Saved in:
Bibliographic Details
Main Authors: Eryuksel, Ogulcan, Ozfuttu, Kamil Anil, Akyon, Fatih Cagatay, Sahin, Kadir, Buyukborekci, Efe, Cavusoglu, Devrim, Altinuc, Sinan
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2407.00830
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910508640108544
author Eryuksel, Ogulcan
Ozfuttu, Kamil Anil
Akyon, Fatih Cagatay
Sahin, Kadir
Buyukborekci, Efe
Cavusoglu, Devrim
Altinuc, Sinan
author_facet Eryuksel, Ogulcan
Ozfuttu, Kamil Anil
Akyon, Fatih Cagatay
Sahin, Kadir
Buyukborekci, Efe
Cavusoglu, Devrim
Altinuc, Sinan
contents Drone detection is a challenging object detection task where visibility conditions and quality of the images may be unfavorable, and detections might become difficult due to complex backgrounds, small visible objects, and hard to distinguish objects. Both provide high confidence for drone detections, and eliminating false detections requires efficient algorithms and approaches. Our previous work, which uses YOLOv5, uses both real and synthetic data and a Kalman-based tracker to track the detections and increase their confidence using temporal information. Our current work improves on the previous approach by combining several improvements. We used a more diverse dataset combining multiple sources and combined with synthetic samples chosen from a large synthetic dataset based on the error analysis of the base model. Also, to obtain more resilient confidence scores for objects, we introduced a classification component that discriminates whether the object is a drone or not. Finally, we developed a more advanced scoring algorithm for object tracking that we use to adjust localization confidence. Furthermore, the proposed technique won 1st Place in the Drone vs. Bird Challenge (Workshop on Small-Drone Surveillance, Detection and Counteraction Techniques at ICIAP 2021).
format Preprint
id arxiv_https___arxiv_org_abs_2407_00830
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle DroBoost: An Intelligent Score and Model Boosting Method for Drone Detection
Eryuksel, Ogulcan
Ozfuttu, Kamil Anil
Akyon, Fatih Cagatay
Sahin, Kadir
Buyukborekci, Efe
Cavusoglu, Devrim
Altinuc, Sinan
Computer Vision and Pattern Recognition
Drone detection is a challenging object detection task where visibility conditions and quality of the images may be unfavorable, and detections might become difficult due to complex backgrounds, small visible objects, and hard to distinguish objects. Both provide high confidence for drone detections, and eliminating false detections requires efficient algorithms and approaches. Our previous work, which uses YOLOv5, uses both real and synthetic data and a Kalman-based tracker to track the detections and increase their confidence using temporal information. Our current work improves on the previous approach by combining several improvements. We used a more diverse dataset combining multiple sources and combined with synthetic samples chosen from a large synthetic dataset based on the error analysis of the base model. Also, to obtain more resilient confidence scores for objects, we introduced a classification component that discriminates whether the object is a drone or not. Finally, we developed a more advanced scoring algorithm for object tracking that we use to adjust localization confidence. Furthermore, the proposed technique won 1st Place in the Drone vs. Bird Challenge (Workshop on Small-Drone Surveillance, Detection and Counteraction Techniques at ICIAP 2021).
title DroBoost: An Intelligent Score and Model Boosting Method for Drone Detection
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2407.00830