Beeld op de omslag

Remote Sensing Imagery for Multi-Stage Vehicle Detection and Classification via YOLOv9 and Deep Learner

Bewaard in:
Gepubliceerd in:Computers, Materials, & Continua vol. 84, no. 3 (2025), p. 4491-4510
Hoofdauteur: Mudawi, Naif
Andere auteurs: Hanzla, Muhammad, Alazeb, Abdulwahab, Alshehri, Mohammed, Alhasson, Haifa, Alhammadi, Dina, Ahmad, Jalal
Gepubliceerd in:
Tech Science Press
Onderwerpen:
Traffic surveillance
Feature extraction
Background noise
Classification
Image resolution
Neural networks
Conditional random fields
Unmanned aerial vehicles
Remote sensing
Aerial photography
Calibration
Urban planning
Weather
Discriminant analysis
Object recognition
Deep learning
Machine learning
Real time
Statistical analysis
Cost effectiveness
Vehicles
Online toegang:Citation/Abstract
Full Text - PDF
Tags: Voeg label toe
Geen labels, Wees de eerste die dit record labelt!
Samenvatting:Unmanned Aerial Vehicles (UAVs) are increasingly employed in traffic surveillance, urban planning, and infrastructure monitoring due to their cost-effectiveness, flexibility, and high-resolution imaging. However, vehicle detection and classification in aerial imagery remain challenging due to scale variations from fluctuating UAV altitudes, frequent occlusions in dense traffic, and environmental noise, such as shadows and lighting inconsistencies. Traditional methods, including sliding-window searches and shallow learning techniques, struggle with computational inefficiency and robustness under dynamic conditions. To address these limitations, this study proposes a six-stage hierarchical framework integrating radiometric calibration, deep learning, and classical feature engineering. The workflow begins with radiometric calibration to normalize pixel intensities and mitigate sensor noise, followed by Conditional Random Field (CRF) segmentation to isolate vehicles. YOLOv9, equipped with a bi-directional feature pyramid network (BiFPN), ensures precise multi-scale object detection. Hybrid feature extraction employs Maximally Stable Extremal Regions (MSER) for stable contour detection, Binary Robust Independent Elementary Features (BRIEF) for texture encoding, and Affine-SIFT (ASIFT) for viewpoint invariance. Quadratic Discriminant Analysis (QDA) enhances feature discrimination, while a Probabilistic Neural Network (PNN) performs Bayesian probability-based classification. Tested on the Roundabout Aerial Imagery (15,474 images, 985K instances) and AU-AIR (32,823 instances, 7 classes) datasets, the model achieves state-of-the-art accuracy of 95.54% and 94.14%, respectively. Its superior performance in detecting small-scale vehicles and resolving occlusions highlights its potential for intelligent traffic systems. Future work will extend testing to nighttime and adverse weather conditions while optimizing real-time UAV inference.
ISSN:1546-2218
1546-2226
DOI:10.32604/cmc.2025.065490
Bron:Publicly Available Content Database