59, no. 10, pp. 663–677, Oct. 2024.

Y. Mei, M. Huang, W. Jiang, G. Yuan, and X. Li, "Formula model established through field application of glass slides and rapid analysis of pesticide residues in vegetables by ion mobility spectrometry," Food Control, vol. 160, June 2024, Art. no. 110316.

W. Qi, Y. Tian, D. Lu, and B. Chen, "Research Progress of Applying Infrared Spectroscopy Technology for Detection of Toxic and Harmful Substances in Food," Foods, vol. 11, no. 7, Mar. 2022, Art. no. 930.

S. Sharma et al., "Raman Spectroscopy-Based Chemometrics for Pesticide Residue Detection: Current Approaches and Future Challenges," ACS Agricultural Science & Technology, vol. 4, no. 4, pp. 389–404, Apr. 2024.

F. Peng, S. Huang, Q. Chen, N. Tong, and Y. Wu, "Rapid Detection of Pesticide Residues in Leaf Vegetables by SERS Technology," Sensors, vol. 25, no. 16, Aug. 2025, Art. no. 4912.

C. N. Ndung’u, M. I. Kaniu, J. M. Wanjohi, K. O. Odongo, L. W. Kiruri, and K. A. Kaduki, "Feasibility for rapid on-site screening of pesticide residues in fresh produce using machine learning-assisted diffuse reflectance spectroscopy," Food and Humanity, vol. 2, May 2024, Art. no. 100204.

N. Sateesh, N. Rawat, S. S. Damre, W. Patel, H. Patil, and R. Maranan, "A Classification Model for Pesticide Residues in Food Based On Machine Learning Methods," in 2024 5th International Conference on Innovative Trends in Information Technology, Kottayam, India, 2024, pp. 1–7.

C. N. Ndung’u, K. A. Kaduki, M. I. Kaniu, and L. W. Kiruri, "Enhanced Detection of Pesticide Residues Using Two-Dimensional Raman Correlation Spectroscopy and Machine Learning," Applied Spectroscopy Practica, vol. 2, no. 4, Dec. 2024, Art. no. 27551857241303466.

S. Wold, M. Sjöström, and L. Eriksson, "PLS-regression: a basic tool of chemometrics," Chemometrics and Intelligent Laboratory Systems, vol. 58, no. 2, pp. 109–130, Oct. 2001.

L. Sun, X. Cui, X. Fan, X. Suo, B. Fan, and X. Zhang, "Automatic detection of pesticide residues on the surface of lettuce leaves using images of feature wavelengths spectrum," Frontiers in Plant Science, vol. 13, Jan. 2023, Art. no. 929999.

B. M. Nicolaï et al., "Nondestructive measurement of fruit and vegetable quality by means of NIR spectroscopy: A review," Postharvest Biology and Technology, vol. 46, no. 2, pp. 99–118, Nov. 2007.

N. Yaemsuk and S. Yammen, "Development of a Portable NIR Spectrometer for Detecting Pesticide Residues," Asian Health, Science and Technology Reports, vol. 32, no. 1, pp. 32–48, Feb. 2024.

S. Castillo-Girones, S. Munera, M. Martínez-Sober, J. Blasco, S. Cubero, and J. Gómez-Sanchis, "Artificial Neural Networks in Agriculture, the core of artificial intelligence: What, When, and Why," Computers and Electronics in Agriculture, vol. 230, Mar. 2025, Art. no. 109938.

"Vegetable chemical residue detection dataset." Kaggle. [Online]. Available: https://www.kaggle.com/datasets/vegetabledataset/mancozeb-and-other-chemical-residues.