Activity recognition using a combination of high gain observer and deep learning computer vision algorithms
•The paper develops a novel method to identify daily living activities of a person using a single wearable sensor on the person's chest.•This paper presents the idea of converting sensor readings to images using spectrograms, so that image processing algorithms can be utilized.•The paper demons...
Uloženo v:
| Vydáno v: | Intelligent systems with applications Ročník 18; s. 200213 |
|---|---|
| Hlavní autoři: | , , |
| Médium: | Journal Article |
| Jazyk: | angličtina |
| Vydáno: |
Elsevier Ltd
01.05.2023
Elsevier |
| Témata: | |
| ISSN: | 2667-3053, 2667-3053 |
| On-line přístup: | Získat plný text |
| Tagy: |
Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
|
| Abstract | •The paper develops a novel method to identify daily living activities of a person using a single wearable sensor on the person's chest.•This paper presents the idea of converting sensor readings to images using spectrograms, so that image processing algorithms can be utilized.•The paper demonstrates how a state estimation observer can highly improve the performance of a deep learning activity recognition algorithm by creating more meaningful input signals for the learning algorithm.•The paper shows how to perform transfer learning from networks pre-trained on millions of images, thus showing how we can train a powerful deep learning network for activity recognition even with just small datasets.•Extensive experimental validation by presenting data from 7 human subjects collected in their home environments.
Inertial sensors have become increasingly popular in human activity classification due to their ease of use and affordability. This paper proposes a novel algorithm for human activity recognition that is a combination of a high-gain observer and deep learning computer vision classification algorithms. The nonlinear high-gain observer designed using Lyapunov analysis accurately estimates the attitude of the chest of a human subject using measurements from a single Inertial Measurement Unit (IMU). The signals processed by the observer are then converted into spectrograms to obtain “images” of the frequency response of the signals. The images for activities from a dataset of 7 human subjects are annotated and used for training/ fine-tuning of several well-known deep learning algorithms for image processing. The results from the best combination of our algorithms shows an exceptional accuracy of 98% for activity recognition. Using deep learning computer vision algorithms, this paper shows how to perform transfer learning from networks pre-trained on millions of images, thus showing how we can train a powerful deep learning network for activity recognition even with just small datasets. The algorithm that uses the high gain observer is shown to perform significantly better than an algorithm based on raw accelerometer and gyro signals. |
|---|---|
| AbstractList | Inertial sensors have become increasingly popular in human activity classification due to their ease of use and affordability. This paper proposes a novel algorithm for human activity recognition that is a combination of a high-gain observer and deep learning computer vision classification algorithms. The nonlinear high-gain observer designed using Lyapunov analysis accurately estimates the attitude of the chest of a human subject using measurements from a single Inertial Measurement Unit (IMU). The signals processed by the observer are then converted into spectrograms to obtain “images” of the frequency response of the signals. The images for activities from a dataset of 7 human subjects are annotated and used for training/ fine-tuning of several well-known deep learning algorithms for image processing. The results from the best combination of our algorithms shows an exceptional accuracy of 98% for activity recognition. Using deep learning computer vision algorithms, this paper shows how to perform transfer learning from networks pre-trained on millions of images, thus showing how we can train a powerful deep learning network for activity recognition even with just small datasets. The algorithm that uses the high gain observer is shown to perform significantly better than an algorithm based on raw accelerometer and gyro signals. •The paper develops a novel method to identify daily living activities of a person using a single wearable sensor on the person's chest.•This paper presents the idea of converting sensor readings to images using spectrograms, so that image processing algorithms can be utilized.•The paper demonstrates how a state estimation observer can highly improve the performance of a deep learning activity recognition algorithm by creating more meaningful input signals for the learning algorithm.•The paper shows how to perform transfer learning from networks pre-trained on millions of images, thus showing how we can train a powerful deep learning network for activity recognition even with just small datasets.•Extensive experimental validation by presenting data from 7 human subjects collected in their home environments. Inertial sensors have become increasingly popular in human activity classification due to their ease of use and affordability. This paper proposes a novel algorithm for human activity recognition that is a combination of a high-gain observer and deep learning computer vision classification algorithms. The nonlinear high-gain observer designed using Lyapunov analysis accurately estimates the attitude of the chest of a human subject using measurements from a single Inertial Measurement Unit (IMU). The signals processed by the observer are then converted into spectrograms to obtain “images” of the frequency response of the signals. The images for activities from a dataset of 7 human subjects are annotated and used for training/ fine-tuning of several well-known deep learning algorithms for image processing. The results from the best combination of our algorithms shows an exceptional accuracy of 98% for activity recognition. Using deep learning computer vision algorithms, this paper shows how to perform transfer learning from networks pre-trained on millions of images, thus showing how we can train a powerful deep learning network for activity recognition even with just small datasets. The algorithm that uses the high gain observer is shown to perform significantly better than an algorithm based on raw accelerometer and gyro signals. |
| ArticleNumber | 200213 |
| Author | McGovern, R. Rajamani, R. Nouriani, A. |
| Author_xml | – sequence: 1 givenname: A. surname: Nouriani fullname: Nouriani, A. email: nouri011@umn.edu organization: Mechanical Engineering at the University of Minnesota – Twin Cities, Minneapolis, MN 55455, United States of America – sequence: 2 givenname: R. surname: McGovern fullname: McGovern, R. email: rmcgover@umn.edu organization: Department of Neurosurgery, University of Minnesota-Twin Cities, Minneapolis, MN 55455, United States of America – sequence: 3 givenname: R. orcidid: 0000-0001-9931-7419 surname: Rajamani fullname: Rajamani, R. email: rajamani@umn.edu organization: Mechanical Engineering at the University of Minnesota – Twin Cities, Minneapolis, MN 55455, United States of America |
| BookMark | eNp9kctqGzEUhoeSQNM0L5DVvIBd3WcM2YTQSyDQTbsWR7fxccaSkRSXvH01dguli2x0OdL3I53vQ3cRU_Rdd0vJmhKqPu3WWH7BmhHG20AY5e-6K6bUsOJE8ot_1u-7m1J2pN0ZKeVCXHXP97biEetrn71NU8SKKfYvBePUQ2_T3mCEUy2FfovTtp8A28YUn48-9xBd77w_9LOHHBeqMYeX2o6OWBYO5illrNt9-dhdBpiLv_kzX3c_v3z-8fBt9fT96-PD_dPKCkrqym4k0FEEIcMmCAGjM9JYRQyzynEQ3DhgMFBquXReSWmIcW4YOLNemsHx6-7xnOsS7PQh4x7yq06A-lRIedKQK9rZa-EY3Ug1bIaghB34OEolQ1BmZIGMAlrWeM6yOZWSfdAW66khNQPOmhK9ONA7vTjQiwN9dtBQ9h_69ylvQndnyLcGHdFnXSz6aL3DJqi2H-Bb-G9rq6Pz |
| CitedBy_id | crossref_primary_10_1051_epjconf_202533002007 crossref_primary_10_1016_j_engappai_2024_109172 |
| Cites_doi | 10.1109/TPAMI.2017.2773081 10.1109/TIM.2015.2504078 10.1016/j.neucom.2021.06.102 10.1109/ACCESS.2020.3037715 10.1016/j.ymssp.2018.02.038 10.3389/fnagi.2023.1117802 10.1016/j.automatica.2020.108814 10.1007/s11263-022-01594-9 10.1016/j.automatica.2010.06.004 10.1016/j.gaitpost.2020.04.010 10.1016/j.automatica.2017.07.067 10.1109/TII.2020.3015934 10.1109/TAC.2018.2882417 10.1109/TAC.2016.2587385 10.1016/j.ifacol.2022.11.152 10.1109/JSEN.2021.3069927 10.1007/978-1-4939-0802-8 10.1016/j.future.2022.12.004 |
| ContentType | Journal Article |
| Copyright | 2023 |
| Copyright_xml | – notice: 2023 |
| DBID | 6I. AAFTH AAYXX CITATION DOA |
| DOI | 10.1016/j.iswa.2023.200213 |
| DatabaseName | ScienceDirect Open Access Titles Elsevier:ScienceDirect:Open Access CrossRef DOAJ Directory of Open Access Journals |
| DatabaseTitle | CrossRef |
| DatabaseTitleList | |
| Database_xml | – sequence: 1 dbid: DOA name: DOAJ Directory of Open Access Journals url: https://www.doaj.org/ sourceTypes: Open Website |
| DeliveryMethod | fulltext_linktorsrc |
| Discipline | Computer Science |
| EISSN | 2667-3053 |
| ExternalDocumentID | oai_doaj_org_article_4d21956797f64c7388565ff6b82f084a 10_1016_j_iswa_2023_200213 S2667305323000388 |
| GroupedDBID | 6I. AAFTH AAXUO ALMA_UNASSIGNED_HOLDINGS AMRAJ FDB GROUPED_DOAJ M41 M~E ROL 0R~ AALRI AAYWO AAYXX ACVFH ADCNI ADVLN AEUPX AFJKZ AFPUW AIGII AITUG AKBMS AKYEP APXCP CITATION |
| ID | FETCH-LOGICAL-c410t-c95a184f45f9f44a8db5bc60b2c6d3a43bda2a711c35de655b0bdd7732ce5b7d3 |
| IEDL.DBID | DOA |
| ISICitedReferencesCount | 2 |
| ISICitedReferencesURI | http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=001307792900008&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D |
| ISSN | 2667-3053 |
| IngestDate | Fri Oct 03 12:50:40 EDT 2025 Tue Nov 18 21:19:02 EST 2025 Sat Nov 29 07:35:15 EST 2025 Sat Nov 04 15:30:39 EDT 2023 |
| IsDoiOpenAccess | true |
| IsOpenAccess | true |
| IsPeerReviewed | true |
| IsScholarly | true |
| Keywords | Deep learning Activity recognition Nonlinear observers Inertial sensors Daily living activities Estimation |
| Language | English |
| License | This is an open access article under the CC BY-NC-ND license. |
| LinkModel | DirectLink |
| MergedId | FETCHMERGED-LOGICAL-c410t-c95a184f45f9f44a8db5bc60b2c6d3a43bda2a711c35de655b0bdd7732ce5b7d3 |
| ORCID | 0000-0001-9931-7419 |
| OpenAccessLink | https://doaj.org/article/4d21956797f64c7388565ff6b82f084a |
| ParticipantIDs | doaj_primary_oai_doaj_org_article_4d21956797f64c7388565ff6b82f084a crossref_citationtrail_10_1016_j_iswa_2023_200213 crossref_primary_10_1016_j_iswa_2023_200213 elsevier_sciencedirect_doi_10_1016_j_iswa_2023_200213 |
| PublicationCentury | 2000 |
| PublicationDate | May 2023 2023-05-00 2023-05-01 |
| PublicationDateYYYYMMDD | 2023-05-01 |
| PublicationDate_xml | – month: 05 year: 2023 text: May 2023 |
| PublicationDecade | 2020 |
| PublicationTitle | Intelligent systems with applications |
| PublicationYear | 2023 |
| Publisher | Elsevier Ltd Elsevier |
| Publisher_xml | – name: Elsevier Ltd – name: Elsevier |
| References | Wang, Rajamani, Bevly (bib0023) 2017; 62 (accessed Aug. 29, 2020). Simonyan, K. & Zisserman, A., “Very Deep Convolutional Networks for Large-Scale Image Recognition,” Sep. 2014. García, Villar, Fáñez, Villar, de la Cal, Cho (bib0043) 2022; 500 Markley, F. & Crassidis, J., “Fundamentals of spacecraft attitude determination and control,” pp. 73–76, 2014. Assidiq, Khalifa, Islam, Khan (bib0013) 2008 Scandaroli, Morin (bib0019) 2011 Mascret, Bielmann, Fall, Bouyer, Gosselin (bib0005) 2018 (accessed Dec. 19, 2021). Tavor (bib0037) 2020 “OptiTrack V120 Duo & Trio Datasheet,” 2018. Rajamani, Jeon, Movahedi, Zemouche (bib0031) 2020; 114 2017. Crassidis, Markley (bib0018) 2006; 30 Nouriani, McGovern, Rajamani (bib48) 2023 Gulli, Pal (bib0035) 2017 Sani, Wiratunga, Massie, Cooper (bib0004) 2017 Zemouche, Rajamani, Phanomchoeng, Boulkroune, Rafaralahy, Zasadzinski (bib0021) 2017; 85 Hung, Thacher, White (bib0011) 1989 Nouriani (bib0046) 2023; 15 Hou (bib0001) 2020 Qiu, Wang, Zhao, Hu (bib0027) 2016; 65 “Runcam 5 Datasheet,” 2020. Mekruksavanich, Jantawong, Jitpattanakul (bib0044) 2022 Krizhevsky, Sutskever, Hinton (bib0040) 2012; 25 Li, Hoiem (bib0010) 2016; 40 Zheng, Wang, Ordieres-Meré (bib0002) 2018; 18 Nouredanesh, Godfrey, Howcroft, Lemaire, Tung (bib0028) 2021; 85 Kong, Fu (bib0030) 2022; 130 Alema Khatun, Abu Yousuf (bib0009) 2020 Park, Kim, Seo (bib0012) 2020 Barra, Lesecq, Zarudniev, Debicki, Mareau, Ouvry (bib0015) 2019 Ramanujam, Perumal, Padmavathi (bib0007) 2021; 21 He, Zhang, Ren, Sun (bib0036) 2016 Khatun (bib0045) 2022; 10 Nouriani, McGovern, Rajamani (bib0024) 2021 Nouriani, McGovern, Rajamani (bib0047) 2022; 55 Demrozi, Pravadelli, Bihorac, Rashidi (bib0029) 2020; 8 Khalil (bib0020) 2015 Boizot, Busvelle, Gauthier (bib0022) 2010; 46 SparkFun OpenLog Artemis - DEV-16832 - SparkFun Electronics. Iandola, F.N., Han, S., Moskewicz, M.W., Ashraf, K., Dally, W.J., & Keutzer, K., “SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and< 0.5MB model size,” Wozniak, Wieczorek, Silka, Polap (bib0026) 2021; 17 Huang, Liu, van der Maaten, Weinberger (bib0041) 2017 Casale, Pujol, Radeva (bib0003) 2011 Li, Liu, Zhang, Hang (bib0014) 2014 Wang, Rajamani (bib0017) 2018; 109 Zemouche, Zhang, Mazenc, Rajamani (bib0032) 2019; 64 Woźniak, Wieczorek, Siłka (bib0025) 2023; 141 Hou (10.1016/j.iswa.2023.200213_bib0001) 2020 Woźniak (10.1016/j.iswa.2023.200213_bib0025) 2023; 141 Hung (10.1016/j.iswa.2023.200213_bib0011) 1989 Wang (10.1016/j.iswa.2023.200213_bib0017) 2018; 109 Li (10.1016/j.iswa.2023.200213_bib0014) 2014 Rajamani (10.1016/j.iswa.2023.200213_bib0031) 2020; 114 Wozniak (10.1016/j.iswa.2023.200213_bib0026) 2021; 17 Nouriani (10.1016/j.iswa.2023.200213_bib0047) 2022; 55 Boizot (10.1016/j.iswa.2023.200213_bib0022) 2010; 46 Qiu (10.1016/j.iswa.2023.200213_bib0027) 2016; 65 Sani (10.1016/j.iswa.2023.200213_bib0004) 2017 Assidiq (10.1016/j.iswa.2023.200213_bib0013) 2008 Mekruksavanich (10.1016/j.iswa.2023.200213_bib0044) 2022 10.1016/j.iswa.2023.200213_bib0033 Kong (10.1016/j.iswa.2023.200213_bib0030) 2022; 130 10.1016/j.iswa.2023.200213_bib0034 Tavor (10.1016/j.iswa.2023.200213_bib0037) 2020 Crassidis (10.1016/j.iswa.2023.200213_bib0018) 2006; 30 Khalil (10.1016/j.iswa.2023.200213_bib0020) 2015 10.1016/j.iswa.2023.200213_bib0016 10.1016/j.iswa.2023.200213_bib0038 Zemouche (10.1016/j.iswa.2023.200213_bib0032) 2019; 64 10.1016/j.iswa.2023.200213_bib0039 Wang (10.1016/j.iswa.2023.200213_bib0023) 2017; 62 García (10.1016/j.iswa.2023.200213_bib0043) 2022; 500 Casale (10.1016/j.iswa.2023.200213_bib0003) 2011 Nouredanesh (10.1016/j.iswa.2023.200213_bib0028) 2021; 85 Demrozi (10.1016/j.iswa.2023.200213_bib0029) 2020; 8 Li (10.1016/j.iswa.2023.200213_bib0010) 2016; 40 Krizhevsky (10.1016/j.iswa.2023.200213_bib0040) 2012; 25 Ramanujam (10.1016/j.iswa.2023.200213_bib0007) 2021; 21 Mascret (10.1016/j.iswa.2023.200213_bib0005) 2018 Zheng (10.1016/j.iswa.2023.200213_bib0002) 2018; 18 Nouriani (10.1016/j.iswa.2023.200213_bib0024) 2021 Gulli (10.1016/j.iswa.2023.200213_bib0035) 2017 Nouriani (10.1016/j.iswa.2023.200213_bib48) 2023 Park (10.1016/j.iswa.2023.200213_bib0012) 2020 Huang (10.1016/j.iswa.2023.200213_bib0041) 2017 10.1016/j.iswa.2023.200213_bib0042 Khatun (10.1016/j.iswa.2023.200213_bib0045) 2022; 10 Barra (10.1016/j.iswa.2023.200213_bib0015) 2019 Alema Khatun (10.1016/j.iswa.2023.200213_bib0009) 2020 Zemouche (10.1016/j.iswa.2023.200213_bib0021) 2017; 85 He (10.1016/j.iswa.2023.200213_bib0036) 2016 Nouriani (10.1016/j.iswa.2023.200213_bib0046) 2023; 15 Scandaroli (10.1016/j.iswa.2023.200213_bib0019) 2011 |
| References_xml | – start-page: 225 year: 2020 end-page: 234 ident: bib0001 article-title: A study on IMU-based human activity recognition using deep learning and traditional machine learning publication-title: 2020 5th Int. Conf. Comput. Commun. Syst. ICCCS 2020 – volume: 62 start-page: 1940 year: 2017 end-page: 1945 ident: bib0023 article-title: Observer design for parameter varying differentiable nonlinear systems, with application to slip angle estimation publication-title: IEEE Trans Automat Contr – year: 2015 ident: bib0020 article-title: Nonlinear control – volume: 114 year: 2020 ident: bib0031 article-title: On the need for switched-gain observers for non-monotonic nonlinear systems publication-title: Automatica – start-page: 4700 year: 2017 end-page: 4708 ident: bib0041 article-title: Densely Connected Convolutional Networks publication-title: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR) – volume: 141 start-page: 489 year: 2023 end-page: 499 ident: bib0025 article-title: BiLSTM deep neural network model for imbalanced medical data of IoT systems publication-title: Futur Gener Comput Syst – year: 2020 ident: bib0009 article-title: Human activity recognition using smartphone sensor based on selective classifiers publication-title: 2020 2nd Int. Conf. Sustain. Technol. Ind. 4.0, STI 2020 – start-page: 239 year: 2018 end-page: 242 ident: bib0005 article-title: Real-time human physical activity recognition with low latency prediction feedback using raw IMU Data publication-title: Proc. Annu. Int. Conf. IEEE Eng. Med. Biol. Soc. EMBS – start-page: 96 year: 2017 end-page: 104 ident: bib0035 article-title: Deep learning with keras – volume: 8 start-page: 210816 year: 2020 end-page: 210836 ident: bib0029 article-title: Human activity recognition using inertial, physiological and environmental sensors: A comprehensive survey publication-title: IEEE access : practical innovations, open solutions – start-page: 7383 year: 2020 end-page: 7390 ident: bib0037 article-title: Do not have enough data? Seep learning to the rescue! publication-title: Proc. AAAI Conf. Artif. Intell. – volume: 18 start-page: 2146 year: 2018 ident: bib0002 article-title: Comparison of data preprocessing approaches for applying deep learning to human activity recognition in the context of industry 4.0 publication-title: Sensors 2018 – start-page: 330 year: 2017 end-page: 344 ident: bib0004 article-title: kNN sampling for personalised human activity recognition publication-title: Lect Notes Comput Sci (including Subser Lect Notes Artif Intell Lect Notes Bioinformatics) – volume: 64 start-page: 3194 year: 2019 end-page: 3209 ident: bib0032 article-title: High-Gain Nonlinear Observer With Lower Tuning Parameter publication-title: IEEE Trans Automat Contr – volume: 55 start-page: 1 year: 2022 end-page: 6 ident: bib0047 article-title: Deep-learning-based human activity recognition using wearable sensors publication-title: IFAC-PapersOnLine – start-page: 1201 year: 2019 end-page: 1206 ident: bib0015 article-title: Localization system in GPS-denied environments using radar and imu measurements: Application to a smart white cane publication-title: 2019 18th Eur. Control Conf. ECC 2019 – start-page: 770 year: 2016 end-page: 778 ident: bib0036 article-title: Deep residual learning for image recognition publication-title: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR) – reference: “OptiTrack V120 Duo & Trio Datasheet,” 2018. – volume: 109 start-page: 268 year: 2018 end-page: 284 ident: bib0017 article-title: Direction cosine matrix estimation with an inertial measurement unit publication-title: Mech Syst Signal Process – volume: 46 start-page: 1483 year: 2010 end-page: 1488 ident: bib0022 article-title: An adaptive high-gain observer for nonlinear systems publication-title: Automatica – volume: 17 start-page: 2101 year: 2021 end-page: 2111 ident: bib0026 article-title: Body pose prediction based on motion sensor data and recurrent neural network publication-title: IEEE Trans Ind Informatics – start-page: 289 year: 2011 end-page: 296 ident: bib0003 article-title: Human activity recognition from accelerometer data using a wearable device publication-title: Lect Notes Comput Sci (including Subser Lect Notes Artif Intell Lect Notes Bioinformatics) – volume: 130 start-page: 1366 year: 2022 end-page: 1401 ident: bib0030 article-title: Human action recognition and prediction: A survey publication-title: Int J Comput Vis – volume: 65 start-page: 939 year: 2016 end-page: 950 ident: bib0027 article-title: Using distributed wearable sensors to measure and evaluate human lower limb motions publication-title: Ieee Transactions on Instrumentation and Measurement – reference: (accessed Aug. 29, 2020). – reference: , 2017. – reference: (accessed Dec. 19, 2021). – volume: 30 start-page: 12 year: 2006 end-page: 28 ident: bib0018 article-title: Survey of nonlinear attitude estimation methods publication-title: arcaiaaorg – volume: 85 start-page: 412 year: 2017 end-page: 425 ident: bib0021 article-title: Circle criterion-based publication-title: Automatica – start-page: 4524 year: 2011 end-page: 4530 ident: bib0019 article-title: Nonlinear filter design for pose and IMU bias estimation publication-title: Proceedings - IEEE International Conference on Robotics and Automation – volume: 85 start-page: 178 year: 2021 end-page: 190 ident: bib0028 article-title: Fall risk assessment in the wild: A critical examination of wearable sensor use in free-living conditions publication-title: Gait & posture – volume: 500 start-page: 231 year: 2022 end-page: 240 ident: bib0043 article-title: Towards effective detection of elderly falls with CNN-LSTM neural networks publication-title: Neurocomputing – start-page: 153 year: 1989 end-page: 158 ident: bib0011 article-title: Calibration of accelerometer triad of an IMU with drifting Z-accelerometer bias publication-title: IEEE proceedings of the national aerospace and electronics conference – year: 2023 ident: bib48 publication-title: Activity Recognition Using a High Gain Observer and Spectrograms, Proceedings of the 2023 American Control Conference, May 31-June 2 – volume: 25 year: 2012 ident: bib0040 article-title: ImageNet Classification with Deep Convolutional Neural Networks publication-title: Advances in Neural Information Processing Systems – reference: Markley, F. & Crassidis, J., “Fundamentals of spacecraft attitude determination and control,” pp. 73–76, 2014. – volume: 10 year: 2022 ident: bib0045 article-title: Deep CNN-LSTM with self-attention model for human activity recognition using wearable sensor publication-title: IEEE J Transl Eng Heal Med – reference: “Runcam 5 Datasheet,” 2020. – start-page: 82 year: 2008 end-page: 88 ident: bib0013 article-title: Real time lane detection for autonomous vehicles publication-title: Proceedings of the International Conference on Computer and Communication Engineering 2008, ICCCE08: Global Links for Human Development – volume: 21 start-page: 1309 year: 2021 end-page: 13040 ident: bib0007 article-title: Human activity recognition with smartphone and wearable sensors using deep learning techniques: a review publication-title: IEEE Sens J – year: 2014 ident: bib0014 article-title: LIDAR/MEMS IMU integrated navigation (SLAM) method for a small UAV in indoor environments publication-title: 2014 DGON Inert. Sensors Syst. ISS 2014 - Proc. – reference: “SparkFun OpenLog Artemis - DEV-16832 - SparkFun Electronics.” – start-page: 342 year: 2022 end-page: 345 ident: bib0044 article-title: LSTM-XGB: a new deep learning model for human activity recognition based on LSTM and XGBoost publication-title: 7th Int. Conf. Digit. Arts, Media Technol. DAMT 2022 5th ECTI North. Sect. Conf. Electr. Electron. Comput. Telecommun. Eng. NCON 2022 – reference: Simonyan, K. & Zisserman, A., “Very Deep Convolutional Networks for Large-Scale Image Recognition,” Sep. 2014. – volume: 15 start-page: 91 year: 2023 ident: bib0046 article-title: Real world validation of activity recognition algorithm and development of novel behavioral biomarkers of falls in aged control and movement disorder patients publication-title: Front Aging Neurosci – reference: Iandola, F.N., Han, S., Moskewicz, M.W., Ashraf, K., Dally, W.J., & Keutzer, K., “SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and< 0.5MB model size,” – volume: 40 start-page: 2935 year: 2016 end-page: 2947 ident: bib0010 article-title: Learning without forgetting publication-title: Ieee Transactions on Pattern Analysis and Machine Intelligence – start-page: 12 year: 2021 end-page: 25 ident: bib0024 article-title: Step length estimation using inertial measurements units publication-title: Proceedings of the 2021 American Control Conference (ACC), May 25-28 – start-page: 800 year: 2020 end-page: 803 ident: bib0012 article-title: Effects of initial attitude estimation errors on loosely coupled smartphone GPS/IMU integration system publication-title: Int. Conf. Control. Autom. Syst. – volume: 10 year: 2022 ident: 10.1016/j.iswa.2023.200213_bib0045 article-title: Deep CNN-LSTM with self-attention model for human activity recognition using wearable sensor publication-title: IEEE J Transl Eng Heal Med – volume: 40 start-page: 2935 issue: 12 year: 2016 ident: 10.1016/j.iswa.2023.200213_bib0010 article-title: Learning without forgetting publication-title: Ieee Transactions on Pattern Analysis and Machine Intelligence doi: 10.1109/TPAMI.2017.2773081 – volume: 65 start-page: 939 issue: 4 year: 2016 ident: 10.1016/j.iswa.2023.200213_bib0027 article-title: Using distributed wearable sensors to measure and evaluate human lower limb motions publication-title: Ieee Transactions on Instrumentation and Measurement doi: 10.1109/TIM.2015.2504078 – volume: 18 start-page: 2146 year: 2018 ident: 10.1016/j.iswa.2023.200213_bib0002 article-title: Comparison of data preprocessing approaches for applying deep learning to human activity recognition in the context of industry 4.0 – volume: 500 start-page: 231 year: 2022 ident: 10.1016/j.iswa.2023.200213_bib0043 article-title: Towards effective detection of elderly falls with CNN-LSTM neural networks publication-title: Neurocomputing doi: 10.1016/j.neucom.2021.06.102 – volume: 8 start-page: 210816 year: 2020 ident: 10.1016/j.iswa.2023.200213_bib0029 article-title: Human activity recognition using inertial, physiological and environmental sensors: A comprehensive survey publication-title: IEEE access : practical innovations, open solutions doi: 10.1109/ACCESS.2020.3037715 – volume: 109 start-page: 268 year: 2018 ident: 10.1016/j.iswa.2023.200213_bib0017 article-title: Direction cosine matrix estimation with an inertial measurement unit publication-title: Mech Syst Signal Process doi: 10.1016/j.ymssp.2018.02.038 – start-page: 12 year: 2021 ident: 10.1016/j.iswa.2023.200213_bib0024 article-title: Step length estimation using inertial measurements units – start-page: 225 year: 2020 ident: 10.1016/j.iswa.2023.200213_bib0001 article-title: A study on IMU-based human activity recognition using deep learning and traditional machine learning – volume: 15 start-page: 91 year: 2023 ident: 10.1016/j.iswa.2023.200213_bib0046 article-title: Real world validation of activity recognition algorithm and development of novel behavioral biomarkers of falls in aged control and movement disorder patients publication-title: Front Aging Neurosci doi: 10.3389/fnagi.2023.1117802 – volume: 114 year: 2020 ident: 10.1016/j.iswa.2023.200213_bib0031 article-title: On the need for switched-gain observers for non-monotonic nonlinear systems publication-title: Automatica doi: 10.1016/j.automatica.2020.108814 – start-page: 770 year: 2016 ident: 10.1016/j.iswa.2023.200213_bib0036 article-title: Deep residual learning for image recognition – ident: 10.1016/j.iswa.2023.200213_bib0039 – volume: 25 year: 2012 ident: 10.1016/j.iswa.2023.200213_bib0040 article-title: ImageNet Classification with Deep Convolutional Neural Networks – volume: 130 start-page: 1366 issue: 5 year: 2022 ident: 10.1016/j.iswa.2023.200213_bib0030 article-title: Human action recognition and prediction: A survey publication-title: Int J Comput Vis doi: 10.1007/s11263-022-01594-9 – year: 2020 ident: 10.1016/j.iswa.2023.200213_bib0009 article-title: Human activity recognition using smartphone sensor based on selective classifiers – start-page: 82 year: 2008 ident: 10.1016/j.iswa.2023.200213_bib0013 article-title: Real time lane detection for autonomous vehicles – volume: 46 start-page: 1483 issue: 9 year: 2010 ident: 10.1016/j.iswa.2023.200213_bib0022 article-title: An adaptive high-gain observer for nonlinear systems publication-title: Automatica doi: 10.1016/j.automatica.2010.06.004 – volume: 30 start-page: 12 issue: 1 year: 2006 ident: 10.1016/j.iswa.2023.200213_bib0018 article-title: Survey of nonlinear attitude estimation methods publication-title: arcaiaaorg – volume: 85 start-page: 178 year: 2021 ident: 10.1016/j.iswa.2023.200213_bib0028 article-title: Fall risk assessment in the wild: A critical examination of wearable sensor use in free-living conditions publication-title: Gait & posture doi: 10.1016/j.gaitpost.2020.04.010 – start-page: 4700 year: 2017 ident: 10.1016/j.iswa.2023.200213_bib0041 article-title: Densely Connected Convolutional Networks – start-page: 239 year: 2018 ident: 10.1016/j.iswa.2023.200213_bib0005 article-title: Real-time human physical activity recognition with low latency prediction feedback using raw IMU Data – ident: 10.1016/j.iswa.2023.200213_bib0033 – start-page: 342 year: 2022 ident: 10.1016/j.iswa.2023.200213_bib0044 article-title: LSTM-XGB: a new deep learning model for human activity recognition based on LSTM and XGBoost – start-page: 800 year: 2020 ident: 10.1016/j.iswa.2023.200213_bib0012 article-title: Effects of initial attitude estimation errors on loosely coupled smartphone GPS/IMU integration system – volume: 85 start-page: 412 year: 2017 ident: 10.1016/j.iswa.2023.200213_bib0021 article-title: Circle criterion-basedH∞ observer design for Lipschitz and monotonic nonlinear systems – Enhanced LMI conditions and constructive discussions publication-title: Automatica doi: 10.1016/j.automatica.2017.07.067 – start-page: 1201 year: 2019 ident: 10.1016/j.iswa.2023.200213_bib0015 article-title: Localization system in GPS-denied environments using radar and imu measurements: Application to a smart white cane – volume: 17 start-page: 2101 issue: 3 year: 2021 ident: 10.1016/j.iswa.2023.200213_bib0026 article-title: Body pose prediction based on motion sensor data and recurrent neural network publication-title: IEEE Trans Ind Informatics doi: 10.1109/TII.2020.3015934 – year: 2014 ident: 10.1016/j.iswa.2023.200213_bib0014 article-title: LIDAR/MEMS IMU integrated navigation (SLAM) method for a small UAV in indoor environments – volume: 64 start-page: 3194 issue: 8 year: 2019 ident: 10.1016/j.iswa.2023.200213_bib0032 article-title: High-Gain Nonlinear Observer With Lower Tuning Parameter publication-title: IEEE Trans Automat Contr doi: 10.1109/TAC.2018.2882417 – start-page: 153 year: 1989 ident: 10.1016/j.iswa.2023.200213_bib0011 article-title: Calibration of accelerometer triad of an IMU with drifting Z-accelerometer bias – start-page: 330 year: 2017 ident: 10.1016/j.iswa.2023.200213_bib0004 article-title: kNN sampling for personalised human activity recognition publication-title: Lect Notes Comput Sci (including Subser Lect Notes Artif Intell Lect Notes Bioinformatics) – ident: 10.1016/j.iswa.2023.200213_bib0042 – year: 2023 ident: 10.1016/j.iswa.2023.200213_bib48 – start-page: 7383 year: 2020 ident: 10.1016/j.iswa.2023.200213_bib0037 article-title: Do not have enough data? Seep learning to the rescue! – start-page: 96 year: 2017 ident: 10.1016/j.iswa.2023.200213_bib0035 – volume: 62 start-page: 1940 issue: 4 year: 2017 ident: 10.1016/j.iswa.2023.200213_bib0023 article-title: Observer design for parameter varying differentiable nonlinear systems, with application to slip angle estimation publication-title: IEEE Trans Automat Contr doi: 10.1109/TAC.2016.2587385 – ident: 10.1016/j.iswa.2023.200213_bib0038 – volume: 55 start-page: 1 issue: 37 year: 2022 ident: 10.1016/j.iswa.2023.200213_bib0047 article-title: Deep-learning-based human activity recognition using wearable sensors publication-title: IFAC-PapersOnLine doi: 10.1016/j.ifacol.2022.11.152 – volume: 21 start-page: 1309 issue: 12 year: 2021 ident: 10.1016/j.iswa.2023.200213_bib0007 article-title: Human activity recognition with smartphone and wearable sensors using deep learning techniques: a review publication-title: IEEE Sens J doi: 10.1109/JSEN.2021.3069927 – ident: 10.1016/j.iswa.2023.200213_bib0016 doi: 10.1007/978-1-4939-0802-8 – volume: 141 start-page: 489 year: 2023 ident: 10.1016/j.iswa.2023.200213_bib0025 article-title: BiLSTM deep neural network model for imbalanced medical data of IoT systems publication-title: Futur Gener Comput Syst doi: 10.1016/j.future.2022.12.004 – start-page: 289 year: 2011 ident: 10.1016/j.iswa.2023.200213_bib0003 article-title: Human activity recognition from accelerometer data using a wearable device publication-title: Lect Notes Comput Sci (including Subser Lect Notes Artif Intell Lect Notes Bioinformatics) – year: 2015 ident: 10.1016/j.iswa.2023.200213_bib0020 – ident: 10.1016/j.iswa.2023.200213_bib0034 – start-page: 4524 year: 2011 ident: 10.1016/j.iswa.2023.200213_bib0019 article-title: Nonlinear filter design for pose and IMU bias estimation |
| SSID | ssj0002811344 |
| Score | 2.27161 |
| Snippet | •The paper develops a novel method to identify daily living activities of a person using a single wearable sensor on the person's chest.•This paper presents... Inertial sensors have become increasingly popular in human activity classification due to their ease of use and affordability. This paper proposes a novel... |
| SourceID | doaj crossref elsevier |
| SourceType | Open Website Enrichment Source Index Database Publisher |
| StartPage | 200213 |
| SubjectTerms | Activity recognition Daily living activities Deep learning Estimation Inertial sensors Nonlinear observers |
| Title | Activity recognition using a combination of high gain observer and deep learning computer vision algorithms |
| URI | https://dx.doi.org/10.1016/j.iswa.2023.200213 https://doaj.org/article/4d21956797f64c7388565ff6b82f084a |
| Volume | 18 |
| WOSCitedRecordID | wos001307792900008&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: PRVAON databaseName: DOAJ Directory of Open Access Journals customDbUrl: eissn: 2667-3053 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0002811344 issn: 2667-3053 databaseCode: DOA dateStart: 20210101 isFulltext: true titleUrlDefault: https://www.doaj.org/ providerName: Directory of Open Access Journals – providerCode: PRVHPJ databaseName: ROAD: Directory of Open Access Scholarly Resources customDbUrl: eissn: 2667-3053 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0002811344 issn: 2667-3053 databaseCode: M~E dateStart: 20210101 isFulltext: true titleUrlDefault: https://road.issn.org providerName: ISSN International Centre |
| link | http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwrV07T8MwELZQxcDCG1Fe8sCGIuLYjpOxoFYsVAwgdYv8LCklrdoCG7-dc5xWZSkLS6REfkTns-476_N9CF2T2PDY5CqCXFlFzB80KUAWUeIAHQsFe4TqWmxC9PvZYJA_rUl9eU5YKA8cDHfLTOKvtIlcuJRpQbMMIIhzqcoSF2eshkaxyNeSqVF9ZEQIrZVcIQDBLgJXa27MBHJXOf_yRYcSGkgK9FdUqov3rwWntYDT20e7DVLEnfCHB2jLVodob6nCgJtNeYTeOjooQOAVGWhSYc9nH2KJwaEg963NjycO--rEeChLeFH-PBYGkpXBxtopbgQkhr5PmCNcPMdyPJzMysXr-_wYvfS6z_cPUSOhEGlG4kWkcy4hh3OMu9wxJjOjuNJprBKdGioZVUYmUhCiKTc25VzFyhghaKItV8LQE9SqJpU9RdgCDOc2lzyFfjzVihhLlSOaE-cybtuILE1Y6Ka-uJe5GBdLItmo8GYvvNmLYPY2uln1mYbqGhtb3_mVWbX0lbHrD-AvReMvxV_-0kZ8ua5FAzICeIChyg2Tn_3H5Odoxw8ZCJMXqLWYfdhLtK0_F-V8dlW7MDwfv7s_Tkf1Bg |
| linkProvider | Directory of Open Access Journals |
| 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=Activity+recognition+using+a+combination+of+high+gain+observer+and+deep+learning+computer+vision+algorithms&rft.jtitle=Intelligent+systems+with+applications&rft.au=Nouriani%2C+A.&rft.au=McGovern%2C+R.&rft.au=Rajamani%2C+R.&rft.date=2023-05-01&rft.issn=2667-3053&rft.eissn=2667-3053&rft.volume=18&rft.spage=200213&rft_id=info:doi/10.1016%2Fj.iswa.2023.200213&rft.externalDBID=n%2Fa&rft.externalDocID=10_1016_j_iswa_2023_200213 |
| thumbnail_l | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=2667-3053&client=summon |
| thumbnail_m | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=2667-3053&client=summon |
| thumbnail_s | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=2667-3053&client=summon |