Abstract:Vehicle detection in images from unmanned aerial vehicles (UAVs) plays an important role in traffic surveillance and urban planning due to the popularity of UAVs. However, the class imbalance problem is an important factor that restricts the performance of vehicle detectors. There are two types of class imbalance in UAV images, i.e., foreground-background imbalance and foreground–foreground imbalance. For anchor-based single stage detector, as many ground truths cannot be assigned to corresponding anchors because of low intersection over union, it makes the foreground-background imbalance problem more severe. Therefore, we propose a novel bag-based single-stage detector, which treats each position on the feature map as a bag. A simple and adaptive definition of bags is proposed along with the positive sample definition method, which is utilized to ensure more ground truths can be assigned to proper bags. In addition, we utilize online hard example mining method to control the proportion of positive and negative samples during the training process. To address the foreground–foreground imbalance, we propose a novel data augmentation algorithm, which allows us to create appropriate visual context for under-represented class. Extensive experiments demonstrate the superiority of the proposed algorithm, compared with other state-of-the-art solutions. Impact Statement—Recently, unmanned aerial vehicles (UAVs) are widely used in intelligent transportation due to their low price and high flexibility, which makes vehicle detection in UAV images important for automatically gathering of traffic information. However, the class imbalance problem, which is common in object detection where some classes have far fewer frequencies in the dataset, has an adverse effect on the performance of vehicle detectors. The data augmentation method and deep learning based vehicle detector proposed in this article are able to reduce the negative impact and improve detection performance by at least 1.27% in mean average precision index. In addition, compared with algorithms with similar detection performance, our method is at least 15 ms faster. The proposed method can benefit users in a wide variety of applications including UAV transportation, traffic surveillance, and urban planning.

Bidirectional Feedback of Optimized Gaussian Mixture Model and Kernel Correlation Filter for Enhancing Simple Detection of Small Pixel Vehicles.

Robust vehicle tracking and detection from UAVs

Traffic Collisions Early Warning Aided by Small Unmanned Aerial Vehicle Companion

Road-Map Aided GM-PHD Filter for Multivehicle Tracking With Automotive Radar

Vehicle Detection and Motion Analysis in Low-Altitude Airborne Video under Urban Environment

Detection of two-wheeled vehicles based on Gaussian mixture model and AdaBoost algorithm

Robust Vehicle Detection in High-Resolution Aerial Images with Imbalanced Data

Enhanced Detection Model and Joint Scoring Strategy for Multi-Vehicle Tracking

A Multi-Scale Feature Fusion Based Lightweight Vehicle Target Detection Network on Aerial Optical Images

Vehicle detection and tracking in airborne videos by multi-motion layer analysis

Vehicle Counting Based on Vehicle Detection and Tracking from Aerial Videos

Real-Time Vehicle Detection from UAV Imagery

Lightweight detection network for arbitrary-oriented vehicles in UAV imagery via precise positional information encoding and bidirectional feature fusion

Accelerating Vehicle Detection in Low-Altitude Airborne Urban Video.

Road Boundary Estimation to Improve Vehicle Detection and Tracking in UAV Video

Mixed Road User Trajectory Extraction from Moving Aerial Videos Based on Convolution Neural Network Detection.

Automatic ground vehicle detection for fixed-wing UAV

Real-Time Vehicle Detection Method Based on Aerial Image in Complex Background

Enhancing UAV Detection in Surveillance Camera Videos through Spatiotemporal Information and Optical Flow

Vehicle Detection in UAV Traffic Video Based on Convolution Neural Network

High-Resolution Vehicle Trajectory Extraction and Denoising from Aerial Videos.