Jornal Paper: Multimodal Optical Biosensing and 3D-CNN Fusion for Basil Water Stress Phenotyping
I am pleased to announce the publication of our latest paper in Agronomy, titled “Multimodal Optical Biosensing and 3D-CNN Fusion for Phenotyping Physiological Responses of Basil Under Water Deficit Stress”.
This work was a collaborative effort with Kyung Hee University, Republic of Korea.
- Title: Multimodal Optical Biosensing and 3D-CNN Fusion for Phenotyping Physiological Responses of Basil Under Water Deficit Stress
- Authors: Yu-Jin Jeon, Hyoung Seok Kim, Taek Sung Lee, Soo Hyun Park, Heesup Yun, and Dae-Hyun Jung
- Journal: Agronomy (MDPI), 2026, 16(1), 55
- Publication Date: December 24, 2025
- DOI: 10.3390/agronomy16010055
Research Summary
This paper presents a non-destructive framework for monitoring water-deficit responses in basil using deep learning. We developed a fusion model combining RGB, Depth, and Chlorophyll Fluorescence (CF) imaging with a 3D Convolutional Neural Network (3D-CNN).
Key findings include:
- The 3D-CNN model achieved a remarkable 96.9% classification accuracy in water stress level tasks (Normal, Resistance, and Recovery).
- This approach effectively captures spatial and temporal-spectral features, outperforming traditional 2D-CNN and machine learning models.
- The results demonstrate the scalability of multi-modal information merging for precision irrigation and agricultural monitoring.
My opinion:
- The study suggests that adopting more advanced fusion techniques, such as Transformers, could further enhance the precision of physiological response phenotyping.
My Contribution:
- Provided editorial feedback and proofread the manuscript to ensure technical clarity and academic rigor.
Read the Full Paper
You can access the full publication here: 👉 MDPI Agronomy - Full Article