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 Concept Videos

Extraction: Advanced Methods00:56

Extraction: Advanced Methods

Metal ions can be separated from one another by complexation with organic ligands–the chelating agent– to form uncharged chelates. Here, the chelating agent must contain hydrophobic groups and behave as a weak acid, losing a proton to bind with the metal. Since most organic ligands used in this process are insoluble or undergo oxidation in the aqueous phase, the chelating agent is initially added to the organic phase and extracted into the aqueous phase. The metal-ligand complex is formed in...
Difference from Background: Limit of Detection01:05

Difference from Background: Limit of Detection

The limit of detection (LOD) is the smallest amount of analyte that can be distinguished from the background noise. The LOD value corresponds to the concentration at which the analyte signal is three times larger than the standard deviation of the blank signal. Below this value, the analyte signal cannot be differentiated from the background noise. It is calculated by dividing the calibration slope by 3 times the standard deviation of the blank signals.
The LOD indicates the presence or absence...
RNA-seq03:21

RNA-seq

RNA sequencing, or RNA-Seq, is a high-throughput sequencing technology used to study the transcriptome of a cell. Transcriptomics helps to interpret the functional elements of a genome and identify the molecular constituents of an organism. Additionally, it also helps in understanding the development of an organism and the occurrence of diseases. 
Before the discovery of RNA-seq, microarray-based methods and Sanger sequencing were used for transcriptome analysis. However, while microarray-based...
Censoring Survival Data01:09

Censoring Survival Data

Survival analysis is a statistical method used to analyze time-to-event data, often employed in fields such as medicine, engineering, and social sciences. One of the key challenges in survival analysis is dealing with incomplete data, a phenomenon known as "censoring." Censoring occurs when the event of interest (such as death, relapse, or system failure) has not occurred for some individuals by the end of the study period or is otherwise unobservable, and it might have many different reasons...

You might also read

Related Articles

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

Sort by
Same author

A targeted PCR approach for the detection of IOLA in canine infectious respiratory disease samples during an atypical CIRD outbreak in winter 2023.

Frontiers in veterinary science·2026
Same author

CAR T cell therapy selectively depletes disease-driving mutant calreticulin cells in xenotransplants and human organoid models of myelofibrosis.

Science translational medicine·2026
Same author

A lightweight graph-enhanced deep learning framework for explainable cucumber leaf disease diagnosis.

Scientific reports·2026
Same author

Benchmarking hybrid CNN and transformer backbones with graph convolution networks (GCN) for flower growth-stage classification.

Scientific reports·2026
Same author

Geographic clustering and population structures of <i>Campylobacter jejuni</i> and <i>Campylobacter coli</i> in South and Southeast Asian poultry systems.

Microbial genomics·2026
Same author

BDFlower: Growth stage flower image dataset for precision agriculture and floriculture.

Data in brief·2026

Related Experiment Videos

Annotation-efficient weed detection using DINOv3-distilled YOLOv12.

Saif Khan1, Osama Bin Qashem1, Mahmudul Hasan Hamim1

  • 1Department of Computer Science and Engineering, East West University, Dhaka, Bangladesh.

Scientific Reports
|July 3, 2026
PubMed
Summary

WEEDINO-YOLOv12 reduces the need for extensive data labeling in precision agriculture. This deep learning framework improves weed detection efficiency with less annotated data, aiding crop yield protection.

Keywords:
DINO label-efficient trainingDeep learningObject detectionYOLO self-supervised learningWeed detection

Related Experiment Videos

Area of Science:

  • Agricultural Science
  • Computer Vision
  • Machine Learning

Background:

  • Weed pressure significantly impacts global crop yields, estimated between 10-34%.
  • High costs associated with bounding-box annotation hinder the large-scale deployment of deep learning-based weed detection systems in diverse agricultural environments.
  • Annotation bottlenecks are a major challenge for precision agriculture adoption.

Purpose of the Study:

  • To address the annotation bottleneck in weed detection for precision agriculture.
  • To propose and evaluate a label-efficient framework, WEEDINO-YOLOv12, for weed detection.
  • To demonstrate the practical accessibility of the framework for end-users.

Main Methods:

  • WEEDINO-YOLOv12 utilizes feature-distribution distillation from a frozen DINOv3 ViT-B/16 teacher model to a YOLOv12n backbone.
  • The framework is trained on unlabeled agricultural imagery, followed by supervised fine-tuning on a limited labeled subset.
  • A controlled empirical benchmark compared WEEDINO-YOLOv12 against fully supervised, semi-supervised (Soft Teacher), and self-supervised (BYOL) training regimes.

Main Results:

  • WEEDINO-YOLOv12 demonstrated annotation-efficiency gains, improving mAP@0.5:0.95 from 0.6402 to 0.6517 on Roboflow Weeds and from 0.7987 to 0.8083 on CottonWeedDet12 at 20% labeled data.
  • While full-label supervision remained superior, the proposed method offered modest but consistent improvements in label efficiency.
  • High-resolution fine-tuning (896x896 pixels) further enhanced localization accuracy, independent of the distillation process.

Conclusions:

  • WEEDINO-YOLOv12 offers a practical solution to reduce annotation costs in precision agriculture weed detection.
  • The framework provides a viable alternative for improving weed detection model performance with limited labeled data.
  • A Streamlit-based prototype showcases the framework's accessibility for agronomists and precision agriculture users.