A Smart Multimodal Machine Learning System for Crop Health Monitoring and Disease Management

Author

R. Kanimozhi1, Dr. V. Maniraj

Abstract

 While traditional precision agriculture systems heavily depend on Internet-of-Things (IoT) hardware, in-field sensor networks face severe real-world bottlenecks including high installation costs, hardware degradation, power constraints, and limited spatial coverage. To bypass physical hardware reliance, this paper presents a Sensor-Less, Multi-Modal Machine Learning Framework for joint crop management optimization and plant disease diagnosis. The proposed system fuses three non-invasive, open-access data modalities: (1) Multispectral Satellite Imagery (Sentinel-2/Landsat-8) for spatial-temporal vegetation index extraction, (2) Gridded Meteorological & Evapotranspiration API Feeds for microclimate tracking, and (3) Ground-Level RGB Leaf Imagery captured via mobile smart devices. An Ensemble Gradient Boosting architecture (XGBoost/LightGBM) processes fusion vectors for crop yield prediction and smart irrigation scheduling, while an Attention-Guided Vision Transformer (ViT-Lite) processes visual leaf inputs for early disease identification. Benchmarking on open agricultural datasets demonstrates a 94.2% crop yield prediction accuracy (R2=0.921) and a 97.6% disease classification accuracy, while achieving an 88.4% correlation with ground-truth physical soil moisture probes. The framework provides scalable, cost-effective decision support for smallholder farming without physical infrastructure investments.

Keywords

Precision Agriculture, Multi-Modal Fusion, Remote Sensing, Machine Learning, Satellite Imagery, Plant Disease Identification, Non-IoT Architecture, Vision Transformers.

Full Text:

Download Paper PDF

References

 [1] L. Breiman, “Random Forests,” Machine Learning, vol. 45, no. 1, pp. 5–32, 2001, doi: 10.1023/A:1010933404324.

[2] T. Chen and C. Guestrin, “XGBoost: A Scalable Tree Boosting System,” in Proc. 22nd ACM SIGKDD Int. Conf. Knowledge Discovery and Data Mining, 2016, pp. 785–794, doi: 10.1145/2939672.2939785.

[3] A. Dosovitskiy et al., “An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale,” in Proc. Int. Conf. Learning Representations (ICLR), 2021.

[4] K. He, X. Zhang, S. Ren, and J. Sun, “Deep Residual Learning for Image Recognition,” in Proc. IEEE Conf. Computer Vision and Pattern Recognition (CVPR), 2016, pp. 770–778, doi: 10.1109/CVPR.2016.90.

[5] M. Tan and Q. V. Le, “EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks,” in Proc. 36th Int. Conf. Machine Learning (ICML), 2019, pp. 6105–6114.

[6] M. Drusch et al., “Sentinel-2: ESA's Optical High-Resolution Mission for GMES Operational Services,” Remote Sensing of Environment, vol. 120, pp. 25–36, 2012, doi: 10.1016/j.rse.2011.11.026.

[7] S. K. McFeeters, “The Use of the Normalized Difference Water Index (NDWI) in the Delineation of Open Water Features,” International Journal of Remote Sensing, vol. 17, no. 7, pp. 1425–1432, 1996, doi: 10.1080/01431169608948714.

[8] T. van Klompenburg, A. Kassahun, and C. Catal, “Crop Yield Prediction Using Machine Learning: A Systematic Literature Review,” Computers and Electronics in Agriculture, vol. 177, Art. no. 105709, 2020, doi: 10.1016/j.compag.2020.105709.

[9] S. Nigam and R. Jain, “Plant Disease Identification Using Deep Learning: A Review,” Indian Journal of Agricultural Sciences, vol. 90, no. 2, pp. 249–257, 2020, doi: 10.56093/ijas.v90i2.98996.

[10] M. Rahmati et al., “Soil Moisture Retrieval From Sentinel-1: Lessons Learned After More Than a Decade in Orbit,” Remote Sensing of Environment, vol. 333, Art. no. 115146, 2026, doi: 10.1016/j.rse.20