Keywords
lightweight feature calibration
illumination shift
paired simulation
reproducibility
Abstract
We evaluated lightweight feature calibration for edge plant-disease detection under 20% illumination shift. A deterministic paired simulation generated 56 cases and preserved a boundary-condition stratum. Mean balanced accuracy changed from 0.585 to 0.629; the paired difference was +0.045 (95% interval +0.043 to +0.047). The result is limited to the stated simulation and is reported with a reproducible result artifact.
References
Zhao, R., Deng, F., Que, H., Liu, M., Yue, X., & Mu, L. (2026). A Lightweight Real-Time Tomato Leaf Disease Detection System for Edge-Based Smart Agriculture. Sensors, 26(11), 3474. https://doi.org/10.3390/s26113474
Manjhvar, A. K., & Parmula, R. (2025). A Lightweight Deep Learning Architecture for Potato Leaf Disease Detection: A Comprehensive Survey. https://doi.org/10.1101/2025.11.16.688731
Mane, D., Deore, M., Ashtagi, R., Shinde, S., & Gurav, Y. (2023). Basil plant leaf disease detection using amalgam based deep learning models. Journal of Autonomous Intelligence, 7(1). https://doi.org/10.32629/jai.v7i1.1002
Yuvraj Narayan Gholap (2026). Plant Leaf Disease Detection Using Machine Learning and Deep Learning: A Review and Experimental Study. Journal of Intelligent Decision Making and Information Science, 3(2s), 142-153. https://doi.org/10.59543/jidmis.v3.635
M Suleman Basha (2025). Paddy Leaf Disease Detection using Deep Learning. Power System Technology, 49(1), 960-970. https://doi.org/10.52783/pst.1644
Arafat, I. S., Aswath, S., & Nishath, S. M. H. (2024). Automatic Early Detection of Tomato Leaf Disease using IoT and Deep Learning. https://doi.org/10.21203/rs.3.rs-3958296/v1
Sulaiya, N. (2025). Enhanced Plant Leaf Disease Detection Using CLAHE-Processed Images and Custom CNN Deep Learning Architecture. International Journal for Research in Applied Science and Engineering Technology, 13(9), 906-915. https://doi.org/10.22214/ijraset.2025.74149
Huang, Z., Fan, S., Ma, C., Yang, Y., Li, P., Li, W., & Zou, H. (2026). Deep learning enables edge deployment for citrus leaf disease recognition in natural orchard scenes. Frontiers in Plant Science, 17, 1896487. https://doi.org/10.3389/fpls.2026.1896487
Salauddin, M., Rashed, M., & Kakon, M. I. (2025). Deep Learning-Based CNN Approach for Tomato Leaf Disease Detection. https://doi.org/10.20944/preprints202506.1251.v1
Peng, Y., & Wang, Y. (2022). Leaf disease image retrieval with object detection and deep metric learning. Frontiers in Plant Science, 13, 963302. https://doi.org/10.3389/fpls.2022.963302