Real Time Detection of Microbial Leaf Diseases Using Deep Learning and Edge Computing on Raspberry Pi 4

Abstract

Bacterial and fungal leaf diseases significantly impact the productivity of agricultural, which causing annually billions of dollars in crop losses and threatening global food security. Conventional detection methods even though effective, but they are labor intensive, consuming more time, and inappropriate for real time applications or large-scale ones. In order to address the limitations of other studies, this study proposes an AI solution that using a fine-tuned ResNet50 model trained on the PlantVillage dataset to classify the plant leaves as Healthy, Bacterial, or Fungal (Mold). The model was optimized using TensorFlow Lite and deployed on a Raspberry Pi 4, achieving 87% accuracy, a recall of 86%, and inference speeds around 1.2 to1.5 seconds per image. To enhance the overall generalization, the data augmentation techniques were applied which including rotation, flipping, and scaling. For early disease detection in agricultural and environmental applications, this research provides a scalable and a cost effective. Compared to traditional methods and other systems, this study provides faster inference speeds and lower costs, making it ideal for designs with limited resource.

Keywords

• Image classification
• Deep learning
• ResNet50
• Edge computing
• Image processing

References

1. Ahmed V, Jouini K, Tuama A, Korbaa O. 2024. Integrating deep and handcrafted features for enhanced remote sensing image classification. IEEE/ACS 21st Int Conf Comput Syst Appl, Sousse, Tunisia, pp: 1-8.
2. Albahar M. 2023. A survey on deep learning and its impact on agriculture: challenges and opportunities. Agricult, 13: 540.
3. Alexandra C, Andreea C, Mihaela T, Florin G, Cecilia C. 2023. Tackling the issue of healthcare associated infections through point-of-care devices. TrAC Trends Anal Chem, 161: 116983.
4. Anyu W, Xiao F, Guangyun H, Ying X, Tian Z, Xi Y. 2023. Recent advances in digital microfluidic chips for food safety analysis: preparation, mechanism and application. Trends Food Sci Technol, 134: 136-148.
5. Bansal M, Kumar M, Sachdeva M, Mittal A. 2023. Transfer learning for image classification using VGG19: caltech-101 image data set. J Ambient Intell Hum Comput, 14: 3609-3620.
6. Beznik T, Smyth P, Lannoy G, John A. 2022. Deep learning to detect bacterial colonies for the production of vaccines. Neurocomputing, 470: 427-431.
7. Deng F, Zhao Z, Wang R, Xiang C, Lv L, Duan Y. 2023. Rapid and online detection of foodborne bacteria via a novel ultraviolet photoionization time-of-flight mass spectrometry. J Agric Food Chem, 71: 10809-10818.
8. Dhaka V, Meena S, Rani G, Sinwar D, Kavita I, Woźniak M. 2021. A survey of deep convolutional neural networks applied for prediction of plant leaf diseases. Sensors, 21: 4749.

9. Durgun Y, Durgun M. 2024. Kenar hesaplama tabanlı, mikrodenetleyici entegreli, çok amaçlı ve düşük maliyetli modül geliştirilmesi: bakteriyel koloni sayımı örneği. Iğdır Univ Fen Bil Enst Der, 14: 531-543.
10. Geetabai S, Soundarya B, Gururaj H, Vinayakumar R. 2024. Classification of various plant leaf disease using pretrained convolutional neural network on imagenet. Open Agric J, 18.
11. Han Q, Siyuan W, Xinge Xi, Yingchao Z, Ying D, Yanbin Li, Jianhan L, Yuanjie L. 2024. Deep learning enhanced multiplex detection of viable foodborne pathogens in digital microfluidic chip. Biosens Bioelectron, 245: 115837.
12. Hasan M, Sundberg C, Hasan H, Kostov Y, Ge X, Choa F. 2023. Rapid bacterial detection and identification of bacterial strains using machine learning methods integrated with a portable multichannel fluorometer. IEEE Access, 11: 86112-86121.
13. Hasan MS, Marsafari M, Tolosa M, Andar A, Ramamurthy SS, Ge X, Kostov Y, Rao G. 2022. Rapid ultrasensitive and high-throughput bioburden detection: microfluidics and instrumentation. Anal Chem, 94: 8683-8692.
14. Hughes DP, Salathé M. 2015. An open access repository of images on plant health to enable the development of mobile disease diagnostics. arXiv, 1511.08060.
15. Kim E, Nam S, Hwang T, Lee J, Park J, Shim I, Kye H, Shin Y, Koo J. 2024. IoT-based tryptophan-like fluorescence portable device to monitor the indicators for microbial quality by E. coli and biochemical oxygen demand (BOD5). Water, 16: 3491.
16. Kumar V, Recupero DR, Riboni D, Helaoui R. 2020. Ensembling classical machine learning and deep learning approaches for morbidity identification from clinical notes. IEEE Access, 9: 7107-7126.
17. Premkumar S, Sigappi A. 2022. Functional framework for edge-based agricultural system. In: Ajith A, Sujata D, Joel J, editors. Intelligent data-centric systems, AI, edge and IoT-based smart agriculture. Academic Press, pp: 71-100.
18. Qi W, Zheng L, Wang S, Huang F, Liu Y, Jiang H, Lin J. 2021. A microfluidic biosensor for rapid and automatic detection of Salmonella using metal-organic framework and raspberry pi. Biosens Bioelectron, 178: 0956-5663.
19. Soundarya B, Poongodi C. 2025. A novel hybrid feature fusion approach using handcrafted features with transfer learning model for enhanced skin cancer classification. Comput Biol Med, 190: 0010-4825.
20. Sun L, Zhang M, Wang B, Tiwari P. 2023. Few-shot class-incremental learning for medical time series classification. IEEE J Biomed Health Inform, Early Access.
21. Upadhyay A, Chandel N, Singh K, Chakraborty S, Nandede B, Kumar M, Subeesh A, Upendar K, Salem A, Elbeltagi A. 2025. Deep learning and computer vision in plant disease detection: a comprehensive review of techniques, models, and trends in precision agriculture. Artif Intell Rev, 58: 92.
22. Xie M, Chen T, Cai Z, Lei B, Dong C. 2024. An all-in-one platform for on-site multiplex foodborne pathogen detection based on channel-digital hybrid microfluidics. Biosensors, 14: 50.
23. Zhang Y, Jiang H, Ye T, Juhas M. 2021. Deep learning for imaging and detection of microorganisms. Trends Microbiol, 29: 569-572.