Related Concept Videos
Difference from Background: Limit of Detection
The LOD indicates the presence or absence...
Light Acquisition
Force Classification
Contact and non-contact forces are two of the most widely used categories of forces. As the name suggests, contact forces require physical contact between two objects to act upon each other. Examples of contact forces include frictional,...
Detection of Black Holes
Their closest cousins are neutron stars, which are composed almost entirely of neutrons packed against each other, making them extremely dense. A neutron star has the same mass as the Sun but its diameter is only a few kilometers. Therefore, the escape velocity from their surface is close to the speed of light.
Not until the 1960s, when the first neutron...
Precipitation Titration: Endpoint Detection Methods
In the Volhard method, a standard excess of AgNO3 is first added to the...
Detection of Gross Error: The Q Test
You might also read
Related Articles
Articles linked to this work by shared authors, journal, and citation graph.
Modeling and Visual Simulation of Bifurcation Aneurysms Using Smoothed Particle Hydrodynamics and Murray's Law.
BPKEM: A biometric-based private key encryption and management framework for blockchain.
GATCF: Graph Attention Collaborative Filtering for Reliable Blockchain Services Selection in BaaS.
Related Experiment Video
Updated: Aug 27, 2025

Author Spotlight: UAV Remote Sensing for Efficient Invasive Plant Biomass Estimation
Published on: February 9, 2024
A Lightweight and Accurate UAV Detection Method Based on YOLOv4.
Hao Cai1, Yuanquan Xie1, Jianlong Xu1
1Department of Computer Science, Shantou University, Shantou 515041, China.
This study introduces a lightweight YOLOv4 model for accurate Unmanned Aerial Vehicle (UAV) detection. The optimized model achieves high accuracy and real-time performance, addressing limitations of existing algorithms.
Area of Science:
- Computer Science
- Artificial Intelligence
- Machine Learning
Background:
- Unmanned Aerial Vehicles (UAVs) are increasingly utilized in civilian and military applications.
- Current UAV detection algorithms often suffer from high parameter counts, hindering real-time performance and accuracy.
Purpose of the Study:
- To develop an accurate and lightweight Unmanned Aerial Vehicle (UAV) detection model.
- To address the challenge of real-time detection with high accuracy for UAVs.
Main Methods:
- A novel UAV detection model based on YOLOv4 was proposed.
- The YOLOv4 backbone was replaced with MobileNet, feature extraction networks were modified, and standard convolutions were replaced with depth-wise separable convolutions.
- A custom UAV dataset with 20,365 images across four types of UAVs was created for validation.
Main Results:
- The proposed lightweight YOLOv4 model achieved 82 FPS (frames per second) and 93.52% mAP (mean Average Precision).
- The model demonstrated a significant reduction in parameters compared to existing methods.
- One-stage detection methods were found to offer superior real-time performance and accuracy for UAV detection.
Conclusions:
- The developed lightweight YOLOv4 model provides an effective solution for efficient and accurate real-time UAV detection.
- The modifications significantly reduced model complexity while maintaining high detection accuracy.
- This research contributes to advancing real-time object detection capabilities for UAVs.

