Thermal Imaging and Disease Detection in Hibiscus Plant

Abstract

The threat of plant diseases poses a significant challenge to agricultural productivity, especially in developing countries where small-scale farmers are highly vulnerable. To avoid crop loss and guarantee food security, early identification of plant stress and disease is crucial. Even if they work well, traditional diagnostic techniques take a lot of time and effort. This study explores the potential of thermal imaging as a non-invasive and efficient solution for early stress detection in Hibiscus plants. The experiment was conducted on a single potted hibiscus plant at Bikaner Technical University over a period of approximately two months (14th December 2024 to 5th February 2025). Initially kept outdoors with regular watering, the plant was moved to a closed indoor setting without sunlight and water from Day 9, allowing for observation of stress progression and disease emergence. Although the thermal dataset was recorded for 16 days, intermediate day observations confirmed consistent stress behavior. Four visual diseases—Leaf Spot, Rust Disease, Botrytis Blight, and Mosaic Virus—were noted, but due to the limited dataset, the prototype focuses on classifying plant health into four thermal stress categories: healthy, mild, significant, and critical. Thermal images were captured from top and front views, and a deep learning model based on MobileNetV2 was developed using a multi-view classification approach. The model was trained using Leave-One-Out Cross-Validation (LOOCV) to ensure robustness with constrained data. Instead of relying solely on traditional performance metrics, a confidence-based interpretation method was adopted to improve decision reliability. The prototype demonstrates the feasibility of using thermal imaging and deep learning for early, non-destructive plant stress classification, paving the way for smarter and more sustainable agricultural monitoring.

Introduction

I. Importance of Agriculture & Problem Statement

• Agriculture is vital for developing nations like India, contributing ~18% to GDP and employing over half of the rural population.

• With the global population rising, food production must increase, but plant diseases severely limit yields and threaten food security.

• Traditional disease detection methods are manual, slow, and subjective, calling for automated and real-time solutions.

II. Thermal Imaging as a Solution

• Thermal imaging captures infrared heat emissions from plants, revealing early stress indicators invisible to the human eye (e.g., water stress, stomatal changes).

• The study focuses on Hibiscus rosa-sinensis, applying thermal imaging and deep learning for classifying plant stress levels.

III. Experimental Setup

• Conducted at Bikaner Technical University (Dec 14, 2024 – Feb 5, 2025).

• A single hibiscus plant was monitored; stress was induced by removing water and sunlight after Day 9.

• Thermal images were taken from three angles (top, front, and leaf) over 16 days.

• Images were categorized into four stress levels: healthy, mild, significant, and critical.

IV. Deep Learning Approach

• A MobileNetV2 CNN model in TensorFlow was used to classify the thermal images.

• Employed Leave-One-Out Cross-Validation (LOOCV) for robust performance despite a small dataset.

• The model demonstrated a non-invasive, real-time, and cost-effective approach to plant health monitoring.

V. Literature Review Highlights

• Previous studies show the value of thermal imaging combined with deep learning for early stress detection in crops like tea, tomato, citrus, maize, and wheat.

• Common limitations include small datasets, limited disease types, and challenges in species-specific modeling.

• This study addresses these gaps by proposing a multi-view thermal classification model for hibiscus.

VI. Hibiscus Plant Overview

• Hibiscus rosa-sinensis is valued for ornamental, medicinal, and cultural uses.

• Contains antioxidants and bioactive compounds; used in herbal medicine, cosmetics, textiles, and eco-friendly packaging.

• Key features: large flowers, fibrous roots, and broad leaves; thrives in tropical climates.

• Health benefits include heart health, liver detox, immunity boost, and blood pressure regulation.

VII. Common Hibiscus Diseases

The plant is susceptible to diseases like:

• Leaf Spot, Powdery Mildew, Root Rot, Rust, Botrytis Blight, Mosaic Virus, Dieback, and Nematodes.

• Causes range from fungal and bacterial infections to poor watering or insect transmission.

• Preventive measures: fungicides, good airflow, virus-free planting material, crop rotation, and organic soil management.

VIII. Methodology Summary

• Controlled experiment simulating stress over time.

• Daily image collection from multiple views with consistent environmental control.

• Image dataset labeled with stress severity for model training and testing.

Conclusion

This study demonstrates that thermal imaging, combined with deep learning, is a viable, scalable, and efficient method for early stress detection in hibiscus plants. It offers an innovative approach to smart agriculture, enabling precision monitoring and proactive crop management, especially in resource-limited settings.

References

[1] S. Adão et al., \"Hyperspectral imaging: A review on UAV-based sensors, data processing and applications for agriculture and forestry,\" *Remote Sensing*, vol. 9, no. 11, pp. 1–34, Nov. 2017, doi: 10.3390/rs9111110.. [2] P. Borodachov, D. Hardin, and E. B. Saff, *Discrete Energy on Rectifiable Sets*. New York, NY, USA: Springer, 2019, doi: 10.1007/978-0-387-84808-7. [3] S. Sun, L. Di, A. Lin, Z. Sun, and Y. Shen, “Infrared thermal imaging-based plant diseases detection,” Environmental Research Communications, vol. 2, no. 9, pp. 1–10, Sep. 2020. DOI: 10.1088/2515-7620/abbc89. [4] L. K. Liu, W. B. Li, X. Zhang, and T. S. Liao, “Application of thermal imaging technology in plant disease and insect pest monitoring,” Computers and Electronics in Agriculture, vol. 137, pp. 254–263, Apr. 2017. DOI: 10.1016/j.compag.2017.04.012. [5] R. C. Y. S. Huang, H. T. Lin, and J. L. Chen, “Thermal imaging applications in plant disease detection,” Journal of Plant Pathology, vol. 102, no. 2, pp. 431–443, Jun. 2020. DOI: 10.1007/s42161-020-00496-9. [6] D. D. Jones, B. E. Kang, M. R. Smith, and T. J. Thompson, “Advances in deep learning for hyperspectral and multispectral image analysis: A review,” IEEE Transactions on Geoscience and Remote Sensing, vol. 60, pp. 1–25, Feb. 2022. DOI: 10.1109/TGRS.2022.3140274. [7] A. P. S. Ferreira, P. M. Silva, and R. G. Oliveira, “Machine learning techniques for plant disease detection using thermal imaging,” Artificial Intelligence in Agriculture, vol. 5, pp. 23–35, Aug. 2021. DOI: 10.1016/j.aiia.2021.08.002. [8] A. Smith and B. Johnson, “Deep learning applications in agricultural image processing,” Computers in Industry, vol. 120, pp. 103–119, Mar. 2020. DOI: 10.1016/j.compind.2020.103225. [9] L. Wang, X. Yang, and Q. Zhang, “Use of UAV thermal imaging for early detection of plant diseases,” Sensors, vol. 19, no. 20, pp. 1–15, Oct. 2019. DOI: 10.3390/s19204345. [10] A. K. Sinha, V. P. Singh, and R. K. Tiwari, “A survey on thermal imaging applications in precision agriculture,” IEEE Access, vol. 9, pp. 134567–134590, Jul. 2021. DOI: 10.1109/ACCESS.2021.3098472. [11] OpenAI, ChatGPT: Artificial Intelligence Chatbot, San Francisco, CA, USA. Available: https://openai.com/chatgpt, Accessed: Feb. 12, 2025. [12] P. Alagumariappan, N. Jamal Dewan, G. Muthukrishnan, B. K. B. Raju, R. A. A. Bilal, and V. Sankaran, “Intelligent plant disease identification system using Machine Learning,” Engineering Proceedings, 2020. DOI: 10.3390/ecsa-7-08160. [13] K. R. Mamatha, S. Singh, and S. A. Hariprasad, “Detection and analysis of plant leaf diseases using Convolutional Neural Network,” Journal of Chemical and Pharmaceutical Sciences, vol. 17, pp. 123–131, 2020. DOI: 10.1166/jctn.2020.8983. [14] N. Shelar, S. Shinde, S. Sawant, S. Dhumal, and K. Fakir, “Plant disease detection using CNN,” ITM Web Conferences, vol. 44, pp. 1–7, 2022. DOI: 10.1051/itmconf/20224403049. [15] G. Geetha, S. Samundeswari, G. Saranya, K. Meenakshi, and M. Nithya, \"Plant leaf disease classification and detection system using machine learning,\" *J. Phys.: Conf. Ser.*, vol. 1712, no. 1, pp. 1–7, 2020, doi: 10.1088/1742-6596/1712/1/012012. [16] C. Ciaburro and B. Venkateswaran, Neural Networks with R: Smart Models using CNN, RNN, Deep Learning, and Artificial Intelligence Principles, Packt Publishing Ltd, 2018. [17] M. Patil and R. Patil, \"Deep learning for image-based mango leaf disease detection,\" *Int. J. Recent Technol. Eng. (IJRTE)*, vol. 8, no. 3S3, Nov. 2019. ISSN: 2277-3878. [18] S. Kumar, D. Nandhini, S. Amutha, and S. P. Syed Ibrahim, \"Detection and identification of healthy and unhealthy sugarcane leaf using convolution neural network system,\" *Springer*, 2023, doi: 10.1007/s12046-023-02309-7.

Copyright

Copyright © 2025 Viraj Tripathi, Abhishek Purohit. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.