Posture Recognition Using Ensemble Deep Models under Various Home Environments
This paper is concerned with posture recognition using ensemble convolutional neural networks (CNNs) in home environments. With the increasing number of elderly people living alone at home, posture recognition is very important for helping elderly people cope with sudden danger. Traditionally, to re...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2020-02-01
|
| Series: | Applied Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2076-3417/10/4/1287 |
| _version_ | 1819024965593202688 |
|---|---|
| author | Yeong-Hyeon Byeon Jae-Yeon Lee Do-Hyung Kim Keun-Chang Kwak |
| author_facet | Yeong-Hyeon Byeon Jae-Yeon Lee Do-Hyung Kim Keun-Chang Kwak |
| author_sort | Yeong-Hyeon Byeon |
| collection | DOAJ |
| description | This paper is concerned with posture recognition using ensemble convolutional neural networks (CNNs) in home environments. With the increasing number of elderly people living alone at home, posture recognition is very important for helping elderly people cope with sudden danger. Traditionally, to recognize posture, it was necessary to obtain the coordinates of the body points, depth, frame information of video, and so on. In conventional machine learning, there is a limitation in recognizing posture directly using only an image. However, with advancements in the latest deep learning, it is possible to achieve good performance in posture recognition using only an image. Thus, we performed experiments based on VGGNet, ResNet, DenseNet, InceptionResNet, and Xception as pre-trained CNNs using five types of preprocessing. On the basis of these deep learning methods, we finally present the ensemble deep model combined by majority and average methods. The experiments were performed by a posture database constructed at the Electronics and Telecommunications Research Institute (ETRI), Korea. This database consists of 51,000 images with 10 postures from 51 home environments. The experimental results reveal that the ensemble system by InceptionResNetV2s with five types of preprocessing shows good performance in comparison to other combination methods and the pre-trained CNN itself. |
| first_indexed | 2024-12-21T05:03:10Z |
| format | Article |
| id | doaj.art-7ad6f1db219d4fb9a204b80e31df900a |
| institution | Directory Open Access Journal |
| issn | 2076-3417 |
| language | English |
| last_indexed | 2024-12-21T05:03:10Z |
| publishDate | 2020-02-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Applied Sciences |
| spelling | doaj.art-7ad6f1db219d4fb9a204b80e31df900a2022-12-21T19:15:11ZengMDPI AGApplied Sciences2076-34172020-02-01104128710.3390/app10041287app10041287Posture Recognition Using Ensemble Deep Models under Various Home EnvironmentsYeong-Hyeon Byeon0Jae-Yeon Lee1Do-Hyung Kim2Keun-Chang Kwak3Department of Control and Instrumentation Engineering, Chosun University, Gwangju 61452, KoreaIntelligent Robotics Research Division, Electronics Telecommunications Research Institute, Daejeon 61452, KoreaIntelligent Robotics Research Division, Electronics Telecommunications Research Institute, Daejeon 61452, KoreaDepartment of Control and Instrumentation Engineering, Chosun University, Gwangju 61452, KoreaThis paper is concerned with posture recognition using ensemble convolutional neural networks (CNNs) in home environments. With the increasing number of elderly people living alone at home, posture recognition is very important for helping elderly people cope with sudden danger. Traditionally, to recognize posture, it was necessary to obtain the coordinates of the body points, depth, frame information of video, and so on. In conventional machine learning, there is a limitation in recognizing posture directly using only an image. However, with advancements in the latest deep learning, it is possible to achieve good performance in posture recognition using only an image. Thus, we performed experiments based on VGGNet, ResNet, DenseNet, InceptionResNet, and Xception as pre-trained CNNs using five types of preprocessing. On the basis of these deep learning methods, we finally present the ensemble deep model combined by majority and average methods. The experiments were performed by a posture database constructed at the Electronics and Telecommunications Research Institute (ETRI), Korea. This database consists of 51,000 images with 10 postures from 51 home environments. The experimental results reveal that the ensemble system by InceptionResNetV2s with five types of preprocessing shows good performance in comparison to other combination methods and the pre-trained CNN itself.https://www.mdpi.com/2076-3417/10/4/1287ensemble deep modelsconvolutional neural networkposture recognitionpreconfigured cnnsposture databasehome environments |
| spellingShingle | Yeong-Hyeon Byeon Jae-Yeon Lee Do-Hyung Kim Keun-Chang Kwak Posture Recognition Using Ensemble Deep Models under Various Home Environments Applied Sciences ensemble deep models convolutional neural network posture recognition preconfigured cnns posture database home environments |
| title | Posture Recognition Using Ensemble Deep Models under Various Home Environments |
| title_full | Posture Recognition Using Ensemble Deep Models under Various Home Environments |
| title_fullStr | Posture Recognition Using Ensemble Deep Models under Various Home Environments |
| title_full_unstemmed | Posture Recognition Using Ensemble Deep Models under Various Home Environments |
| title_short | Posture Recognition Using Ensemble Deep Models under Various Home Environments |
| title_sort | posture recognition using ensemble deep models under various home environments |
| topic | ensemble deep models convolutional neural network posture recognition preconfigured cnns posture database home environments |
| url | https://www.mdpi.com/2076-3417/10/4/1287 |
| work_keys_str_mv | AT yeonghyeonbyeon posturerecognitionusingensembledeepmodelsundervarioushomeenvironments AT jaeyeonlee posturerecognitionusingensembledeepmodelsundervarioushomeenvironments AT dohyungkim posturerecognitionusingensembledeepmodelsundervarioushomeenvironments AT keunchangkwak posturerecognitionusingensembledeepmodelsundervarioushomeenvironments |