Search research articles

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

Effectiveness of drying in digestive endoscope reprocessing: Systematic review on residual droplets and microorganisms.

Endoscopy international open·2026

Same author

A Review of Robotic Weeding Modalities for Site-Specific Weed Management.

Sensors (Basel, Switzerland)·2026

Same author

Development and validation of a diagnostic nomogram for Wagner grade≥2 diabetic foot ulcers in hospitalized patients with type 2 diabetes.

Frontiers in endocrinology·2026

Same author

Research progress on precision orchard yield estimation based on multi-source information perception sensors.

Frontiers in plant science·2026

Same author

Adaptive Sparse Self-Attention for Efficient Image Super-Resolution and Beyond.

IEEE transactions on pattern analysis and machine intelligence·2026

Same author

Pre-priming cell sheet therapy enabled by dynamic wrinkled electroactive substrate for muscle reconstruction.

Bioactive materials·2026

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 7, 2026

SwarmSight: Real-time Tracking of Insect Antenna Movements and Proboscis Extension Reflex Using a Common Preparation and Conventional Hardware

SwarmSight: Real-time Tracking of Insect Antenna Movements and Proboscis Extension Reflex Using a Common Preparation and Conventional Hardware

Published on: December 25, 2017

Multi-Strategy Improved Cantaloupe Pest Detection Algorithm.

Hongyan Zou¹, Zishuo Weng¹, Maocheng Zhao¹

¹College of Mechanical and Electronic Engineering, Nanjing Forestry University, Nanjing 210037, China.

|December 30, 2025

Summary

A new YOLOv12-based algorithm enhances pest detection in cantaloupe fields by improving feature fusion and incorporating an attention mechanism. This advanced model achieves higher accuracy and efficiency for effective pest management.

Keywords:

YOLOv12 cantaloupe pests crop protection deep learning object detection

More Related Videos

Flying Insect Detection and Classification with Inexpensive Sensors

Flying Insect Detection and Classification with Inexpensive Sensors

Published on: October 15, 2014

A Precise and Autonomous System for the Detection of Insect Emergence Patterns

A Precise and Autonomous System for the Detection of Insect Emergence Patterns

Published on: January 9, 2019

Related Experiment Videos

Last Updated: Jan 7, 2026

SwarmSight: Real-time Tracking of Insect Antenna Movements and Proboscis Extension Reflex Using a Common Preparation and Conventional Hardware

SwarmSight: Real-time Tracking of Insect Antenna Movements and Proboscis Extension Reflex Using a Common Preparation and Conventional Hardware

Published on: December 25, 2017

Flying Insect Detection and Classification with Inexpensive Sensors

Flying Insect Detection and Classification with Inexpensive Sensors

Published on: October 15, 2014

A Precise and Autonomous System for the Detection of Insect Emergence Patterns

A Precise and Autonomous System for the Detection of Insect Emergence Patterns

Published on: January 9, 2019

Area of Science:

Agricultural Science
Computer Vision
Machine Learning

Background:

Pest detection in cantaloupe fields faces challenges due to dense pest populations, varying sizes, and complex pest characteristics.
Accurate pest identification is crucial for effective crop protection and yield optimization in agriculture.

Purpose of the Study:

To develop an improved pest detection algorithm for cantaloupe leaves, addressing limitations of existing methods.
To enhance the accuracy and efficiency of pest identification in agricultural settings.

Main Methods:

Utilized the Melon Cantaloupe Pest dataset, involving image processing for data augmentation.
Modified the YOLOv12 model by integrating an EMA attention mechanism, optimizing feature fusion strategies (Concat and Detect layers), introducing the WIoU v3 loss function, and enhancing the C3k2 module.
Conducted ablation and generalization experiments on additional pest datasets (rice and corn).

Main Results:

The proposed multi-strategy dynamic feature fusion algorithm achieved high performance metrics: mAP50 (85.06%), Precision (86.76%), Recall (79.94%), mAP50-95 (54.15%), and F1-score (83.08%).
Outperformed comparative models in most key metrics, demonstrating superior pest detection capabilities.
Showed significant improvements over the original YOLOv12 model, with enhanced accuracy and reduced parameter count, indicating increased efficiency.

Conclusions:

The improved YOLOv12 model effectively enhances pest detection performance in diverse agricultural scenarios.
The proposed algorithm offers a valuable technological reference for developing advanced pest control strategies in cantaloupe cultivation and beyond.