Real-time people and vehicle detection from UAV imagery

  title={Real-time people and vehicle detection from UAV imagery},
  author={Anna Gaszczak and T. Breckon and Jiwan Han},
  booktitle={Electronic Imaging},
A generic and robust approach for the real-time detection of people and vehicles from an Unmanned Aerial Vehicle (UAV) is an important goal within the framework of fully autonomous UAV deployment for aerial reconnaissance and surveillance. Here we present an approach for the automatic detection of vehicles based on using multiple trained cascaded Haar classifiers with secondary confirmation in thermal imagery. Additionally we present a related approach for people detection in thermal imagery… 
Multi-modal People Detection from Aerial Video
It is shown that, by fusing visible and infrared imagery and tracking detections over time, the false positive rate is reduced to a minimum and the existing state of the art part-based people detection algorithms can be used with considerable success in aerial imagery.
Pedestrian Detection and Tracking from Low-Resolution Unmanned Aerial Vehicle Thermal Imagery
This research proposes an approach for pedestrian tracking that employs the feature tracker with the update of detected pedestrian location to track pedestrian objects from the registered videos and extracts the motion trajectory data.
Real-time person detection using an embedded system for UAV Search and Rescue operations
This paper applies colour segmentation and graph theory techniques in order to locate a static target in an image that differs in colour to its surroundings and develops a real-time victim detection algorithm to address this problem.
Robust Vehicle Detection in Aerial Images Using Bag-of-Words and Orientation Aware Scanning
A novel approach to automatically detect and count cars in different aerial images, which can be satellite or unmanned aerial vehicle (UAV) images, based on the bag-of-words (BoW) model is presented, which is applicable to a wider range of aerial images.
UAV-Based Vehicle Detection by Multi-source Images
The rich information from the multi-source data is fully exploited in the proposed detection framework to further improve the accuracy of the detection result.
Real Time Human Detection and Localization Using Consumer Grade Camera and Commercial UAV
Based on flying height - human size relationship and tradeoff between detection accuracy vs computation time, the study came up with optimal parameters for OpenCV’s cv2.cascadeClassifier.
Salient object detection approach in UAV video
This proposed real-time approach for the detection of salient objects on road, e.g. stationary and moving vehicle or people, is proposed, which is based on region segmentation and saliency detection within related domains and performs excellently in isolated rural environments.
Real-Time Vehicle-Detection Method in Bird-View Unmanned-Aerial-Vehicle Imagery
The DRFBNet300 trained on the UAV-cars dataset with SIP method operates at 33 FPS, enabling real-time vehicle detection, and obtains the highest AP score at altitudes of 20–50 m.
Detection and Analysis Methods for unmanned aerial Vehicle Images
The proposed methodological contributions include a methodology which detects classes of objects by exploiting a nonlinear filter which combines image gradient features at different orders and Gaussian process (GP) modeling; a novel strategy to describe extremely high resolution images using various representation and matching strategies.
Marine object detection in UAV full-motion video
A generic object detection system that maintains a high level of performance without making prior assumptions about foreground-background characteristics and does not experience abrupt performance degradation when subjected to variations in lighting, background characteristics, video quality, abrupt changes in video perspective, size, appearance and number of the targets.


Autonomous Real-time Vehicle Detection from a Medium-Level UAV
A generic and robust approach for the detection of road vehicles from an Unmanned Aerial Vehicle (UAV) is an important goal within the framework of fully autonomous UAV deployment for aerial
Human Body Detection and Geolocalization for UAV Search and Rescue Missions Using Color and Thermal Imagery
A technique which allows detecting humans at a high frame rate on standard hardware onboard an autonomous UAV in a real-world outdoor environment using thermal and color imagery and a map of points of interest is built.
Vehicle and Person Tracking in Aerial Videos
Two tracking approaches from Sarnoff Corporation to detect moving vehicles and person in the videos taken from aerial platform or plane are presented and a HOG (Hierarchy Of Gradient) based vehicle v.s. person classifier is designed and integrated with the tracking post-processing.
A Hybrid Moving Object Detection Method for Aerial Images
A hybrid method for moving object detection in aerial videos that compensated the ego motion of airborne vehicle by feature-point based image alignment on consecutive frames, and applied an accumulative frame differencing method to detect the pixels with motion.
Geometric Constraints for Human Detection in Aerial Imagery
This paper proposes a method for detecting humans in imagery taken from a UAV that bypasses the global motion compensation process, and allows for detection of stationary and slow moving humans, while avoiding the search across the entire image, which makes it more accurate and very fast.
Person Tracking in UAV Video
This work found that the best results were obtained using a combination of intensity thresholding (for IR imagery), motion compensation, interest-point detection and correspondence, and pattern classification for UAV person tracking.
Rapid object detection using a boosted cascade of simple features
  • Paul A. Viola, Michael J. Jones
  • Computer Science
    Proceedings of the 2001 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. CVPR 2001
  • 2001
A machine learning approach for visual object detection which is capable of processing images extremely rapidly and achieving high detection rates and the introduction of a new image representation called the "integral image" which allows the features used by the detector to be computed very quickly.
Human identity recognition in aerial images
This paper proposes an algorithm to address the novel problem of human identity recognition over a set of unordered low quality aerial images by implementing a weighted voter-candidate formulation and identifies the candidate with the highest weighted vote as the target.