Self-supervised Model using Transfer Learning to Predict Generic Disease in Paddy

In this study, we developed a self-supervised Domain Adaptation Technique (DAA) model built on transfer learning to address the challenging task of detecting a variety of diseases and pests in paddy crops. The model delivered high accuracy and F1-scores, demonstrating consistent reliability across different conditions. Through the combination of contrastive learning and effective pre- processing, it was able to extract meaningful features even when working with limited labeled data.For future work, we aim to extend this framework by incorporating an Adaptive Crop Stage-Aware Transfer Learning approach. This enhancement will enable the model to account for the specific growth stage of the crop during prediction, helping deliver more context-aware diagnoses and reducing false alarms. We also plan to expand the dataset with geo tagged images and seasonal information, improving the system’s ability to generalize across varying regions and farming practices. These upgrades will make the solution even more practical and scalable for real-world agricultural applications.

1.  Chen, Xinlei, and Kaiming He. “Exploring Simple Siamese Representation Learning.” Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2021, pp. 15750– 15758. 2.  Conde, Sofia, et al. “Rice Pests and Diseases Around the World: Literature-Based Assessment with Emphasis on Africa and Asia.” Agriculture, vol. 15, no. 7, 2025, p. 667, doi:10.3390/agriculture15070667. 3.  Ferentinos, Konstantinos P. “Deep Learning Models for Plant Disease Detection and Diagnosis.” Computers and Electronics in Agriculture, vol. 145, 2018, pp. 311–318. 4.  Kamilaris, Andreas, and Francesc X. Prenafeta-Boldú. “Deep Learning in Agriculture: A Survey.” Computers and Electronics in Agriculture, vol. 147, 2018, pp. 70–90. 5.  Liu, Bin, et al. “Identification of Apple Leaf Diseases Based on Deep Convolutional Neural Networks.” Symmetry, vol. 10, no. 1, 2018, p. 11. 6.  Mohanty, Sharada P., David P. Hughes, and Marcel Salathé. “Using Deep Learning for Image-Based Plant Disease Detection.” Frontiers in Plant Science, vol. 7, 2016, p. 1419. 7.  Ramcharan, Alex, et al. “Deep Learning for Image-Based Cassava Disease Detection.” Frontiers in Plant Science, vol. 8, 2017, p. 1852. 8.  M Suleman Basha , K Rathi , K Naga Harshitha , P Sudha Aruna , K Manisha” Paddy Leaf Disease Detection using Deep Learning”. 9.  Journal of visual and performing arts - PADDY leaf detection using YOLOV8- siava subramaniyan 10.  Rice disease classification using fusion vision approach – Naresh kumar.B & sakthivel s, journal: nature portfolio-2025