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

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

Generative morphodynamic forecasting enables robust zero-shot volumetric medical segmentation.

Medical image analysis·2026

Same author

Both chronological age and individual differences in aging are the two indispensable components for predicting biological age.

Computer methods in biomechanics and biomedical engineering·2026

Same author

ACE-ProtoNet: Adaptive covariance eigen-gate and uncertainty-aware prototype learning for coronary artery segmentation.

Medical image analysis·2026

Same author

Crop GraphRAG: pest and disease knowledge base Q&A system for sustainable crop protection.

Frontiers in plant science·2026

Same author

Mitigating Quality Deterioration of Reduced-Fat Pork Sausages During Cold Storage via Resistant Starch Incorporation: Gel Properties and Protein Conformation Study.

Gels (Basel, Switzerland)·2025

Same author

A short recorded pulse dataset for vascular age prediction in China.

Scientific data·2025

Search research articles

Related Experiment Video

Updated: Sep 4, 2025

Imaging and Analysis for Quantifying Maize (Zea mays) Abiotic Stress Phenotypes

Imaging and Analysis for Quantifying Maize (Zea mays) Abiotic Stress Phenotypes

Published on: March 28, 2025

An Industrial-Grade Solution for Crop Disease Image Detection Tasks.

Guowei Dai¹, Jingchao Fan^1,2

¹National Agriculture Science Data Center, Agricultural Information Institute, Chinese Academy of Agricultural Sciences, Beijing, China.

Frontiers in Plant Science

|July 14, 2022

Summary

This study introduces YOLO V5-CAcT, a novel network for fast and accurate crop disease detection. The model achieves 94.24% accuracy with a 72% reduction in inference time, aiding agricultural industrialization.

Keywords:

activate quantitative convolutional neural network crop disease detection knowledge distillation model compression model deployment

More Related Videos

Cereal Crop Ear Counting in Field Conditions Using Zenithal RGB Images

Cereal Crop Ear Counting in Field Conditions Using Zenithal RGB Images

Published on: February 2, 2019

Using Flatbed Scanners to Collect High-resolution Time-lapsed Images of the Arabidopsis Root Gravitropic Response

Using Flatbed Scanners to Collect High-resolution Time-lapsed Images of the Arabidopsis Root Gravitropic Response

Published on: January 25, 2014

Related Experiment Videos

Last Updated: Sep 4, 2025

Imaging and Analysis for Quantifying Maize (Zea mays) Abiotic Stress Phenotypes

Imaging and Analysis for Quantifying Maize (Zea mays) Abiotic Stress Phenotypes

Published on: March 28, 2025

Cereal Crop Ear Counting in Field Conditions Using Zenithal RGB Images

Cereal Crop Ear Counting in Field Conditions Using Zenithal RGB Images

Published on: February 2, 2019

Using Flatbed Scanners to Collect High-resolution Time-lapsed Images of the Arabidopsis Root Gravitropic Response

Using Flatbed Scanners to Collect High-resolution Time-lapsed Images of the Arabidopsis Root Gravitropic Response

Published on: January 25, 2014

Area of Science:

Agricultural Science
Computer Vision
Machine Learning

Background:

Accurate crop leaf disease detection is crucial for agricultural health and industrialization.
Current machine learning methods face challenges in recognition time consumption and accuracy.

Purpose of the Study:

To propose a novel network architecture, YOLO V5-CAcT, for efficient and accurate crop disease identification.
To optimize the model for deployment in agricultural industrialization settings.

Main Methods:

Utilized YOLO V5 as the base network, enhanced with Repeated Augmentation, FocalLoss, and SmoothBCE strategies.
Employed Early Stopping for improved model convergence.
Applied model pruning, knowledge distillation, and parameter compression (ActNN) for diverse hardware conditions.
Optimized using INT8 quantization and deployed on the NCNN platform.

Main Results:

Achieved an average recognition accuracy of 94.24% across 59 crop disease categories for 10 crop species.
Reduced model size by 88% and inference time by 72% (1.563 ms per sample).
Demonstrated significant performance advantages in accuracy and computational cost.

Conclusions:

The YOLO V5-CAcT model offers a robust solution for agricultural disease image detection.
The model's efficiency and accuracy meet the demands of agricultural industrialization.