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

Search research articles

Related Experiment Video

Updated: May 28, 2026

Computer Vision-Based Biomass Estimation for Invasive Plants

Computer Vision-Based Biomass Estimation for Invasive Plants

Published on: February 9, 2024

Vehicle detection in aerial surveillance using dynamic Bayesian networks.

Hsu-Yung Cheng¹, Chih-Chia Weng, Yi-Ying Chen

¹Department of Computer Science and Information Engineering, National Central University, Chungli 320, Taiwan. chengsy@csie.ncu.edu.tw

IEEE Transactions on Image Processing : a Publication of the IEEE Signal Processing Society

|October 25, 2011

Summary

This study introduces a novel pixelwise classification system for automatic vehicle detection in aerial surveillance. The method effectively identifies vehicles by preserving spatial relationships and utilizing color and local features for improved accuracy.

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

Pseudotumor Cerebri Syndrome and Essential Thrombocythemia: A Case Report.

Clinical case reports·2026

Same author

Importation risks and local transmission of Japanese encephalitis virus in Taiwan.

One health (Amsterdam, Netherlands)·2026

Same author

Neuroprotective mechanism of acupuncture against brain injury during delayed thrombolysis for acute ischemic stroke.

Journal of integrative medicine·2026

Same author

Effects of Iron Supplementation during the Menstrual Period on Hemoglobin Levels and Fatigue in Female Athletes.

Journal of physiological investigation·2026

Same author

Lung-Targeting Cepharanthine Polymer Micelles Modified with Mannose: Effectiveness Against Acute Lung Injury Evaluated Using in vitro and in vivo Analyses.

Current drug metabolism·2025

Same author

Answer to comments on "Bamboo ecosystem services in 25 years: a systematic literature review of trends, insights, and knowledge gaps".

Environmental science and pollution research international·2025

Area of Science:

Computer Vision
Artificial Intelligence
Remote Sensing

Background:

Traditional vehicle detection in aerial surveillance relies on region-based or sliding window methods.
Existing frameworks often struggle with variations in image scale, angle, and lighting conditions inherent in aerial surveillance.

Purpose of the Study:

To develop an innovative, pixelwise classification system for automatic vehicle detection in aerial surveillance.
To overcome the limitations of conventional region-based and sliding window approaches.
To enhance the accuracy and adaptability of vehicle detection in diverse aerial imagery.

Main Methods:

A pixelwise classification approach is employed, preserving spatial relationships among neighboring pixels.
Vehicle color extraction utilizes a color transform for effective separation of vehicle and non-vehicle colors.

Related Experiment Videos

Last Updated: May 28, 2026

Computer Vision-Based Biomass Estimation for Invasive Plants

Computer Vision-Based Biomass Estimation for Invasive Plants

Published on: February 9, 2024

Edge detection incorporates moment-preserving techniques to automatically adjust Canny edge detector thresholds.

A dynamic Bayesian network (DBN) is constructed for classification, converting regional local features into quantitative observations.

Main Results:

The proposed system demonstrates flexibility and robust generalization capabilities across various aerial surveillance datasets.
Experimental results validate the effectiveness of the pixelwise classification method, even with images captured at different altitudes and camera angles.
The integration of color transforms and adaptive edge detection significantly improves detection accuracy.

Conclusions:

The developed pixelwise classification system offers a significant advancement in automatic vehicle detection for aerial surveillance.
The method's ability to preserve spatial context and adapt to varying image conditions makes it highly suitable for challenging real-world scenarios.
This approach provides a more accurate and adaptable solution compared to traditional vehicle detection techniques in aerial imagery.