Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Video

Updated: May 1, 2026

Author Spotlight: AI-Driven Trypanosome Species Detection from Microscopic Images
08:20

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

Published on: October 27, 2023

2.9K

Transformer augmented hybrid deep learning for explainable multi class pest classification.

Vivek Kumar Verma1, Ashish Kumar2, Varda Pareek3

  • 1Department of Information Technology, Manipal University Jaipur, Jaipur, Rajasthan, India.

Scientific Reports
|April 4, 2026
PubMed
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

Meta-analysis Comparing Combined Use of Eicosapentaenoic Acid and Statin to Statin Alone.

The American journal of cardiology·2019
Same author

Correction to "MEMS impedance flow cytometry designs for effective manipulation of micro entities in health care applications" [Biosens. Bioelectron], 2019, 142, 111526.

Biosensors & bioelectronics·2019
Same author

Structure-function relationship of H2A-H2B specific plant histone chaperones.

Cell stress & chaperones·2019
Same author

Scope and Limitations of γ-Valerolactone (GVL) as a Green Solvent to be Used with Base for Fmoc Removal in Solid Phase Peptide Synthesis.

Molecules (Basel, Switzerland)·2019
Same author

Joint Consensus Statement of the Indian National Association for Study of the Liver and Indian Radiological and Imaging Association for the Diagnosis and Imaging of Hepatocellular Carcinoma Incorporating Liver Imaging Reporting and Data System.

Journal of clinical and experimental hepatology·2019
Same author

Do we have enough evidence to support the use of SGLT2 inhibitors for MACE reduction in patients with advance kidney disease?

European journal of internal medicine·2019
Same journal

Serum vitamin D level and its association with vertigo frequency and severity in Meniere disease.

Scientific reports·2026
Same journal

PFA-Net: a physics-informed feature enhancement and attention network for interpretable bearing fault diagnosis under strong noise.

Scientific reports·2026
Same journal

Circulating inflammatory, redox, and apoptosis-related alterations in drug-naive idiopathic pulmonary fibrosis: an exploratory case-control study.

Scientific reports·2026
Same journal

A baseline-oriented dynamic aggregation approach for demand-side heterogeneous controllable resources.

Scientific reports·2026
Same journal

Temporal precision and accuracy in schizophrenia: an exploratory study.

Scientific reports·2026
Same journal

Prefrontal EEG spectral and nonlinear signatures of subthreshold depression during resting state and affectively valenced picture/video viewing: a participant-level analysis.

Scientific reports·2026
See all related articles
This summary is machine-generated.

Deep learning models, especially hybrid CNN-Transformer designs, show promise for identifying agricultural pests. Attention-augmented models like Hybrid EfficientNetV2-S+Transformer achieved high accuracy, aiding precision agriculture and global food security.

Area of Science:

  • Agricultural Science
  • Computer Science
  • Artificial Intelligence

Background:

  • Agricultural pests pose significant threats to global food security, necessitating accurate and early detection for effective management.
  • Manual pest scouting and conventional methods have limitations, driving the need for advanced digital solutions in agriculture.
  • Deep learning offers a promising approach for automated pest recognition, overcoming traditional monitoring challenges.

Purpose of the Study:

  • To comprehensively evaluate various deep learning architectures for multi-class pest classification across 19 species.
  • To investigate the impact of segmentation-driven preprocessing on image analysis for pest identification.
  • To identify the most effective deep learning models for enhancing precision agriculture systems.

Main Methods:

Keywords:
Agricultural pest classificationDeep learningEfficientNetHybrid CNN–Transformer modelsImage segmentationPrecision agricultureReliability assessmentSelf-attention mechanisms

More Related Videos

Author Spotlight: UAV Remote Sensing for Efficient Invasive Plant Biomass Estimation
08:47

Author Spotlight: UAV Remote Sensing for Efficient Invasive Plant Biomass Estimation

Published on: February 9, 2024

2.0K

Related Experiment Videos

Last Updated: May 1, 2026

Author Spotlight: AI-Driven Trypanosome Species Detection from Microscopic Images
08:20

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

Published on: October 27, 2023

2.9K
Author Spotlight: UAV Remote Sensing for Efficient Invasive Plant Biomass Estimation
08:47

Author Spotlight: UAV Remote Sensing for Efficient Invasive Plant Biomass Estimation

Published on: February 9, 2024

2.0K
  • Evaluated classical CNNs (MobileNetV2, VGG16), compound-scaled models (EfficientNet), residual architectures (ResNet50), and NAS models (Xception).
  • Developed and tested novel hybrid CNN-Transformer models, including attention mechanisms (CBAM).
  • Employed segmentation techniques (GrabCut, Watershed, SLIC, Felzenszwalb) for image preprocessing to isolate pests from backgrounds.

Main Results:

  • Attention-augmented hybrid models significantly outperformed standalone CNNs in pest classification.
  • The Hybrid EfficientNetV2-S+Transformer model achieved the highest performance: 0.8800 validation accuracy, 0.849 macro-F1, and 0.4560 validation loss.
  • Segmentation preprocessing effectively enhanced foreground isolation and reduced background complexity in field images.

Conclusions:

  • Combining convolutional feature extraction with global self-attention mechanisms is highly effective for multi-species pest classification.
  • The study provides valuable insights for developing intelligent systems for precision agriculture and pest management.
  • Advanced deep learning models offer a robust solution for improving early pest detection and safeguarding crop yields.