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

Difference from Background: Limit of Detection

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...

Light Acquisition

Light Acquisition

In order to produce glucose, plants need to capture sufficient light energy. Many modern plants have evolved leaves specialized for light acquisition. Leaves can be only millimeters in width or tens of meters wide, depending on the environment. Due to competition for sunlight, evolution has driven the evolution of increasingly larger leaves and taller plants, to avoid shading by their neighbors with contaminant elaboration of root architecture and mechanisms to transport water and nutrients.

You might also read

Related Articles

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

Sort by

Same author

Whole-Genome Sequencing Pilot of the Central Asian Genomic Diversity Project Reveals Distinct Histories, Adaptation, and Introgression.

Advanced science (Weinheim, Baden-Wurttemberg, Germany)·2026

Same author

Large-scale mitogenome analysis reveals complex maternal genetic connections between Sino-Tibetan- and Altaic-speaking populations.

Human genetics·2026

Same author

Conformational Snapshots of CydDC in a Native Lipid Bilayer Coupling Heme Transport to Antibiotic Resistance.

Advanced science (Weinheim, Baden-Wurttemberg, Germany)·2026

Same author

Contrasting ancestry patterns inferred from Y chromosome and mitochondrial DNA in Nanjing people from southwestern China.

Human genetics·2026

Same author

Tracing the genetic legacy of Altaic-speaking populations suggested long-distance migration and multi-source admixture shape the genomic diversity of Xibe and Daur.

Human genetics·2026

Same author

Large-scale human genomic resources drive innovations toward forensic intelligent source attribution and precision identification.

International journal of legal medicine·2026

Search research articles

Related Experiment Video

Updated: Sep 18, 2025

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

A Lightweight Multi-Scale Context Detail Network for Efficient Target Detection in Resource-Constrained Environments.

Kaipeng Wang¹, Guanglin He¹, Xinmin Li¹

¹Science and Technology on Electromechanical Dynamic Control Laboratory, Beijing Institute of Technology, Beijing 100081, China.

Sensors (Basel, Switzerland)

|June 27, 2025

Summary

This summary is machine-generated.

This study introduces MSCDNet, a lightweight deep learning model for efficient target detection in resource-limited edge computing environments. MSCDNet enhances accuracy for camouflaged targets and diverse conditions, outperforming existing models.

Keywords:

context-aware modulation edge computing lightweight neural network multi-scale feature fusion target detection

More Related Videos

A Real-Time Interactive System for Studying Confrontational Pursuit Behavior in Rodents

A Real-Time Interactive System for Studying Confrontational Pursuit Behavior in Rodents

Published on: May 16, 2025

Design and Analysis for Fall Detection System Simplification

Design and Analysis for Fall Detection System Simplification

Published on: April 6, 2020

Related Experiment Videos

Last Updated: Sep 18, 2025

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

A Real-Time Interactive System for Studying Confrontational Pursuit Behavior in Rodents

A Real-Time Interactive System for Studying Confrontational Pursuit Behavior in Rodents

Published on: May 16, 2025

Design and Analysis for Fall Detection System Simplification

Design and Analysis for Fall Detection System Simplification

Published on: April 6, 2020

Area of Science:

Computer Vision
Artificial Intelligence
Edge Computing

Background:

Target detection is challenging in resource-constrained environments due to camouflage, varied target sizes, and harsh conditions.
Edge computing demands efficient solutions with limited computational power.

Purpose of the Study:

To propose MSCDNet (Multi-Scale Context Detail Network), a lightweight architecture for efficient target detection in edge computing.
To address challenges of camouflage, scale variation, and environmental conditions in target detection.

Main Methods:

Developed MSCDNet, integrating Multi-Scale Fusion, Context Merge, and Detail Enhance Modules.
Evaluated MSCDNet on target detection tasks, comparing performance against YOLO-family variants and baseline models.
Conducted generalization tests on VisDrone2019 and BDD100K datasets.

Main Results:

MSCDNet achieved 40.1% mAP50-95, 86.1% precision, and 68.1% recall with 2.22 M parameters and 6.0 G FLOPs.
Outperformed YOLO variants by 1.9% in mAP50-95 and used 14% fewer parameters.
Demonstrated robustness with improved mAP on VisDrone2019 (+1.1%) and BDD100K (+1.2%).

Conclusions:

MSCDNet offers an effective and efficient solution for target detection in resource-limited tactical deployments.
The model's architecture is well-suited for edge computing scenarios requiring reliable performance.
MSCDNet provides a strong balance between accuracy and computational efficiency.