國立虎尾科技大學 |

Improving the Performance of Yolo-Based Detection Algorithms for Small Object Detection in Uav-Taken Images = = Kucuk Nesne Tanima uZerine Kullanilan Yolo Tabanli Nesne Tanima Algoritmalarinin Iyilestirilmesi.

Record Type:	Language materials, manuscript : Monograph/item
Title/Author:	Improving the Performance of Yolo-Based Detection Algorithms for Small Object Detection in Uav-Taken Images =/
Reminder of title:	Kucuk Nesne Tanima uZerine Kullanilan Yolo Tabanli Nesne Tanima Algoritmalarinin Iyilestirilmesi.
remainder title:	Kucuk Nesne Tanima uZerine Kullanilan Yolo Tabanli Nesne Tanima Algoritmalarinin Iyilestirilmesi.
Author:	Sahin, Oyku.
Description:	1 online resource (121 pages)
Notes:	Source: Masters Abstracts International, Volume: 84-10.
Contained By:	Masters Abstracts International84-10.
Subject:	Cameras. -
Online resource:	click for full text (PQDT)
ISBN:	9798377692768

Improving the Performance of Yolo-Based Detection Algorithms for Small Object Detection in Uav-Taken Images = = Kucuk Nesne Tanima uZerine Kullanilan Yolo Tabanli Nesne Tanima Algoritmalarinin Iyilestirilmesi.
Sahin, Oyku.

Improving the Performance of Yolo-Based Detection Algorithms for Small Object Detection in Uav-Taken Images =Kucuk Nesne Tanima uZerine Kullanilan Yolo Tabanli Nesne Tanima Algoritmalarinin Iyilestirilmesi.Kucuk Nesne Tanima uZerine Kullanilan Yolo Tabanli Nesne Tanima Algoritmalarinin Iyilestirilmesi. - 1 online resource (121 pages)

Source: Masters Abstracts International, Volume: 84-10.

Thesis (M.S.)--Bilkent Universitesi (Turkey), 2023.

Includes bibliographical references

Recent advances in computer vision yield emerging novel applications for cameraequipped unmanned aerial vehicles such as object detection. The accuracy of the existing object detection solutions running on images acquired by Unmanned Aerial Vehicles (UAVs) is limited when compared to the performance of the object detection solutions designed for ground-taken images. Existing object detection solutions demonstrate lower performance on aerial datasets because of the reasons originating from the nature of the UAVs. These reasons can be summarized as: (i) the lack of large drone datasets with different types of objects, (ii) the larger variance in both scale and orientation of objects in drone images, and (iii) the difference in shape and texture of the features between the ground and the aerial images. Due to these reasons, YOLO-based models, a popular family of one-stage object detectors, perform insufficiently in UAV-based applications. In this thesis, two improved YOLO models: YOLODrone and YOLODrone+ are introduced for detecting objects in drone images. The performance of the models are tested on VisDrone2019 and SkyDataV1 datasets and improved results are reported when compared to the original YOLOv3 and YOLOv5 models.

Electronic reproduction.
Ann Arbor, Mich. :
ProQuest,
2024

Mode of access: World Wide Web

ISBN: 9798377692768Subjects--Topical Terms:

812890
Cameras.
Index Terms--Genre/Form:

554714
Electronic books.

Improving the Performance of Yolo-Based Detection Algorithms for Small Object Detection in Uav-Taken Images = = Kucuk Nesne Tanima uZerine Kullanilan Yolo Tabanli Nesne Tanima Algoritmalarinin Iyilestirilmesi.
LDR:04050ntm a22004217 4500 001 1148150
005 20240916070013.5
006 m o d
007 cr bn ---uuuuu
008 250605s2023 xx obm 000 0 eng d
020 $a 9798377692768
035 $a (MiAaPQ)AAI30394584
035 $a (MiAaPQ)Bilkent_11693111198
035 $a AAI30394584
040 $a MiAaPQ $b eng $c MiAaPQ $d NTU
100 1 $a Sahin, Oyku. $3 1474061
245 1 0 $a Improving the Performance of Yolo-Based Detection Algorithms for Small Object Detection in Uav-Taken Images = $b Kucuk Nesne Tanima uZerine Kullanilan Yolo Tabanli Nesne Tanima Algoritmalarinin Iyilestirilmesi.
246 3 1 $a Kucuk Nesne Tanima uZerine Kullanilan Yolo Tabanli Nesne Tanima Algoritmalarinin Iyilestirilmesi.
264 0 $c 2023
300 $a 1 online resource (121 pages)
336 $a text $b txt $2 rdacontent
337 $a computer $b c $2 rdamedia
338 $a online resource $b cr $2 rdacarrier
500 $a Source: Masters Abstracts International, Volume: 84-10.
500 $a Advisor: Korpeoglu, Ibrahim.
502 $a Thesis (M.S.)--Bilkent Universitesi (Turkey), 2023.
504 $a Includes bibliographical references
520 $a Recent advances in computer vision yield emerging novel applications for cameraequipped unmanned aerial vehicles such as object detection. The accuracy of the existing object detection solutions running on images acquired by Unmanned Aerial Vehicles (UAVs) is limited when compared to the performance of the object detection solutions designed for ground-taken images. Existing object detection solutions demonstrate lower performance on aerial datasets because of the reasons originating from the nature of the UAVs. These reasons can be summarized as: (i) the lack of large drone datasets with different types of objects, (ii) the larger variance in both scale and orientation of objects in drone images, and (iii) the difference in shape and texture of the features between the ground and the aerial images. Due to these reasons, YOLO-based models, a popular family of one-stage object detectors, perform insufficiently in UAV-based applications. In this thesis, two improved YOLO models: YOLODrone and YOLODrone+ are introduced for detecting objects in drone images. The performance of the models are tested on VisDrone2019 and SkyDataV1 datasets and improved results are reported when compared to the original YOLOv3 and YOLOv5 models.
520 $a Bilgisayarli gorunun onemli alanlarindan biri olan nesne tanima alanindaki son gelismeler, kamera donanimli Insansiz Hava Araclari (IHA) icin ortaya cikan yeni cozumler saglar. Insansiz Hava Araclari'ndan alinan goruntuler uzerinde calisan, nesne tanima cozumlerinin basarimi, yerden alinan goruntuler icin nesne tanima cozumlerinin performansi ile karsilastirildiginda daha siniridir. Mev- cut nesne tanima cozumleri, IHA'larin dogasindan kaynaklanan nedenlerle IHA veri kumelerinde daha dusuk performans gostermektedir. Bu nedenler: (i) farkli nesne boyutlarina sahip buyuk drone veri setlerinin olmamasi, (ii) drone goruntulerinin heni olcek hem de yonelim olarak daha buyuk varyanslara sahip olmasi (iii) yer ve hava arasindaki sekil ve doku ozelliklerindeki farklilik. Bu sebeplerden dolayi, tek kademeli nesne tanima yontemlerinin bir parcasi olan YOLO tabanli modeller, IHA tabanli uygulamalarda daha dusuk performans gostermektedir. Bu tezde, YOLO algoritmasi uzerine YOLODrone ve YOLO- Drone, IHA goruntulerindeki nesneleri tanimak icin gelistirilmistir. Modellerin performansi VisDrone2019 ve SkyDataV1 veri ktimelerinde test edilip; YOLOv3 ve YOLOv5 ile karsilastirildiginda sonuclarda iyilestirme gozlemlenmistir.
533 $a Electronic reproduction. $b Ann Arbor, Mich. : $c ProQuest, $d 2024
538 $a Mode of access: World Wide Web
650 4 $a Cameras. $3 812890
650 4 $a Deep learning. $3 1127428
650 4 $a Sensors. $3 1003702
650 4 $a Design. $3 595500
650 4 $a Unmanned aerial vehicles. $3 1372845
650 4 $a Boxes. $3 1465626
650 4 $a Drones. $3 1346227
650 4 $a Algorithms. $3 527865
650 4 $a Aerospace engineering. $3 686400
650 4 $a Engineering. $3 561152
650 4 $a Robotics. $3 561941
650 4 $a Transportation. $3 558117
655 7 $a Electronic books. $2 local $3 554714
690 $a 0389
690 $a 0729
690 $a 0538
690 $a 0800
690 $a 0537
690 $a 0771
690 $a 0709
710 2 $a ProQuest Information and Learning Co. $3 1178819
710 2 $a Bilkent Universitesi (Turkey). $3 1413631
773 0 $t Masters Abstracts International $g 84-10.
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=30394584 $z click for full text (PQDT)