Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Search research articles

Related Experiment Video

Updated: Jan 14, 2026

Author Spotlight: AI-Driven Trypanosome Species Detection from Microscopic Images

Author Spotlight: AI-Driven Trypanosome Species Detection from Microscopic Images

Published on: October 27, 2023

LDSL framework: a lightweight dual-stream learning framework for wheat disease detection.

Lei Feng^1,2, Mingliang Li¹, Guanshi Ye¹

¹College of Electrical and Information Engineering, Jilin Agricultural Science and Technology University, Jilin, 132101, China.

|October 24, 2025

Summary

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by

Same author

The Performance Evolution of Porous Asphalt Mixtures in Hot In-Place Recycling with the Addition of Different Rejuvenators.

Materials (Basel, Switzerland)·2026

Same author

Multi-dimensional laboratory characterization: VOCs emission and evolution in warm-mix synchronous rejuvenated (WMA-SR) asphalt pavement throughout construction stages.

Environmental pollution (Barking, Essex : 1987)·2026

Same author

Genome-wide association studies and QTL mapping for traits deviating from normal distribution.

National science review·2026

Same author

A Comparative Analysis of the Properties of Coal Liquefaction Residues and Limestone Fine Aggregates.

Materials (Basel, Switzerland)·2026

Same author

Spiritual leadership and service performance among Chinese flight attendants: The mediating effects of meaningful work and work engagement.

PloS one·2026

Same author

Graphene oxide-modified Ni<sub>1.5</sub>Mg<sub>1.5</sub>Al<sub>1</sub>-LDH/g-C<sub>3</sub>N<sub>4</sub> composites for efficient photocatalytic degradation toward ciprofloxacin.

Environmental research·2026

Same journal

Establishment of a protoplast isolation and transient transformation system for tung tree (Vernicia fordii).

Plant methods·2026

Same journal

Deep aerenchyma: a transformer-based pipeline for scalable phenotyping of rice root aerenchyma lacunae across environments.

Plant methods·2026

Same journal

Comparative analysis of SP3 and S-Trap sample preparation protocols for proteomic profiling associated with somatic embryogenesis efficiency in Olea europaea L.

Plant methods·2026

Same journal

DAPR-AM-Net: an end-to-end smart farming system powered by dual-attention progressive refinement and adaptive MixUp for explainable tomato leaf disease classification and forecasting.

Plant methods·2026

Same journal

Deep soil layers show the most pronounced genetic variation in wheat root length.

Plant methods·2026

Same journal

A three-dimensional reconstruction method for seedlings based on improved DIFIX3D.

Plant methods·2026

See all related articles

This summary is machine-generated.

A new lightweight dual-stream learning (LDSL) framework accurately detects wheat diseases using global and local feature analysis. This efficient method shows promise for real-world agricultural applications with limited resources.

Area of Science:

Agricultural Science
Computer Vision
Machine Learning

Background:

Wheat diseases significantly impact crop yield and quality.
Deep learning models struggle with complex backgrounds and varied lesion shapes in field conditions.
Existing methods face challenges with limited computational resources and edge device constraints.

Purpose of the Study:

To develop a lightweight and accurate wheat disease detection framework.
To address the limitations of current deep learning models in real-world agricultural scenarios.
To enable efficient disease recognition on edge devices.

Main Methods:

Introduced a lightweight dual-stream learning (LDSL) framework.
Employed a global-local dual-stream architecture for comprehensive feature extraction.

Keywords:

Computer vision Deep learning Precision agriculture Wheat disease detection

More Related Videos

Fully Automated Leg Tracking in Freely Moving Insects using Feature Learning Leg Segmentation and Tracking FLLIT

Fully Automated Leg Tracking in Freely Moving Insects using Feature Learning Leg Segmentation and Tracking FLLIT

Published on: April 23, 2020

Related Experiment Videos

Last Updated: Jan 14, 2026

Author Spotlight: AI-Driven Trypanosome Species Detection from Microscopic Images

Author Spotlight: AI-Driven Trypanosome Species Detection from Microscopic Images

Published on: October 27, 2023

Fully Automated Leg Tracking in Freely Moving Insects using Feature Learning Leg Segmentation and Tracking FLLIT

Fully Automated Leg Tracking in Freely Moving Insects using Feature Learning Leg Segmentation and Tracking FLLIT

Published on: April 23, 2020

Utilized a dynamic-static dual attention (DSDA) mechanism for fine-grained analysis.

Implemented Kullback-Leibler (KL) divergence perturbation for enhanced robustness.

Main Results:

Achieved high accuracy (94.44%), precision (94.47%), recall (94.44%), and F1-score (94.45%).
Outperformed mainstream models like ConvNeXt-T.
Demonstrated a lightweight design with 4.41M parameters and 1.71G FLOPs.
Showcased efficient performance on edge devices (NVIDIA Jetson Orin Nano) with low storage and memory requirements.

Conclusions:

The LDSL framework significantly improves wheat disease detection metrics.
The framework's low computational cost and parameter count make it suitable for practical deployment.
This research offers a novel and efficient solution for agricultural disease recognition.