Performance Analysis of Deep Learning based Human Activity Recognition Methods
Human Activity Recognition (HAR) is one of the most important branches of human-centered research activities. Along with the development of artificial intelligence, deep learning techniques have gained remarkable success in computer vision. In recent years, there is a growing interest in Human Acti...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
UNIMAS Publisher
2022-10-01
|
| Series: | Journal of Applied Science & Process Engineering |
| Subjects: | |
| Online Access: | https://publisher.unimas.my/ojs/index.php/JASPE/article/view/4639 |
| _version_ | 1811256931171434496 |
|---|---|
| author | Mst. Farzana Aktter Md Anwar Hossain Sohag Sarker AFM Zainul Abadin Mirza AFM Rashidul Hasan |
| author_facet | Mst. Farzana Aktter Md Anwar Hossain Sohag Sarker AFM Zainul Abadin Mirza AFM Rashidul Hasan |
| author_sort | Mst. Farzana Aktter |
| collection | DOAJ |
| description |
Human Activity Recognition (HAR) is one of the most important branches of human-centered research activities. Along with the development of artificial intelligence, deep learning techniques have gained remarkable success in computer vision. In recent years, there is a growing interest in Human Activity Recognition systems applied in healthcare, security surveillance, and human motion-based activities. A HAR system is essentially made of a wearable device equipped with a set of sensors (like accelerometers, gyroscopes, magnetometers, heart-rate sensors, etc.). Different methods are being applied for improving the accuracy and performance of the HAR system. In this paper, we implement Artificial Neural Network (ANN), and Convolutional Neural Network (CNN) in combination with Long Short-term Memory (LSTM) methods with different layers and compare their outputs towards the accuracy in the HAR system. We compare the accuracy of different HAR methods and observed that the performance of our proposed model of CNN 2 layers with LSTM 1 layer is the best.
|
| first_indexed | 2024-04-12T17:48:43Z |
| format | Article |
| id | doaj.art-09f63083700e4dcea65560a5b3e2d946 |
| institution | Directory Open Access Journal |
| issn | 2289-7771 |
| language | English |
| last_indexed | 2024-04-12T17:48:43Z |
| publishDate | 2022-10-01 |
| publisher | UNIMAS Publisher |
| record_format | Article |
| series | Journal of Applied Science & Process Engineering |
| spelling | doaj.art-09f63083700e4dcea65560a5b3e2d9462022-12-22T03:22:34ZengUNIMAS PublisherJournal of Applied Science & Process Engineering2289-77712022-10-019210.33736/jaspe.4639.2022Performance Analysis of Deep Learning based Human Activity Recognition Methods Mst. Farzana Aktter0Md Anwar Hossain1Sohag Sarker2AFM Zainul Abadin3Mirza AFM Rashidul Hasan4Pabna University of Science and Technology, BangladeshPabna University of Science and Technology, BangladeshPabna University of Science and Technology, BangladeshPabna University of Science and Technology, BangladeshUniversity of Rajshahi, Bangladesh Human Activity Recognition (HAR) is one of the most important branches of human-centered research activities. Along with the development of artificial intelligence, deep learning techniques have gained remarkable success in computer vision. In recent years, there is a growing interest in Human Activity Recognition systems applied in healthcare, security surveillance, and human motion-based activities. A HAR system is essentially made of a wearable device equipped with a set of sensors (like accelerometers, gyroscopes, magnetometers, heart-rate sensors, etc.). Different methods are being applied for improving the accuracy and performance of the HAR system. In this paper, we implement Artificial Neural Network (ANN), and Convolutional Neural Network (CNN) in combination with Long Short-term Memory (LSTM) methods with different layers and compare their outputs towards the accuracy in the HAR system. We compare the accuracy of different HAR methods and observed that the performance of our proposed model of CNN 2 layers with LSTM 1 layer is the best. https://publisher.unimas.my/ojs/index.php/JASPE/article/view/4639Human Activity RecognitionArtificial Neural NetworkConvolutional Neural NetworkLong Short-term Memory |
| spellingShingle | Mst. Farzana Aktter Md Anwar Hossain Sohag Sarker AFM Zainul Abadin Mirza AFM Rashidul Hasan Performance Analysis of Deep Learning based Human Activity Recognition Methods Journal of Applied Science & Process Engineering Human Activity Recognition Artificial Neural Network Convolutional Neural Network Long Short-term Memory |
| title | Performance Analysis of Deep Learning based Human Activity Recognition Methods |
| title_full | Performance Analysis of Deep Learning based Human Activity Recognition Methods |
| title_fullStr | Performance Analysis of Deep Learning based Human Activity Recognition Methods |
| title_full_unstemmed | Performance Analysis of Deep Learning based Human Activity Recognition Methods |
| title_short | Performance Analysis of Deep Learning based Human Activity Recognition Methods |
| title_sort | performance analysis of deep learning based human activity recognition methods |
| topic | Human Activity Recognition Artificial Neural Network Convolutional Neural Network Long Short-term Memory |
| url | https://publisher.unimas.my/ojs/index.php/JASPE/article/view/4639 |
| work_keys_str_mv | AT mstfarzanaaktter performanceanalysisofdeeplearningbasedhumanactivityrecognitionmethods AT mdanwarhossain performanceanalysisofdeeplearningbasedhumanactivityrecognitionmethods AT sohagsarker performanceanalysisofdeeplearningbasedhumanactivityrecognitionmethods AT afmzainulabadin performanceanalysisofdeeplearningbasedhumanactivityrecognitionmethods AT mirzaafmrashidulhasan performanceanalysisofdeeplearningbasedhumanactivityrecognitionmethods |