國立虎尾科技大學 |

A Deep Neural Network for Real Time Mesoscale Ocean Eddy Detection.

紀錄類型:	書目-語言資料,手稿 : Monograph/item
正題名/作者:	A Deep Neural Network for Real Time Mesoscale Ocean Eddy Detection./
作者:	Witherell, Brianna Kimberly.
面頁冊數:	1 online resource (86 pages)
附註:	Source: Masters Abstracts International, Volume: 84-11.
Contained By:	Masters Abstracts International84-11.
標題:	Ocean engineering. -
電子資源:	click for full text (PQDT)
ISBN:	9798379527808

A Deep Neural Network for Real Time Mesoscale Ocean Eddy Detection.
Witherell, Brianna Kimberly.

A Deep Neural Network for Real Time Mesoscale Ocean Eddy Detection. - 1 online resource (86 pages)

Source: Masters Abstracts International, Volume: 84-11.

Thesis (M.S.)--Southern Illinois University at Edwardsville, 2023.

Includes bibliographical references

Accurate and timely detection of mesoscale ocean eddies is paramount to ensure the safety and efficiency of maritime vessels. Existing methods of mesoscale ocean eddy detection exist but are limited in three ways: infrequent passes over a region of interest, insufficient scan swath size, and inadequate data resolution. To combat these issues, this work presents and validates an artificial intelligence based mesoscale ocean eddy detection algorithm that is trained on a novel data set of several thousand eddies gathered from imagery from the GOES-16 Advanced Baseline Imager (ABI). The ABI onboard GOES-16 shows promise for resolving issues from polar-orbiting satellites for quick and accurate mesoscale ocean eddy detection.There are three major steps in this work: data collection and preprocessing, training the object detection scheme, and validating the object detection scheme. The pre-processed data is input into an object detection scheme, which is well trained with 43,776 files from 152 days from March-July of 2022. This work proposes a Faster R-CNN object detection scheme with a MobileNet V3 feature extraction network. Several competing object detection scheme's performances were compared including: Single Shot Multibox Detector, Fully Convolutional One-stage Object Detection, and Focal Loss. It was found the proposed Faster R-CNN approach with a MobileNet V3 model backbone shows optimal performance in both accuracy and robustness, allowing for real-time mesoscale ocean eddy detection.

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

Mode of access: World Wide Web

ISBN: 9798379527808Subjects--Topical Terms:

857658
Ocean engineering.
Subjects--Index Terms:

Artificial intelligenceIndex Terms--Genre/Form:

554714
Electronic books.

A Deep Neural Network for Real Time Mesoscale Ocean Eddy Detection.
LDR:02868ntm a22003977 4500 001 1150251
005 20241028114727.5
006 m o d
007 cr bn ---uuuuu
008 250605s2023 xx obm 000 0 eng d
020 $a 9798379527808
035 $a (MiAaPQ)AAI30422198
035 $a AAI30422198
040 $a MiAaPQ $b eng $c MiAaPQ $d NTU
100 1 $a Witherell, Brianna Kimberly. $3 1476698
245 1 2 $a A Deep Neural Network for Real Time Mesoscale Ocean Eddy Detection.
264 0 $c 2023
300 $a 1 online resource (86 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: 84-11.
500 $a Advisor: Wang, Yadong.
502 $a Thesis (M.S.)--Southern Illinois University at Edwardsville, 2023.
504 $a Includes bibliographical references
520 $a Accurate and timely detection of mesoscale ocean eddies is paramount to ensure the safety and efficiency of maritime vessels. Existing methods of mesoscale ocean eddy detection exist but are limited in three ways: infrequent passes over a region of interest, insufficient scan swath size, and inadequate data resolution. To combat these issues, this work presents and validates an artificial intelligence based mesoscale ocean eddy detection algorithm that is trained on a novel data set of several thousand eddies gathered from imagery from the GOES-16 Advanced Baseline Imager (ABI). The ABI onboard GOES-16 shows promise for resolving issues from polar-orbiting satellites for quick and accurate mesoscale ocean eddy detection.There are three major steps in this work: data collection and preprocessing, training the object detection scheme, and validating the object detection scheme. The pre-processed data is input into an object detection scheme, which is well trained with 43,776 files from 152 days from March-July of 2022. This work proposes a Faster R-CNN object detection scheme with a MobileNet V3 feature extraction network. Several competing object detection scheme's performances were compared including: Single Shot Multibox Detector, Fully Convolutional One-stage Object Detection, and Focal Loss. It was found the proposed Faster R-CNN approach with a MobileNet V3 model backbone shows optimal performance in both accuracy and robustness, allowing for real-time mesoscale ocean eddy detection.
533 $a Electronic reproduction. $b Ann Arbor, Mich. : $c ProQuest, $d 2024
538 $a Mode of access: World Wide Web
650 4 $a Ocean engineering. $3 857658
650 4 $a Meteorology. $3 554820
650 4 $a Remote sensing. $3 557272
653 $a Artificial intelligence
653 $a Faster R-CNN
653 $a GOES
653 $a Mesoscale ocean eddy
653 $a Deep Neural Network
655 7 $a Electronic books. $2 local $3 554714
690 $a 0799
690 $a 0557
690 $a 0547
710 2 $a Southern Illinois University at Edwardsville. $b Electrical and Computer Engineering. $3 1476699
710 2 $a ProQuest Information and Learning Co. $3 1178819
773 0 $t Masters Abstracts International $g 84-11.
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=30422198 $z click for full text (PQDT)