Improvement of ablation experiment results for YOLOv7 modules.

<p>(a) The original YOLO model (b)YOLOv7+SimAM. (c)YOLOv7+SimAM+Pconv.</p>

محفوظ في:
التفاصيل البيبلوغرافية
المؤلف الرئيسي: Guoqing Zhang (151441) (author)
مؤلفون آخرون: Jiandong Liu (6064643) (author), Yongxiang Zhao (671364) (author), Wei Luo (80175) (author), Keyu Mei (20542959) (author), Penggang Wang (6038210) (author), Yubin Song (3205194) (author), Xiaoliang Li (720274) (author)
منشور في: 2025
الموضوعات:
الوسوم: إضافة وسم
لا توجد وسوم, كن أول من يضع وسما على هذه التسجيلة!
_version_ 1852023684318887936
author Guoqing Zhang (151441)
author2 Jiandong Liu (6064643)
Yongxiang Zhao (671364)
Wei Luo (80175)
Keyu Mei (20542959)
Penggang Wang (6038210)
Yubin Song (3205194)
Xiaoliang Li (720274)
author2_role author
author
author
author
author
author
author
author_facet Guoqing Zhang (151441)
Jiandong Liu (6064643)
Yongxiang Zhao (671364)
Wei Luo (80175)
Keyu Mei (20542959)
Penggang Wang (6038210)
Yubin Song (3205194)
Xiaoliang Li (720274)
author_role author
dc.creator.none.fl_str_mv Guoqing Zhang (151441)
Jiandong Liu (6064643)
Yongxiang Zhao (671364)
Wei Luo (80175)
Keyu Mei (20542959)
Penggang Wang (6038210)
Yubin Song (3205194)
Xiaoliang Li (720274)
dc.date.none.fl_str_mv 2025-01-10T20:14:09Z
dc.identifier.none.fl_str_mv 10.1371/journal.pone.0316933.g012
dc.relation.none.fl_str_mv https://figshare.com/articles/figure/Improvement_of_ablation_experiment_results_for_YOLOv7_modules_/28187220
dc.rights.none.fl_str_mv CC BY 4.0
info:eu-repo/semantics/openAccess
dc.subject.none.fl_str_mv Science Policy
Space Science
Biological Sciences not elsewhere classified
Information Systems not elsewhere classified
unmanned aerial vehicles
uavs ), utilizing
transfer learning techniques
suppressing irrelevant information
reducing missed detections
partially occluded targets
offering valuable insights
global economy expands
deep sort algorithms
deep sort algorithm
become increasingly crucial
ship monitoring scenarios
model &# 8217
limited ship data
deep sort model
emphasizing salient features
enhanced system achieves
demonstrated robust performance
improves detection accuracy
ship detection
model training
stable features
xlink ">
waterway transportation
unlinked tracks
significant challenges
partial convolution
logistics sector
irregularly shaped
iou metric
invalid regions
growth presents
feature extraction
false negatives
extensive evaluation
diou metric
detected objects
critical reference
computational demands
artificial intelligence
article introduces
also reduces
address issues
dc.title.none.fl_str_mv Improvement of ablation experiment results for YOLOv7 modules.
dc.type.none.fl_str_mv Image
Figure
info:eu-repo/semantics/publishedVersion
image
description <p>(a) The original YOLO model (b)YOLOv7+SimAM. (c)YOLOv7+SimAM+Pconv.</p>
eu_rights_str_mv openAccess
id Manara_f13d64d904473f7facd82f20b85cded0
identifier_str_mv 10.1371/journal.pone.0316933.g012
network_acronym_str Manara
network_name_str ManaraRepo
oai_identifier_str oai:figshare.com:article/28187220
publishDate 2025
repository.mail.fl_str_mv
repository.name.fl_str_mv
repository_id_str
rights_invalid_str_mv CC BY 4.0
spelling Improvement of ablation experiment results for YOLOv7 modules.Guoqing Zhang (151441)Jiandong Liu (6064643)Yongxiang Zhao (671364)Wei Luo (80175)Keyu Mei (20542959)Penggang Wang (6038210)Yubin Song (3205194)Xiaoliang Li (720274)Science PolicySpace ScienceBiological Sciences not elsewhere classifiedInformation Systems not elsewhere classifiedunmanned aerial vehiclesuavs ), utilizingtransfer learning techniquessuppressing irrelevant informationreducing missed detectionspartially occluded targetsoffering valuable insightsglobal economy expandsdeep sort algorithmsdeep sort algorithmbecome increasingly crucialship monitoring scenariosmodel &# 8217limited ship datadeep sort modelemphasizing salient featuresenhanced system achievesdemonstrated robust performanceimproves detection accuracyship detectionmodel trainingstable featuresxlink ">waterway transportationunlinked trackssignificant challengespartial convolutionlogistics sectorirregularly shapediou metricinvalid regionsgrowth presentsfeature extractionfalse negativesextensive evaluationdiou metricdetected objectscritical referencecomputational demandsartificial intelligencearticle introducesalso reducesaddress issues<p>(a) The original YOLO model (b)YOLOv7+SimAM. (c)YOLOv7+SimAM+Pconv.</p>2025-01-10T20:14:09ZImageFigureinfo:eu-repo/semantics/publishedVersionimage10.1371/journal.pone.0316933.g012https://figshare.com/articles/figure/Improvement_of_ablation_experiment_results_for_YOLOv7_modules_/28187220CC BY 4.0info:eu-repo/semantics/openAccessoai:figshare.com:article/281872202025-01-10T20:14:09Z
spellingShingle Improvement of ablation experiment results for YOLOv7 modules.
Guoqing Zhang (151441)
Science Policy
Space Science
Biological Sciences not elsewhere classified
Information Systems not elsewhere classified
unmanned aerial vehicles
uavs ), utilizing
transfer learning techniques
suppressing irrelevant information
reducing missed detections
partially occluded targets
offering valuable insights
global economy expands
deep sort algorithms
deep sort algorithm
become increasingly crucial
ship monitoring scenarios
model &# 8217
limited ship data
deep sort model
emphasizing salient features
enhanced system achieves
demonstrated robust performance
improves detection accuracy
ship detection
model training
stable features
xlink ">
waterway transportation
unlinked tracks
significant challenges
partial convolution
logistics sector
irregularly shaped
iou metric
invalid regions
growth presents
feature extraction
false negatives
extensive evaluation
diou metric
detected objects
critical reference
computational demands
artificial intelligence
article introduces
also reduces
address issues
status_str publishedVersion
title Improvement of ablation experiment results for YOLOv7 modules.
title_full Improvement of ablation experiment results for YOLOv7 modules.
title_fullStr Improvement of ablation experiment results for YOLOv7 modules.
title_full_unstemmed Improvement of ablation experiment results for YOLOv7 modules.
title_short Improvement of ablation experiment results for YOLOv7 modules.
title_sort Improvement of ablation experiment results for YOLOv7 modules.
topic Science Policy
Space Science
Biological Sciences not elsewhere classified
Information Systems not elsewhere classified
unmanned aerial vehicles
uavs ), utilizing
transfer learning techniques
suppressing irrelevant information
reducing missed detections
partially occluded targets
offering valuable insights
global economy expands
deep sort algorithms
deep sort algorithm
become increasingly crucial
ship monitoring scenarios
model &# 8217
limited ship data
deep sort model
emphasizing salient features
enhanced system achieves
demonstrated robust performance
improves detection accuracy
ship detection
model training
stable features
xlink ">
waterway transportation
unlinked tracks
significant challenges
partial convolution
logistics sector
irregularly shaped
iou metric
invalid regions
growth presents
feature extraction
false negatives
extensive evaluation
diou metric
detected objects
critical reference
computational demands
artificial intelligence
article introduces
also reduces
address issues