Comparison of Deep Learning Frameworks For Rice Disease Mapping From UAV Multispectral Imaging

2026-06-04

AI Summary

This study addresses the challenge of efficiently and accurately segmenting bacterial leaf blight (BLB) in rice using unmanned aerial vehicle (UAV)-acquired multispectral imagery. The authors systematically evaluate multiple deep learning models under three input configurations—multispectral-only, multispectral combined with NDVI, and multispectral combined with NDRE—employing a consistent training protocol. Architectures based on both CNNs (U-Net, U-Net++, DeepLabV3+) and Transformers (SegFormer) are compared, utilizing backbones such as ResNet-101 and EfficientNet-B3/B7. Results indicate that lightweight CNN-based models offer greater reliability for real-world deployment, and incorporating vegetation indices consistently yields modest but stable performance gains. The best-performing model, U-Net++ with an EfficientNet-B3 backbone, achieves a mean Intersection-over-Union (mIoU) of 97.62%, while SegFormer, though faster at inference, exhibits slightly lower accuracy, underscoring the practical advantages of CNN architectures for disease mapping in agricultural remote sensing.

Abstract

In this study, UAV multispectral imagery is used to segment the severity of bacterial leaf blight (BLB) in rice using convolutional neural networks (CNNs) and transformer-based models. The evaluated architectures include U-Net with a ResNet- 101 encoder, U-Net++ with EfficientNet-B3 and EfficientNetB7, DeepLabV3+, and SegFormer, all trained under a common pipeline with three input configurations (multispectral only, multispectral+NDVI, and multispectral+NDRE). Experiments are conducted using the publicly available BLB dataset with performance reported using mean IoU (mIoU), mean F1 (mF1), mean accuracy (mAcc), precision, and recall. U-Net++ with EfficientNet-B3 achieved the highest performance, with an mIoU of 97.62%. SegFormer obtained lower segmentation accuracy but comparable inference speed. Overall, the results indicate that lightweight CNN backbones remain more reliable for operational BLB monitoring while integration of vegetation indices provides small and consistent improvements. The study also highlights the value of standardised UAV datasets to compare disease mapping methods and encourages the use of CNN architectures for field implementation.

Problem

• rice disease mapping
• bacterial leaf blight
• UAV multispectral imaging
• disease severity segmentation
• precision agriculture

Innovation

• UAV multispectral imaging
• rice disease mapping
• U-Net++
• vegetation indices
• deep learning comparison