Architecture of LSTM-RBM.

<div><p>This research addresses the imperative need for efficient underwater exploration in the domain of deep-sea resource development, highlighting the importance of autonomous operations to mitigate the challenges posed by high-stress underwater environments. The proposed approach int...

وصف كامل

محفوظ في:
التفاصيل البيبلوغرافية
المؤلف الرئيسي: Manimurugan S. (20314652) (author)
مؤلفون آخرون: Karthikeyan P. (20314655) (author), Narmatha C. (20314658) (author), Majed M. Aborokbah (20314661) (author), Anand Paul (19646164) (author), Subramaniam Ganesan (18173638) (author), Rajendran T. (20314664) (author), Mohammad Ammad-Uddin (20314667) (author)
منشور في: 2024
الموضوعات:
الوسوم: إضافة وسم
لا توجد وسوم, كن أول من يضع وسما على هذه التسجيلة!
_version_ 1852024974914617344
author Manimurugan S. (20314652)
author2 Karthikeyan P. (20314655)
Narmatha C. (20314658)
Majed M. Aborokbah (20314661)
Anand Paul (19646164)
Subramaniam Ganesan (18173638)
Rajendran T. (20314664)
Mohammad Ammad-Uddin (20314667)
author2_role author
author
author
author
author
author
author
author_facet Manimurugan S. (20314652)
Karthikeyan P. (20314655)
Narmatha C. (20314658)
Majed M. Aborokbah (20314661)
Anand Paul (19646164)
Subramaniam Ganesan (18173638)
Rajendran T. (20314664)
Mohammad Ammad-Uddin (20314667)
author_role author
dc.creator.none.fl_str_mv Manimurugan S. (20314652)
Karthikeyan P. (20314655)
Narmatha C. (20314658)
Majed M. Aborokbah (20314661)
Anand Paul (19646164)
Subramaniam Ganesan (18173638)
Rajendran T. (20314664)
Mohammad Ammad-Uddin (20314667)
dc.date.none.fl_str_mv 2024-11-22T18:25:17Z
dc.identifier.none.fl_str_mv 10.1371/journal.pone.0313708.g002
dc.relation.none.fl_str_mv https://figshare.com/articles/figure/Architecture_of_LSTM-RBM_/27891563
dc.rights.none.fl_str_mv CC BY 4.0
info:eu-repo/semantics/openAccess
dc.subject.none.fl_str_mv Microbiology
Neuroscience
Evolutionary Biology
Inorganic Chemistry
Science Policy
Biological Sciences not elsewhere classified
Information Systems not elsewhere classified
vanishing gradient problem
underwater object detection
stress underwater environments
star fish object
sea resource development
restricted boltzmann machine
efficient underwater exploration
effective feature learning
big fish object
architecture handles variable
achieves enhanced significance
proposed approach introduces
processing sequences bidirectionally
directional long short
capturing complex patterns
big fish 98
capturing long
rbm approach
length sequences
xlink ">
urpc dataset
term memory
term dependencies
robust solution
research addresses
rbm ).
model benefits
imperative need
hybrid model
future contexts
findings underscore
enhance comprehension
comprehensive evaluations
challenges posed
autonomous operations
abstract representations
dc.title.none.fl_str_mv Architecture of LSTM-RBM.
dc.type.none.fl_str_mv Image
Figure
info:eu-repo/semantics/publishedVersion
image
description <div><p>This research addresses the imperative need for efficient underwater exploration in the domain of deep-sea resource development, highlighting the importance of autonomous operations to mitigate the challenges posed by high-stress underwater environments. The proposed approach introduces a hybrid model for Underwater Object Detection (UOD), combining Bi-directional Long Short-Term Memory (Bi-LSTM) with a Restricted Boltzmann Machine (RBM). Bi-LSTM excels at capturing long-term dependencies and processing sequences bidirectionally to enhance comprehension of both past and future contexts. The model benefits from effective feature learning, aided by RBMs that enable the extraction of hierarchical and abstract representations. Additionally, this architecture handles variable-length sequences, mitigates the vanishing gradient problem, and achieves enhanced significance by capturing complex patterns in the data. Comprehensive evaluations on brackish, and URPC 2020 datasets demonstrate superior performance, with the BiLSTM-RBM model showcasing notable accuracies, such as big fish 98.5 for the big fish object in the brackish dataset and 98 for the star fish object in the URPC dataset. Overall, these findings underscore the BiLSTM-RBM model’s suitability for UOD, positioning it as a robust solution for effective underwater object detection in challenging underwater environments.</p></div>
eu_rights_str_mv openAccess
id Manara_e96f96cddc9c1982e18ae32b4eab20bb
identifier_str_mv 10.1371/journal.pone.0313708.g002
network_acronym_str Manara
network_name_str ManaraRepo
oai_identifier_str oai:figshare.com:article/27891563
publishDate 2024
repository.mail.fl_str_mv
repository.name.fl_str_mv
repository_id_str
rights_invalid_str_mv CC BY 4.0
spelling Architecture of LSTM-RBM.Manimurugan S. (20314652)Karthikeyan P. (20314655)Narmatha C. (20314658)Majed M. Aborokbah (20314661)Anand Paul (19646164)Subramaniam Ganesan (18173638)Rajendran T. (20314664)Mohammad Ammad-Uddin (20314667)MicrobiologyNeuroscienceEvolutionary BiologyInorganic ChemistryScience PolicyBiological Sciences not elsewhere classifiedInformation Systems not elsewhere classifiedvanishing gradient problemunderwater object detectionstress underwater environmentsstar fish objectsea resource developmentrestricted boltzmann machineefficient underwater explorationeffective feature learningbig fish objectarchitecture handles variableachieves enhanced significanceproposed approach introducesprocessing sequences bidirectionallydirectional long shortcapturing complex patternsbig fish 98capturing longrbm approachlength sequencesxlink ">urpc datasetterm memoryterm dependenciesrobust solutionresearch addressesrbm ).model benefitsimperative needhybrid modelfuture contextsfindings underscoreenhance comprehensioncomprehensive evaluationschallenges posedautonomous operationsabstract representations<div><p>This research addresses the imperative need for efficient underwater exploration in the domain of deep-sea resource development, highlighting the importance of autonomous operations to mitigate the challenges posed by high-stress underwater environments. The proposed approach introduces a hybrid model for Underwater Object Detection (UOD), combining Bi-directional Long Short-Term Memory (Bi-LSTM) with a Restricted Boltzmann Machine (RBM). Bi-LSTM excels at capturing long-term dependencies and processing sequences bidirectionally to enhance comprehension of both past and future contexts. The model benefits from effective feature learning, aided by RBMs that enable the extraction of hierarchical and abstract representations. Additionally, this architecture handles variable-length sequences, mitigates the vanishing gradient problem, and achieves enhanced significance by capturing complex patterns in the data. Comprehensive evaluations on brackish, and URPC 2020 datasets demonstrate superior performance, with the BiLSTM-RBM model showcasing notable accuracies, such as big fish 98.5 for the big fish object in the brackish dataset and 98 for the star fish object in the URPC dataset. Overall, these findings underscore the BiLSTM-RBM model’s suitability for UOD, positioning it as a robust solution for effective underwater object detection in challenging underwater environments.</p></div>2024-11-22T18:25:17ZImageFigureinfo:eu-repo/semantics/publishedVersionimage10.1371/journal.pone.0313708.g002https://figshare.com/articles/figure/Architecture_of_LSTM-RBM_/27891563CC BY 4.0info:eu-repo/semantics/openAccessoai:figshare.com:article/278915632024-11-22T18:25:17Z
spellingShingle Architecture of LSTM-RBM.
Manimurugan S. (20314652)
Microbiology
Neuroscience
Evolutionary Biology
Inorganic Chemistry
Science Policy
Biological Sciences not elsewhere classified
Information Systems not elsewhere classified
vanishing gradient problem
underwater object detection
stress underwater environments
star fish object
sea resource development
restricted boltzmann machine
efficient underwater exploration
effective feature learning
big fish object
architecture handles variable
achieves enhanced significance
proposed approach introduces
processing sequences bidirectionally
directional long short
capturing complex patterns
big fish 98
capturing long
rbm approach
length sequences
xlink ">
urpc dataset
term memory
term dependencies
robust solution
research addresses
rbm ).
model benefits
imperative need
hybrid model
future contexts
findings underscore
enhance comprehension
comprehensive evaluations
challenges posed
autonomous operations
abstract representations
status_str publishedVersion
title Architecture of LSTM-RBM.
title_full Architecture of LSTM-RBM.
title_fullStr Architecture of LSTM-RBM.
title_full_unstemmed Architecture of LSTM-RBM.
title_short Architecture of LSTM-RBM.
title_sort Architecture of LSTM-RBM.
topic Microbiology
Neuroscience
Evolutionary Biology
Inorganic Chemistry
Science Policy
Biological Sciences not elsewhere classified
Information Systems not elsewhere classified
vanishing gradient problem
underwater object detection
stress underwater environments
star fish object
sea resource development
restricted boltzmann machine
efficient underwater exploration
effective feature learning
big fish object
architecture handles variable
achieves enhanced significance
proposed approach introduces
processing sequences bidirectionally
directional long short
capturing complex patterns
big fish 98
capturing long
rbm approach
length sequences
xlink ">
urpc dataset
term memory
term dependencies
robust solution
research addresses
rbm ).
model benefits
imperative need
hybrid model
future contexts
findings underscore
enhance comprehension
comprehensive evaluations
challenges posed
autonomous operations
abstract representations