MOBILE-BASED PEST DETECTION AND ALERT SYSTEM FOR SESAME PLANTS USING DEEP LEARNING ALGORITHM, SPECIFICALLY A CONVOLUTIONAL NEURAL NETWORK (CNN)

Author

Ofoh Onyinyechi franklyn, Tsokwa Idibore Danasebe, Mercy Danbeki, Daniel Rimamtatang


Abstract

 Sesame (Sesamum indicum L.) is a major oilseed crop and a critical foreign exchange earner for Nigeria and West Africa. However, its production is severely threatened by foliar diseases particularly bacterial blight and phyllody capable of causing yield losses of 40–60% when not detected early. Existing disease management practices rely heavily on manual visual inspection, a method that is labor-intensive, subjective, and inadequate given the limited reach of agricultural extension services in rural farming communities. This study addresses the complete absence of any mobile or artificial intelligence-based diagnostic tool specifically designed for sesame diseases by designing, developing, and deploying Sesame Scan a mobile-based sesame plant pest detection and alert system powered by a custom Convolutional Neural Network (CNN). A dataset of 640 labeled sesame leaf images spanning three classes Bacterial Blight Infected, Healthy Leaf, and Phyllody Infected was curated from diverse field conditions. A custom CNN architecture comprising three progressive convolutional blocks was trained from scratch on this dataset using the Adam optimizer with categorical cross-entropy loss, and augmented with structured data preprocessing, on-the-fly augmentation, and training callbacks including Early Stopping, Model Checkpoint, and Reduce LR On Plateau. The trained model achieved an overall classification accuracy of 97.9%, with macro-average precision, recall, and F1-score of 97.7% across all three disease classes. The model was exported to TensorFlow Lite format (63 MB) and integrated into a cross-platform mobile application built with React Native and Expo Go, enabling fully offline on-device inference on both iOS and Android smartphones without internet connectivity. The Sesame Scan application comprises three functional modules: a Scan Module for real-time disease diagnosis, an Analytics Dashboard for longitudinal disease prevalence monitoring, and an Explore Module for delivering disease management advisory content. The system demonstrates that a compact, custom-trained CNN, combined with a thoughtfully curated crop-specific dataset and a lightweight mobile deployment pipeline, can deliver precision agriculture capabilities directly to smallholder sesame farmers bridging a critical technological gap and contributing to improved food security and agricultural productivity in Nigeria and West Africa.



Keywords

Sesame disease detection; Convolutional Neural Network; Deep learning; Mobile application; Precision agriculture; TensorFlow Lite



Full Text:

Download Paper PDF


References


  • Abade, A., Ferreira, P. A., & de Barros Vidal, F. (2021). Plant diseases recognition on images using convolutional neural networks: A systematic review. Computers and Electronics in Agriculture, 185, 106125. https://doi.org/10.1016/j.compag.2021.106125
  • Asif, M., Rayamajhi, A., & Mahmud, M. S. (2025). Technological progress toward peanut disease management: A review. Sensors, 25(4), 1255. https://doi.org/10.3390/s25041255
  • Atila, U., Ucar, M., Akyol, K., & Ucar, E. (2021). Plant leaf disease classification using EfficientNet deep learning model. Ecological Informatics, 61, 101182. https://doi.org/10.1016/j.ecoinf.2020.101182
  • Desoky, E. S. M., Alharbi, K., Rady, M. M., Elnahal, A. S., Selem, E., Arnaout, S. M., & Mansour, E. (2023). Physiological, biochemical, anatomical, and agronomic responses of sesame to exogenously applied polyamines under different irrigation regimes. Agronomy, 13(3), 875. https://doi.org/10.3390/agronomy13030875
  • Ferentinos, K. P. (2018). Deep learning models for plant disease detection and diagnosis. Computers and Electronics in Agriculture, 145, 311–318. https://doi.org/10.1016/j.compag.2018.01.009Girmay, A. B. (2018). Sesame production, challenges and opportunities in Ethiopia. Vegetos: An International Journal of Plant Research, 31(1), 51. https://doi.org/10.5958/2229-4473.2018.00008.5
  • Ibrahim, A. G., Aliyu, J., & Aba, D. A. (2020). Occurrence of sesame (Sesamum indicum) diseases and their effects on crop production in Nigeria. African Journal of Agricultural Research, 15(8), 1020–1030. https://doi.org/10.5897/AJAR2019.14432
  • Jasrotia, S., Yadav, J., Rajpal, N., Arora, M., & Chaudhary, J. (2023). Convolutional neural network-based maize plant disease identification. Procedia Computer Science, 218, 1712–1721. https://doi.org/10.1016/j.procs.2023.01.149
  • LeCun, Y., Bengio, Y., & Hinton, G. (2015). Deep learning. Nature, 521(7553), 436–444. https://doi.org/10.1038/nature14539
  • Lu, Y., Zhou, G., Hao, X., Wang, H., & Hu, K. (2024). Image recognition of rice leaf diseases using atrous convolutional neural network and improved transfer learning algorithm. Multimedia Tools and Applications, 83(5), 12799–12817. https://doi.org/10.1007/s11042-023-16122-9
  • Mohanty, S. P., Hughes, D. P., & Salathe, M. (2016). Using deep learning for image-based plant disease detection. Frontiers in Plant Science, 7, 1419. https://doi.org/10.3389/fpls.2016.01419
  • Ramcharan, A., McCloskey, P., Baranowski, K., Selvaraj, S., Ngambiri, M., Niraula, H., & Hughes, D. (2019). A mobile-based deep learning model for cassava disease diagnosis. Frontiers in Plant Science, 10, 272. https://doi.org/10.3389/fpls.2019.00272
  • Ransingh, N., Khamari, B., & Adhikary, N. K. (2021). Modern approaches for management of sesame diseases. In Innovative approaches in diagnosis and management of crop diseases(pp. 123–162). Apple Academic Press.
  • Saleem, M. H., Potgieter, J., & Arif, K. M. (2019). Plant disease detection and classification by deep learning. Plants, 8(11), 468. https://doi.org/10.3390/plants8110468

 

Share your valuable work from Social Media Buttons