RamanCMP: A Raman spectral classification acceleration method based on lightweight model and model compression techniques

In recent years, Raman spectroscopy combined with deep learning techniques has been widely used in various fields such as medical, chemical, and geological. However, there is still room for optimization of deep learning techniques and model compression algorithms for processing Raman spectral data....

Full description

Saved in:

Bibliographic Details
Published in:	Analytica chimica acta Vol. 1278; p. 341758
Main Authors:	Gong, Zengyun, Chen, Chen, Chen, Cheng, Li, Chenxi, Tian, Xuecong, Gong, Zhongcheng, Lv, Xiaoyi
Format:	Journal Article
Language:	English
Published:	Elsevier B.V 16.10.2023
Subjects:	Lightweight model Model compression Raman spectroscopy RamanCMP Lightweight model Raman spectroscopy RamanCMP Model compression
ISSN:	0003-2670, 1873-4324, 1873-4324
Online Access:	Get full text
Tags:	Add Tag No Tags, Be the first to tag this record!

Abstract	In recent years, Raman spectroscopy combined with deep learning techniques has been widely used in various fields such as medical, chemical, and geological. However, there is still room for optimization of deep learning techniques and model compression algorithms for processing Raman spectral data. To further optimize deep learning models applied to Raman spectroscopy, in this study time, accuracy, sensitivity, specificity and floating point operations numbers(FLOPs) are used as evaluation metrics to optimize the model, which is named RamanCompact(RamanCMP). The experimental data used in this research are selected from the RRUFF public dataset, which consists of 723 Raman spectroscopy data samples from 10 different mineral categories. In this paper, 1D-EfficientNet adapted to the spectral data as well as 1D-DRSN are proposed to improve the model classification accuracy. To achieve better classification accuracy while optimizing the time parameters, three model compression methods are designed: knowledge distillation using 1D-EfficientNet model as a teacher model to train convolutional neural networks(CNN), proposing a channel conversion method to optimize 1D-DRSN model, and using 1D-DRSN model as a feature extractor in combination with linear discriminant analysis(LDA) model for classification. Compared with the traditional LDA and CNN models, the accuracy of 1D-EfficientNet and 1D-DRSN is improved by more than 20%. The time of the distilled model is reduced by 9680.9s compared with the teacher model 1D-EfficientNet under the condition of losing 2.07% accuracy. The accuracy of the distilled model is improved by 20% compared to the CNN student model while keeping inference efficiency constant. The 1D-DRSN optimized with channel conversion method saves 60% inference time of the original 1D-DRSN model. Feature extraction reduces the inference time of 1D-DRSN model by 93% with 94.48% accuracy. This study innovatively combines lightweight models and model compression algorithms to improve the classification speed of deep learning models in the field of Raman spectroscopy, forming a complete set of analysis methods and laying the foundation for future research. [Display omitted] •Proposed RamanCMP, a Raman spectral classification acceleration method.•Proposed 1D-EfficientNet and 1D-DRSN for spectral data.•Proposed channel conversion method and 1D-DRSN feature extractor.•Testing RamanCMP using the RRUFF public dataset and the external validation dataset for Sjögren's syndrome.
AbstractList	In recent years, Raman spectroscopy combined with deep learning techniques has been widely used in various fields such as medical, chemical, and geological. However, there is still room for optimization of deep learning techniques and model compression algorithms for processing Raman spectral data. To further optimize deep learning models applied to Raman spectroscopy, in this study time, accuracy, sensitivity, specificity and floating point operations numbers(FLOPs) are used as evaluation metrics to optimize the model, which is named RamanCompact(RamanCMP). The experimental data used in this research are selected from the RRUFF public dataset, which consists of 723 Raman spectroscopy data samples from 10 different mineral categories. In this paper, 1D-EfficientNet adapted to the spectral data as well as 1D-DRSN are proposed to improve the model classification accuracy. To achieve better classification accuracy while optimizing the time parameters, three model compression methods are designed: knowledge distillation using 1D-EfficientNet model as a teacher model to train convolutional neural networks(CNN), proposing a channel conversion method to optimize 1D-DRSN model, and using 1D-DRSN model as a feature extractor in combination with linear discriminant analysis(LDA) model for classification. Compared with the traditional LDA and CNN models, the accuracy of 1D-EfficientNet and 1D-DRSN is improved by more than 20%. The time of the distilled model is reduced by 9680.9s compared with the teacher model 1D-EfficientNet under the condition of losing 2.07% accuracy. The accuracy of the distilled model is improved by 20% compared to the CNN student model while keeping inference efficiency constant. The 1D-DRSN optimized with channel conversion method saves 60% inference time of the original 1D-DRSN model. Feature extraction reduces the inference time of 1D-DRSN model by 93% with 94.48% accuracy. This study innovatively combines lightweight models and model compression algorithms to improve the classification speed of deep learning models in the field of Raman spectroscopy, forming a complete set of analysis methods and laying the foundation for future research. [Display omitted] •Proposed RamanCMP, a Raman spectral classification acceleration method.•Proposed 1D-EfficientNet and 1D-DRSN for spectral data.•Proposed channel conversion method and 1D-DRSN feature extractor.•Testing RamanCMP using the RRUFF public dataset and the external validation dataset for Sjögren's syndrome. In recent years, Raman spectroscopy combined with deep learning techniques has been widely used in various fields such as medical, chemical, and geological. However, there is still room for optimization of deep learning techniques and model compression algorithms for processing Raman spectral data. To further optimize deep learning models applied to Raman spectroscopy, in this study time, accuracy, sensitivity, specificity and floating point operations numbers(FLOPs) are used as evaluation metrics to optimize the model, which is named RamanCompact(RamanCMP). The experimental data used in this research are selected from the RRUFF public dataset, which consists of 723 Raman spectroscopy data samples from 10 different mineral categories. In this paper, 1D-EfficientNet adapted to the spectral data as well as 1D-DRSN are proposed to improve the model classification accuracy. To achieve better classification accuracy while optimizing the time parameters, three model compression methods are designed: knowledge distillation using 1D-EfficientNet model as a teacher model to train convolutional neural networks(CNN), proposing a channel conversion method to optimize 1D-DRSN model, and using 1D-DRSN model as a feature extractor in combination with linear discriminant analysis(LDA) model for classification. Compared with the traditional LDA and CNN models, the accuracy of 1D-EfficientNet and 1D-DRSN is improved by more than 20%. The time of the distilled model is reduced by 9680.9s compared with the teacher model 1D-EfficientNet under the condition of losing 2.07% accuracy. The accuracy of the distilled model is improved by 20% compared to the CNN student model while keeping inference efficiency constant. The 1D-DRSN optimized with channel conversion method saves 60% inference time of the original 1D-DRSN model. Feature extraction reduces the inference time of 1D-DRSN model by 93% with 94.48% accuracy. This study innovatively combines lightweight models and model compression algorithms to improve the classification speed of deep learning models in the field of Raman spectroscopy, forming a complete set of analysis methods and laying the foundation for future research.In recent years, Raman spectroscopy combined with deep learning techniques has been widely used in various fields such as medical, chemical, and geological. However, there is still room for optimization of deep learning techniques and model compression algorithms for processing Raman spectral data. To further optimize deep learning models applied to Raman spectroscopy, in this study time, accuracy, sensitivity, specificity and floating point operations numbers(FLOPs) are used as evaluation metrics to optimize the model, which is named RamanCompact(RamanCMP). The experimental data used in this research are selected from the RRUFF public dataset, which consists of 723 Raman spectroscopy data samples from 10 different mineral categories. In this paper, 1D-EfficientNet adapted to the spectral data as well as 1D-DRSN are proposed to improve the model classification accuracy. To achieve better classification accuracy while optimizing the time parameters, three model compression methods are designed: knowledge distillation using 1D-EfficientNet model as a teacher model to train convolutional neural networks(CNN), proposing a channel conversion method to optimize 1D-DRSN model, and using 1D-DRSN model as a feature extractor in combination with linear discriminant analysis(LDA) model for classification. Compared with the traditional LDA and CNN models, the accuracy of 1D-EfficientNet and 1D-DRSN is improved by more than 20%. The time of the distilled model is reduced by 9680.9s compared with the teacher model 1D-EfficientNet under the condition of losing 2.07% accuracy. The accuracy of the distilled model is improved by 20% compared to the CNN student model while keeping inference efficiency constant. The 1D-DRSN optimized with channel conversion method saves 60% inference time of the original 1D-DRSN model. Feature extraction reduces the inference time of 1D-DRSN model by 93% with 94.48% accuracy. This study innovatively combines lightweight models and model compression algorithms to improve the classification speed of deep learning models in the field of Raman spectroscopy, forming a complete set of analysis methods and laying the foundation for future research.
ArticleNumber	341758
Author	Chen, Chen Tian, Xuecong Gong, Zhongcheng Chen, Cheng Li, Chenxi Gong, Zengyun Lv, Xiaoyi
Author_xml	– sequence: 1 givenname: Zengyun surname: Gong fullname: Gong, Zengyun email: g18999390232@sina.com organization: College of Software, Xinjiang University, Urumqi, 830046, Xinjiang, China – sequence: 2 givenname: Chen surname: Chen fullname: Chen, Chen email: 1343432873@qq.com organization: College of Information Science and Engineering, Xinjiang University, Urumqi, 830046, Xinjian, China – sequence: 3 givenname: Cheng surname: Chen fullname: Chen, Cheng email: chenchengoptics@gmail.com organization: College of Software, Xinjiang University, Urumqi, 830046, Xinjiang, China – sequence: 4 givenname: Chenxi surname: Li fullname: Li, Chenxi email: lichenxiuke@gmail.com organization: Oncological Department of Oral and Maxillofacial Surgery, The First Affiliated Hospital of Xinjiang Medical University, Urumqi, 830054, Xinjiang, China – sequence: 5 givenname: Xuecong surname: Tian fullname: Tian, Xuecong email: xuecongtian@gmail.com organization: College of Information Science and Engineering, Xinjiang University, Urumqi, 830046, Xinjian, China – sequence: 6 givenname: Zhongcheng surname: Gong fullname: Gong, Zhongcheng email: 565249755@qq.com organization: Oncological Department of Oral and Maxillofacial Surgery, The First Affiliated Hospital of Xinjiang Medical University, Urumqi, 830054, Xinjiang, China – sequence: 7 givenname: Xiaoyi orcidid: 0000-0002-8023-4119 surname: Lv fullname: Lv, Xiaoyi email: xjuwawj01@163.com organization: College of Software, Xinjiang University, Urumqi, 830046, Xinjiang, China
BookMark	eNp9kctu2zAQRYnAAeI8PiA7LruRy4ciSu3KMJq2QIoEQbImRsNxTUMSXVJJ4b8PHWWVhTfDucC9Q87hOZsNYSDGrqVYSCGrr9sFICyUUHqhS2lu6hM2l7XRRalVOWNzIYQuVGXEGTtPaZulkqKcs_0j9DCs_jx840v-3vO0IxwjdBw7SMmvPcLow8ABkTqKk-hp3ATHW0jkeNad_7sZ_9Oh8j446jgM7qPD0O8i5VHZNxJuBv_vhdIlO11Dl-jq47xgz7c_nla_irv7n79Xy7sCtRZj4WQjWgV107YGGpBCV2WLlUKd1zaNAyFzIVXn1xksTWW0g3XdGE0IWLb6gn2Z5u5iONw72t6nvEkHA4WXZFVd3ZjaSG2yVU5WjCGlSGu7i76HuLdS2ANmu7UZsz1gthPmnDGfMujHd0aZoe-OJr9PScrbv3qKNqGnAcn5mH_AuuCPpN8AveWbUw
CitedBy_id	crossref_primary_10_1016_j_heliyon_2024_e32087 crossref_primary_10_1038_s41598_024_64621_4 crossref_primary_10_1016_j_artmed_2024_103053
Cites_doi	10.1016/j.fuel.2019.02.096 10.1002/jrs.4340 10.1002/jrs.5214 10.1016/j.jallcom.2015.09.273 10.1016/j.ymssp.2020.107398 10.1002/jrs.1574 10.1038/s42254-020-0171-y 10.1016/j.saa.2003.12.021 10.1016/j.cosrev.2021.100378 10.3390/bios9020057 10.1038/s41598-021-84565-3 10.3390/ijms20184414 10.1016/j.cej.2020.128208 10.3389/frobt.2015.00036 10.1208/s12249-022-02335-4 10.3390/bios11060187 10.1016/j.eswa.2017.05.039 10.1007/s10618-019-00619-1 10.1109/TII.2019.2943898 10.1016/j.earscirev.2019.102936 10.1002/adsu.201900062 10.1111/maps.13106 10.1145/3439726 10.1007/s00269-016-0824-7 10.1016/j.patrec.2020.07.042 10.1016/j.molstruc.2009.01.001 10.1016/j.saa.2018.07.079 10.1186/s11671-019-3039-2
ContentType	Journal Article
Copyright	2023 Elsevier B.V. Copyright © 2023 Elsevier B.V. All rights reserved.
Copyright_xml	– notice: 2023 Elsevier B.V. – notice: Copyright © 2023 Elsevier B.V. All rights reserved.
DBID	AAYXX CITATION 7X8
DOI	10.1016/j.aca.2023.341758
DatabaseName	CrossRef MEDLINE - Academic
DatabaseTitle	CrossRef MEDLINE - Academic
DatabaseTitleList	MEDLINE - Academic
Database_xml	– sequence: 1 dbid: 7X8 name: MEDLINE - Academic url: https://search.proquest.com/medline sourceTypes: Aggregation Database
DeliveryMethod	fulltext_linktorsrc
Discipline	Chemistry
EISSN	1873-4324
ExternalDocumentID	10_1016_j_aca_2023_341758 S0003267023009790
GroupedDBID	--- --K --M -~X .~1 0R~ 1B1 1RT 1~. 1~5 23M 4.4 457 4G. 5GY 5VS 6J9 7-5 71M 8P~ 9JM 9JN AABNK AACTN AAEDT AAEDW AAIAV AAIKJ AAKOC AALRI AAOAW AAQFI AARLI AAXUO ABFNM ABFRF ABFYP ABGSF ABJNI ABLST ABMAC ABUDA ABYKQ ACBEA ACCUC ACDAQ ACGFO ACGFS ACIWK ACNCT ACPRK ACRLP ADBBV ADECG ADEZE ADUVX AEBSH AEFWE AEHWI AEKER AENEX AFKWA AFRAH AFTJW AFXIZ AFZHZ AGHFR AGUBO AGYEJ AHEUO AHHHB AIEXJ AIKHN AITUG AJOXV AJSZI AKIFW ALMA_UNASSIGNED_HOLDINGS AMFUW AMRAJ AXJTR BKOJK BLECG BLXMC CS3 DOVZS EBS EFJIC EFLBG EO8 EO9 EP2 EP3 F5P FDB FIRID FLBIZ FNPLU FYGXN G-Q GBLVA IHE J1W K-O KCYFY KOM M36 M41 MO0 N9A O-L O9- OAUVE OZT P-8 P-9 P2P PC. Q38 RNS ROL RPZ SCC SCH SDF SDG SDP SES SEW SPC SPCBC SSJ SSK SSU SSZ T5K TN5 TWZ UPT WH7 YK3 ZMT ~02 ~G- .GJ 3O- 53G 9DU AAHBH AAQXK AATTM AAXKI AAYJJ AAYWO AAYXX ABDPE ABEFU ABWVN ABXDB ACKIV ACLOT ACNNM ACRPL ACVFH ADCNI ADMUD ADNMO AEIPS AEUPX AFJKZ AFPUW AGQPQ AGRDE AI. AIGII AIIUN AJQLL AKBMS AKRWK AKYEP ANKPU APXCP ASPBG AVWKF AZFZN CITATION EFKBS EJD FA8 FEDTE FGOYB HMU HVGLF HZ~ H~9 MVM NHB R2- SCB T9H UQL VH1 WUQ XOL XPP ZCG ZXP ZY4 ~HD 7X8
ID	FETCH-LOGICAL-c330t-d190b2a89bb7a9a10364bc62c301679da019dae28cce7c47673daf8973ecac4b3
ISICitedReferencesCount	6
ISICitedReferencesURI	http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=001070863600001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN	0003-2670 1873-4324
IngestDate	Sun Sep 28 10:22:48 EDT 2025 Sat Nov 29 07:01:01 EST 2025 Tue Nov 18 21:40:02 EST 2025 Fri Feb 23 02:35:48 EST 2024
IsPeerReviewed	true
IsScholarly	true
Keywords	Lightweight model Raman spectroscopy RamanCMP Model compression
Language	English
LinkModel	OpenURL
MergedId	FETCHMERGED-LOGICAL-c330t-d190b2a89bb7a9a10364bc62c301679da019dae28cce7c47673daf8973ecac4b3
Notes	ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 23
ORCID	0000-0002-8023-4119
PQID	2865787137
PQPubID	23479
ParticipantIDs	proquest_miscellaneous_2865787137 crossref_primary_10_1016_j_aca_2023_341758 crossref_citationtrail_10_1016_j_aca_2023_341758 elsevier_sciencedirect_doi_10_1016_j_aca_2023_341758
PublicationCentury	2000
PublicationDate	2023-10-16
PublicationDateYYYYMMDD	2023-10-16
PublicationDate_xml	– month: 10 year: 2023 text: 2023-10-16 day: 16
PublicationDecade	2020
PublicationTitle	Analytica chimica acta
PublicationYear	2023
Publisher	Elsevier B.V
Publisher_xml	– name: Elsevier B.V
References	Xu (bib11) 2020; 49 Carlomagno (bib4) 2021; 11 Menghani (bib25) 2021 Hua (bib19) 2018 Wang (bib15) 2021; 141 Armbruster (bib26) 2015 Petersen (bib5) 2021; 11 Lauer (bib6) 2019; 3 Henry (bib8) 2019; 198 Karwowski (bib28) 2013; 44 Hinton (bib24) 2015; vol. 2 Jehlička (bib31) 2017; 48 Srinivas (bib40) 2016; 2 Mu (bib16) 2019; 13 Sheng (bib9) 2022; 23 Zhang (bib33) 2018; 53 Zhao (bib39) 2019; 16 Zhu (bib23) 2017 Juroszek (bib30) 2018; 205 Chollet (bib20) 2017 Frost (bib29) 2007; 38 Lee (bib42) 2020; 67 Schulte (bib35) 1995 Roelofs (bib37) 2019; 32 Jones (bib1) 2019; 14 Minaee (bib12) 2021; 54 Tan (bib38) 2019 Anowar (bib43) 2021; 40 Makreski (bib36) 2009; 924 Duch (bib44) 2006 Varshni (bib46) 2019 Cho (bib47) 2019 Hsiao (bib21) 2021 Li (bib32) 2016; 657 Chen (bib10) 2019; 246 Pilot (bib3) 2019; 9 Qin (bib34) 2016; 43 Cai (bib7) 2021; 410 Wang (bib2) 2020; 2 Ismail Fawaz (bib13) 2019; 33 Wang (bib17) 2019 Guo (bib18) 2019 Sompairac (bib45) 2019; 20 Zhang (bib22) 2018 Kiranyaz (bib41) 2021; 151 Affonso (bib14) 2017; 85 Gopal (bib27) 2004; 60 Hua (10.1016/j.aca.2023.341758_bib19) 2018 Lee (10.1016/j.aca.2023.341758_bib42) 2020; 67 Anowar (10.1016/j.aca.2023.341758_bib43) 2021; 40 Minaee (10.1016/j.aca.2023.341758_bib12) 2021; 54 Guo (10.1016/j.aca.2023.341758_bib18) 2019 Hsiao (10.1016/j.aca.2023.341758_bib21) 2021 Zhao (10.1016/j.aca.2023.341758_bib39) 2019; 16 Jones (10.1016/j.aca.2023.341758_bib1) 2019; 14 Schulte (10.1016/j.aca.2023.341758_bib35) 1995 Sompairac (10.1016/j.aca.2023.341758_bib45) 2019; 20 Carlomagno (10.1016/j.aca.2023.341758_bib4) 2021; 11 Gopal (10.1016/j.aca.2023.341758_bib27) 2004; 60 Tan (10.1016/j.aca.2023.341758_bib38) 2019 Wang (10.1016/j.aca.2023.341758_bib15) 2021; 141 Mu (10.1016/j.aca.2023.341758_bib16) 2019; 13 Roelofs (10.1016/j.aca.2023.341758_bib37) 2019; 32 Juroszek (10.1016/j.aca.2023.341758_bib30) 2018; 205 Cho (10.1016/j.aca.2023.341758_bib47) 2019 Wang (10.1016/j.aca.2023.341758_bib2) 2020; 2 Ismail Fawaz (10.1016/j.aca.2023.341758_bib13) 2019; 33 Makreski (10.1016/j.aca.2023.341758_bib36) 2009; 924 Petersen (10.1016/j.aca.2023.341758_bib5) 2021; 11 Jehlička (10.1016/j.aca.2023.341758_bib31) 2017; 48 Chollet (10.1016/j.aca.2023.341758_bib20) 2017 Pilot (10.1016/j.aca.2023.341758_bib3) 2019; 9 Henry (10.1016/j.aca.2023.341758_bib8) 2019; 198 Zhang (10.1016/j.aca.2023.341758_bib22) 2018 Cai (10.1016/j.aca.2023.341758_bib7) 2021; 410 Menghani (10.1016/j.aca.2023.341758_bib25) 2021 Sheng (10.1016/j.aca.2023.341758_bib9) 2022; 23 Karwowski (10.1016/j.aca.2023.341758_bib28) 2013; 44 Frost (10.1016/j.aca.2023.341758_bib29) 2007; 38 Qin (10.1016/j.aca.2023.341758_bib34) 2016; 43 Affonso (10.1016/j.aca.2023.341758_bib14) 2017; 85 Wang (10.1016/j.aca.2023.341758_bib17) 2019 Xu (10.1016/j.aca.2023.341758_bib11) 2020; 49 Zhu (10.1016/j.aca.2023.341758_bib23) 2017 Duch (10.1016/j.aca.2023.341758_bib44) 2006 Kiranyaz (10.1016/j.aca.2023.341758_bib41) 2021; 151 Lauer (10.1016/j.aca.2023.341758_bib6) 2019; 3 Armbruster (10.1016/j.aca.2023.341758_bib26) 2015 Zhang (10.1016/j.aca.2023.341758_bib33) 2018; 53 Hinton (10.1016/j.aca.2023.341758_bib24) 2015; vol. 2 Varshni (10.1016/j.aca.2023.341758_bib46) 2019 Li (10.1016/j.aca.2023.341758_bib32) 2016; 657 Srinivas (10.1016/j.aca.2023.341758_bib40) 2016; 2 Chen (10.1016/j.aca.2023.341758_bib10) 2019; 246
References_xml	– year: 2019 ident: bib17 article-title: Fully learnable group convolution for acceleration of deep neural networks publication-title: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition – year: 2021 ident: bib25 article-title: Efficient Deep Learning: A Survey on Making Deep Learning Models Smaller, Faster, and Better – year: 2018 ident: bib19 article-title: Pointwise convolutional neural networks publication-title: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition – year: 2019 ident: bib38 article-title: Efficientnet: rethinking model scaling for convolutional neural networks publication-title: International Conference on Machine Learning – volume: 9 start-page: 57 year: 2019 ident: bib3 article-title: A review on surface-enhanced Raman scattering publication-title: Biosensors – volume: 3 year: 2019 ident: bib6 article-title: Durable cellulose–sulfur composites derived from agricultural and petrochemical waste publication-title: Adv. Sustain. Syst. – volume: 410 year: 2021 ident: bib7 article-title: Analysis of environmental nanoplastics: progress and challenges publication-title: Chem. Eng. J. – volume: 246 start-page: 60 year: 2019 end-page: 68 ident: bib10 article-title: The quantitative assessment of coke morphology based on the Raman spectroscopic characterization of serial petroleum cokes publication-title: Fuel – volume: 151 year: 2021 ident: bib41 article-title: 1D convolutional neural networks and applications: a survey publication-title: Mech. Syst. Signal Process. – volume: 205 start-page: 582 year: 2018 end-page: 592 ident: bib30 article-title: Raman spectroscopy and structural study of baryte-hashemite solid solution from pyrometamorphic rocks of the Hatrurim Complex, Israel publication-title: Spectrochim. Acta Mol. Biomol. Spectrosc. – volume: 16 start-page: 4681 year: 2019 end-page: 4690 ident: bib39 article-title: Deep residual shrinkage networks for fault diagnosis publication-title: IEEE Trans. Ind. Inf. – volume: 11 start-page: 187 year: 2021 ident: bib5 article-title: Application of Raman spectroscopic methods in food safety: a review publication-title: Biosensors – year: 2018 ident: bib22 article-title: Shufflenet: an extremely efficient convolutional neural network for mobile devices publication-title: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition – year: 2019 ident: bib46 article-title: Pneumonia detection using CNN based feature extraction publication-title: 2019 IEEE International Conference on Electrical, Computer and Communication Technologies (ICECCT) – volume: 14 start-page: 1 year: 2019 end-page: 34 ident: bib1 article-title: Raman techniques: fundamentals and frontiers publication-title: Nanoscale Res. Lett. – year: 2019 ident: bib18 article-title: Depthwise convolution is all you need for learning multiple visual domains publication-title: Proceedings of the AAAI Conference on Artificial Intelligence – year: 1995 ident: bib35 article-title: Raman spectroscopy of fluorophosphate and fluorovanadate laser crystals publication-title: UV and Visible Lasers and Laser Crystal Growth – volume: vol. 2 year: 2015 ident: bib24 publication-title: Distilling the Knowledge in a Neural Network – volume: 657 start-page: 152 year: 2016 end-page: 156 ident: bib32 article-title: The effect of strong magnetic field on the microstructure of pure diopside and diopside doped with Fe3+ or Mn2+ publication-title: J. Alloys Compd. – volume: 67 start-page: 1344 year: 2020 end-page: 1353 ident: bib42 article-title: Channel attention module with multiscale grid average pooling for breast cancer segmentation in an ultrasound image publication-title: IEEE Trans. Ultrason. Ferroelectrics Freq. Control – start-page: 89 year: 2006 end-page: 117 ident: bib44 article-title: Feature Extraction: Foundations and Applications – volume: 49 year: 2020 ident: bib11 article-title: Raman mineral recognition method based on all-optical diffraction deep neural network publication-title: 红外与激光工程 – volume: 40 year: 2021 ident: bib43 article-title: Conceptual and empirical comparison of dimensionality reduction algorithms (pca, kpca, lda, mds, svd, lle, isomap, le, ica, t-sne) publication-title: Comput. Sci. Rev. – year: 2019 ident: bib47 article-title: On the efficacy of knowledge distillation publication-title: Proceedings of the IEEE/CVF International Conference on Computer Vision – volume: 13 start-page: 1738 year: 2019 end-page: 1764 ident: bib16 article-title: A review of deep learning research publication-title: KSII Transactions on Internet and Information Systems (TIIS) – volume: 43 start-page: 649 year: 2016 end-page: 659 ident: bib34 article-title: High-pressure behavior of natural single-crystal epidote and clinozoisite up to 40 GPa publication-title: Phys. Chem. Miner. – volume: 85 start-page: 114 year: 2017 end-page: 122 ident: bib14 article-title: Deep learning for biological image classification publication-title: Expert Syst. Appl. – year: 2021 ident: bib21 article-title: Efficient computation of depthwise separable convolution in MoblieNet deep neural network models publication-title: 2021 IEEE International Conference on Consumer Electronics-Taiwan (ICCE-TW) – volume: 33 start-page: 917 year: 2019 end-page: 963 ident: bib13 article-title: Deep learning for time series classification: a review publication-title: Data Min. Knowl. Discov. – volume: 32 year: 2019 ident: bib37 article-title: A meta-analysis of overfitting in machine learning publication-title: Adv. Neural Inf. Process. Syst. – year: 2017 ident: bib23 article-title: To Prune, or Not to Prune: Exploring the Efficacy of Pruning for Model Compression – volume: 924 start-page: 413 year: 2009 end-page: 419 ident: bib36 article-title: Minerals from Macedonia. XXIV. Spectra-structure characterization of tectosilicates publication-title: J. Mol. Struct. – volume: 141 start-page: 61 year: 2021 end-page: 67 ident: bib15 article-title: Comparative analysis of image classification algorithms based on traditional machine learning and deep learning publication-title: Pattern Recogn. Lett. – volume: 53 start-page: 2067 year: 2018 end-page: 2077 ident: bib33 article-title: Raman spectroscopy of shocked enstatite‐rich meteorites publication-title: Meteoritics Planet Sci. – start-page: 1 year: 2015 end-page: 30 ident: bib26 article-title: Highlights in Mineralogical Crystallography – volume: 2 start-page: 253 year: 2020 end-page: 271 ident: bib2 article-title: Fundamental understanding and applications of plasmon-enhanced Raman spectroscopy publication-title: Nat. Rev. Phys. – volume: 23 start-page: 1 year: 2022 end-page: 12 ident: bib9 article-title: Visible particle identification using Raman spectroscopy and machine learning publication-title: AAPS PharmSciTech – volume: 54 start-page: 1 year: 2021 end-page: 40 ident: bib12 article-title: Deep learning--based text classification: a comprehensive review publication-title: ACM Comput. Surv. – volume: 48 start-page: 1289 year: 2017 end-page: 1299 ident: bib31 article-title: Comparison of seven portable Raman spectrometers: beryl as a case study publication-title: J. Raman Spectrosc. – volume: 20 start-page: 4414 year: 2019 ident: bib45 article-title: Independent component analysis for unraveling the complexity of cancer omics datasets publication-title: Int. J. Mol. Sci. – year: 2017 ident: bib20 article-title: Xception: deep learning with depthwise separable convolutions publication-title: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition – volume: 60 start-page: 2441 year: 2004 end-page: 2448 ident: bib27 article-title: EPR, optical, infrared and Raman spectral studies of Actinolite mineral publication-title: Spectrochim. Acta Mol. Biomol. Spectrosc. – volume: 38 start-page: 135 year: 2007 end-page: 141 ident: bib29 article-title: Raman spectroscopy of the borosilicate mineral ferroaxinite publication-title: J. Raman Spectrosc.: An International Journal for Original Work in all Aspects of Raman Spectroscopy, Including Higher Order Processes, and also Brillouin and Rayleigh Scattering – volume: 44 start-page: 1181 year: 2013 end-page: 1186 ident: bib28 article-title: Raman spectra of selected mineral phases of the Morasko iron meteorite publication-title: J. Raman Spectrosc. – volume: 11 start-page: 1 year: 2021 end-page: 13 ident: bib4 article-title: COVID-19 salivary Raman fingerprint: innovative approach for the detection of current and past SARS-CoV-2 infections publication-title: Sci. Rep. – volume: 2 start-page: 36 year: 2016 ident: bib40 article-title: A taxonomy of deep convolutional neural nets for computer vision publication-title: Front. Robot. AI – volume: 198 year: 2019 ident: bib8 article-title: Raman spectroscopy as a tool to determine the thermal maturity of organic matter: application to sedimentary, metamorphic and structural geology publication-title: Earth Sci. Rev. – volume: 246 start-page: 60 year: 2019 ident: 10.1016/j.aca.2023.341758_bib10 article-title: The quantitative assessment of coke morphology based on the Raman spectroscopic characterization of serial petroleum cokes publication-title: Fuel doi: 10.1016/j.fuel.2019.02.096 – volume: 44 start-page: 1181 issue: 8 year: 2013 ident: 10.1016/j.aca.2023.341758_bib28 article-title: Raman spectra of selected mineral phases of the Morasko iron meteorite publication-title: J. Raman Spectrosc. doi: 10.1002/jrs.4340 – volume: 48 start-page: 1289 issue: 10 year: 2017 ident: 10.1016/j.aca.2023.341758_bib31 article-title: Comparison of seven portable Raman spectrometers: beryl as a case study publication-title: J. Raman Spectrosc. doi: 10.1002/jrs.5214 – year: 1995 ident: 10.1016/j.aca.2023.341758_bib35 article-title: Raman spectroscopy of fluorophosphate and fluorovanadate laser crystals – year: 2019 ident: 10.1016/j.aca.2023.341758_bib17 article-title: Fully learnable group convolution for acceleration of deep neural networks – start-page: 1 year: 2015 ident: 10.1016/j.aca.2023.341758_bib26 – volume: 657 start-page: 152 year: 2016 ident: 10.1016/j.aca.2023.341758_bib32 article-title: The effect of strong magnetic field on the microstructure of pure diopside and diopside doped with Fe3+ or Mn2+ publication-title: J. Alloys Compd. doi: 10.1016/j.jallcom.2015.09.273 – year: 2019 ident: 10.1016/j.aca.2023.341758_bib38 article-title: Efficientnet: rethinking model scaling for convolutional neural networks – volume: 13 start-page: 1738 issue: 4 year: 2019 ident: 10.1016/j.aca.2023.341758_bib16 article-title: A review of deep learning research publication-title: KSII Transactions on Internet and Information Systems (TIIS) – volume: 151 year: 2021 ident: 10.1016/j.aca.2023.341758_bib41 article-title: 1D convolutional neural networks and applications: a survey publication-title: Mech. Syst. Signal Process. doi: 10.1016/j.ymssp.2020.107398 – volume: 67 start-page: 1344 issue: 7 year: 2020 ident: 10.1016/j.aca.2023.341758_bib42 article-title: Channel attention module with multiscale grid average pooling for breast cancer segmentation in an ultrasound image publication-title: IEEE Trans. Ultrason. Ferroelectrics Freq. Control – volume: 38 start-page: 135 issue: 2 year: 2007 ident: 10.1016/j.aca.2023.341758_bib29 article-title: Raman spectroscopy of the borosilicate mineral ferroaxinite publication-title: J. Raman Spectrosc.: An International Journal for Original Work in all Aspects of Raman Spectroscopy, Including Higher Order Processes, and also Brillouin and Rayleigh Scattering doi: 10.1002/jrs.1574 – start-page: 89 year: 2006 ident: 10.1016/j.aca.2023.341758_bib44 – volume: 2 start-page: 253 issue: 5 year: 2020 ident: 10.1016/j.aca.2023.341758_bib2 article-title: Fundamental understanding and applications of plasmon-enhanced Raman spectroscopy publication-title: Nat. Rev. Phys. doi: 10.1038/s42254-020-0171-y – volume: 60 start-page: 2441 issue: 11 year: 2004 ident: 10.1016/j.aca.2023.341758_bib27 article-title: EPR, optical, infrared and Raman spectral studies of Actinolite mineral publication-title: Spectrochim. Acta Mol. Biomol. Spectrosc. doi: 10.1016/j.saa.2003.12.021 – volume: 40 year: 2021 ident: 10.1016/j.aca.2023.341758_bib43 article-title: Conceptual and empirical comparison of dimensionality reduction algorithms (pca, kpca, lda, mds, svd, lle, isomap, le, ica, t-sne) publication-title: Comput. Sci. Rev. doi: 10.1016/j.cosrev.2021.100378 – volume: 9 start-page: 57 issue: 2 year: 2019 ident: 10.1016/j.aca.2023.341758_bib3 article-title: A review on surface-enhanced Raman scattering publication-title: Biosensors doi: 10.3390/bios9020057 – volume: 11 start-page: 1 issue: 1 year: 2021 ident: 10.1016/j.aca.2023.341758_bib4 article-title: COVID-19 salivary Raman fingerprint: innovative approach for the detection of current and past SARS-CoV-2 infections publication-title: Sci. Rep. doi: 10.1038/s41598-021-84565-3 – volume: 20 start-page: 4414 issue: 18 year: 2019 ident: 10.1016/j.aca.2023.341758_bib45 article-title: Independent component analysis for unraveling the complexity of cancer omics datasets publication-title: Int. J. Mol. Sci. doi: 10.3390/ijms20184414 – year: 2018 ident: 10.1016/j.aca.2023.341758_bib22 article-title: Shufflenet: an extremely efficient convolutional neural network for mobile devices – volume: 410 year: 2021 ident: 10.1016/j.aca.2023.341758_bib7 article-title: Analysis of environmental nanoplastics: progress and challenges publication-title: Chem. Eng. J. doi: 10.1016/j.cej.2020.128208 – volume: 2 start-page: 36 year: 2016 ident: 10.1016/j.aca.2023.341758_bib40 article-title: A taxonomy of deep convolutional neural nets for computer vision publication-title: Front. Robot. AI doi: 10.3389/frobt.2015.00036 – year: 2019 ident: 10.1016/j.aca.2023.341758_bib46 article-title: Pneumonia detection using CNN based feature extraction – volume: 23 start-page: 1 issue: 6 year: 2022 ident: 10.1016/j.aca.2023.341758_bib9 article-title: Visible particle identification using Raman spectroscopy and machine learning publication-title: AAPS PharmSciTech doi: 10.1208/s12249-022-02335-4 – volume: 11 start-page: 187 issue: 6 year: 2021 ident: 10.1016/j.aca.2023.341758_bib5 article-title: Application of Raman spectroscopic methods in food safety: a review publication-title: Biosensors doi: 10.3390/bios11060187 – volume: 49 issue: 10 year: 2020 ident: 10.1016/j.aca.2023.341758_bib11 article-title: Raman mineral recognition method based on all-optical diffraction deep neural network publication-title: 红外与激光工程 – volume: 85 start-page: 114 year: 2017 ident: 10.1016/j.aca.2023.341758_bib14 article-title: Deep learning for biological image classification publication-title: Expert Syst. Appl. doi: 10.1016/j.eswa.2017.05.039 – volume: 33 start-page: 917 issue: 4 year: 2019 ident: 10.1016/j.aca.2023.341758_bib13 article-title: Deep learning for time series classification: a review publication-title: Data Min. Knowl. Discov. doi: 10.1007/s10618-019-00619-1 – year: 2021 ident: 10.1016/j.aca.2023.341758_bib25 – volume: 16 start-page: 4681 issue: 7 year: 2019 ident: 10.1016/j.aca.2023.341758_bib39 article-title: Deep residual shrinkage networks for fault diagnosis publication-title: IEEE Trans. Ind. Inf. doi: 10.1109/TII.2019.2943898 – volume: 32 year: 2019 ident: 10.1016/j.aca.2023.341758_bib37 article-title: A meta-analysis of overfitting in machine learning publication-title: Adv. Neural Inf. Process. Syst. – volume: 198 year: 2019 ident: 10.1016/j.aca.2023.341758_bib8 article-title: Raman spectroscopy as a tool to determine the thermal maturity of organic matter: application to sedimentary, metamorphic and structural geology publication-title: Earth Sci. Rev. doi: 10.1016/j.earscirev.2019.102936 – volume: 3 issue: 10 year: 2019 ident: 10.1016/j.aca.2023.341758_bib6 article-title: Durable cellulose–sulfur composites derived from agricultural and petrochemical waste publication-title: Adv. Sustain. Syst. doi: 10.1002/adsu.201900062 – year: 2017 ident: 10.1016/j.aca.2023.341758_bib23 – year: 2019 ident: 10.1016/j.aca.2023.341758_bib47 article-title: On the efficacy of knowledge distillation – volume: 53 start-page: 2067 issue: 10 year: 2018 ident: 10.1016/j.aca.2023.341758_bib33 article-title: Raman spectroscopy of shocked enstatite‐rich meteorites publication-title: Meteoritics Planet Sci. doi: 10.1111/maps.13106 – volume: 54 start-page: 1 issue: 3 year: 2021 ident: 10.1016/j.aca.2023.341758_bib12 article-title: Deep learning--based text classification: a comprehensive review publication-title: ACM Comput. Surv. doi: 10.1145/3439726 – year: 2018 ident: 10.1016/j.aca.2023.341758_bib19 article-title: Pointwise convolutional neural networks – volume: 43 start-page: 649 year: 2016 ident: 10.1016/j.aca.2023.341758_bib34 article-title: High-pressure behavior of natural single-crystal epidote and clinozoisite up to 40 GPa publication-title: Phys. Chem. Miner. doi: 10.1007/s00269-016-0824-7 – year: 2019 ident: 10.1016/j.aca.2023.341758_bib18 article-title: Depthwise convolution is all you need for learning multiple visual domains – year: 2017 ident: 10.1016/j.aca.2023.341758_bib20 article-title: Xception: deep learning with depthwise separable convolutions – volume: 141 start-page: 61 year: 2021 ident: 10.1016/j.aca.2023.341758_bib15 article-title: Comparative analysis of image classification algorithms based on traditional machine learning and deep learning publication-title: Pattern Recogn. Lett. doi: 10.1016/j.patrec.2020.07.042 – volume: 924 start-page: 413 year: 2009 ident: 10.1016/j.aca.2023.341758_bib36 article-title: Minerals from Macedonia. XXIV. Spectra-structure characterization of tectosilicates publication-title: J. Mol. Struct. doi: 10.1016/j.molstruc.2009.01.001 – volume: 205 start-page: 582 year: 2018 ident: 10.1016/j.aca.2023.341758_bib30 article-title: Raman spectroscopy and structural study of baryte-hashemite solid solution from pyrometamorphic rocks of the Hatrurim Complex, Israel publication-title: Spectrochim. Acta Mol. Biomol. Spectrosc. doi: 10.1016/j.saa.2018.07.079 – volume: vol. 2 year: 2015 ident: 10.1016/j.aca.2023.341758_bib24 – volume: 14 start-page: 1 year: 2019 ident: 10.1016/j.aca.2023.341758_bib1 article-title: Raman techniques: fundamentals and frontiers publication-title: Nanoscale Res. Lett. doi: 10.1186/s11671-019-3039-2 – year: 2021 ident: 10.1016/j.aca.2023.341758_bib21 article-title: Efficient computation of depthwise separable convolution in MoblieNet deep neural network models
SSID	ssj0002104
Score	2.459535
Snippet	In recent years, Raman spectroscopy combined with deep learning techniques has been widely used in various fields such as medical, chemical, and geological....
SourceID	proquest crossref elsevier
SourceType	Aggregation Database Enrichment Source Index Database Publisher
StartPage	341758
SubjectTerms	Lightweight model Model compression Raman spectroscopy RamanCMP
Title	RamanCMP: A Raman spectral classification acceleration method based on lightweight model and model compression techniques
URI	https://dx.doi.org/10.1016/j.aca.2023.341758 https://www.proquest.com/docview/2865787137
Volume	1278
WOSCitedRecordID	wos001070863600001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
hasFullText	1
inHoldings	1
isFullTextHit
isPrint
journalDatabaseRights	– providerCode: PRVESC databaseName: Elsevier SD Freedom Collection Journals 2021 customDbUrl: eissn: 1873-4324 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0002104 issn: 0003-2670 databaseCode: AIEXJ dateStart: 19950110 isFulltext: true titleUrlDefault: https://www.sciencedirect.com providerName: Elsevier
link	http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwtV1bb9MwFLa6DgleEFcxLpOREA9UQaud1jZvVdVxUSkT6qSKF8tx0qlTycraju5P8Js5x5e0G2MaD7ykzlGSRv2--hz73Ah5lQqusiYbJ6wt0iTNhUoUECdhaaGsMqxtC1dnti8GAzkaqYNa7VfMhTmbirKUq5Wa_VeoQQZgY-rsP8BdPRQEMAbQ4Qiww_FGwH81303Z_Xzgc87dWcPlU2IqvkVjGaODPO7GWlA7gQS-mXQD9VqOPoQprtt_uq1T3zDHORr8CCPRfQRt2ajKwM43LV1X7QQ3yjFfHGsSYNmOSge8D5HA34ry6HxZEbQbckXw8yrZURT2J1G0mmxuXDAXAufzKsNcKwVPsCCgV0VXyOIEzXyXnzDHgt4Vvtz7H9O_34k4BipiSSnG366vvVhqe_BF7x_2-3rYGw1fz34k2IUMvfWhJcsW2WaipWSdbHc-9kafKt0OC2QXpxBfM_rJXcTgpW_9m6VzSec7Q2Z4j9wNKxDa8cy5T2pF-YDc7sbGfw_JeWTQO9qhbkwjf-hF_tBN_lDPH-r4Q-F8gz_UsYYCf8Jogz90zZ9H5HC_N-x-SEKHjsRyvrdIcjAnM2akyjJhlGmiUzuzbWY5Zreo3MACIjcFk_A2wqaiLXhuxlIJXlhj04w_JvXypCyeEAqTSZtbKXkm81QV0vCMZ1lLNo0ct3ia7pC9-GtqG8rXYxeVqY5xiscaANAIgPYA7JA31S0zX7vluovTCJEOxqc3KjWQ67rbXkY4NcCE3jZTFifLucaUb9CGTS6e3uCaZ-TO-j_ynNQXp8viBbllzxaT-eku2RIjuRvI-BuEsrKm
linkProvider	Elsevier
openUrl	ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=RamanCMP%3A+A+Raman+spectral+classification+acceleration+method+based+on+lightweight+model+and+model+compression+techniques&rft.jtitle=Analytica+chimica+acta&rft.au=Gong%2C+Zengyun&rft.au=Chen%2C+Chen&rft.au=Chen%2C+Cheng&rft.au=Li%2C+Chenxi&rft.date=2023-10-16&rft.issn=1873-4324&rft.eissn=1873-4324&rft.volume=1278&rft.spage=341758&rft_id=info:doi/10.1016%2Fj.aca.2023.341758&rft.externalDBID=NO_FULL_TEXT
thumbnail_l	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=0003-2670&client=summon
thumbnail_m	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=0003-2670&client=summon
thumbnail_s	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=0003-2670&client=summon