GLBRF: Group-Based Lightweight Human Behavior Recognition Framework in Video Camera
Behavioral recognition is an important technique for recognizing actions by analyzing human behavior. It is used in various fields, such as anomaly detection and health estimation. For this purpose, deep learning models are used to recognize and classify the features and patterns of each behavior. H...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2024-03-01
|
| Series: | Applied Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2076-3417/14/6/2424 |
| _version_ | 1797242226513805312 |
|---|---|
| author | Young-Chan Lee So-Yeon Lee Byeongchang Kim Dae-Young Kim |
| author_facet | Young-Chan Lee So-Yeon Lee Byeongchang Kim Dae-Young Kim |
| author_sort | Young-Chan Lee |
| collection | DOAJ |
| description | Behavioral recognition is an important technique for recognizing actions by analyzing human behavior. It is used in various fields, such as anomaly detection and health estimation. For this purpose, deep learning models are used to recognize and classify the features and patterns of each behavior. However, video-based behavior recognition models require a lot of computational power as they are trained using large datasets. Therefore, there is a need for a lightweight learning framework that can efficiently recognize various behaviors. In this paper, we propose a group-based lightweight human behavior recognition framework (GLBRF) that achieves both low computational burden and high accuracy in video-based behavior recognition. The GLBRF system utilizes a relatively small dataset to reduce computational cost using a 2D CNN model and improves behavior recognition accuracy by applying location-based grouping to recognize interaction behaviors between people. This enables efficient recognition of multiple behaviors in various services. With grouping, the accuracy was as high as 98%, while without grouping, the accuracy was relatively low at 68%. |
| first_indexed | 2024-04-24T18:35:51Z |
| format | Article |
| id | doaj.art-9de8467dfdc74123af8335e61f4f7122 |
| institution | Directory Open Access Journal |
| issn | 2076-3417 |
| language | English |
| last_indexed | 2024-04-24T18:35:51Z |
| publishDate | 2024-03-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Applied Sciences |
| spelling | doaj.art-9de8467dfdc74123af8335e61f4f71222024-03-27T13:19:40ZengMDPI AGApplied Sciences2076-34172024-03-01146242410.3390/app14062424GLBRF: Group-Based Lightweight Human Behavior Recognition Framework in Video CameraYoung-Chan Lee0So-Yeon Lee1Byeongchang Kim2Dae-Young Kim3Department of Computer Software, Daegu Catholic University, Gyeongsan 38430, Republic of KoreaDepartment of Software Convergence, Soonchunhyang University, Asan 31538, Republic of KoreaSchool of Computer Software, Daegu Catholic University, Gyeongsan 38430, Republic of KoreaDepartment of Computer Software Engineering, Soonchunhyang University, Asan 31538, Republic of KoreaBehavioral recognition is an important technique for recognizing actions by analyzing human behavior. It is used in various fields, such as anomaly detection and health estimation. For this purpose, deep learning models are used to recognize and classify the features and patterns of each behavior. However, video-based behavior recognition models require a lot of computational power as they are trained using large datasets. Therefore, there is a need for a lightweight learning framework that can efficiently recognize various behaviors. In this paper, we propose a group-based lightweight human behavior recognition framework (GLBRF) that achieves both low computational burden and high accuracy in video-based behavior recognition. The GLBRF system utilizes a relatively small dataset to reduce computational cost using a 2D CNN model and improves behavior recognition accuracy by applying location-based grouping to recognize interaction behaviors between people. This enables efficient recognition of multiple behaviors in various services. With grouping, the accuracy was as high as 98%, while without grouping, the accuracy was relatively low at 68%.https://www.mdpi.com/2076-3417/14/6/2424behavior recognitiondeep learningvideo-basedlightweight learning frameworklocation-based grouping |
| spellingShingle | Young-Chan Lee So-Yeon Lee Byeongchang Kim Dae-Young Kim GLBRF: Group-Based Lightweight Human Behavior Recognition Framework in Video Camera Applied Sciences behavior recognition deep learning video-based lightweight learning framework location-based grouping |
| title | GLBRF: Group-Based Lightweight Human Behavior Recognition Framework in Video Camera |
| title_full | GLBRF: Group-Based Lightweight Human Behavior Recognition Framework in Video Camera |
| title_fullStr | GLBRF: Group-Based Lightweight Human Behavior Recognition Framework in Video Camera |
| title_full_unstemmed | GLBRF: Group-Based Lightweight Human Behavior Recognition Framework in Video Camera |
| title_short | GLBRF: Group-Based Lightweight Human Behavior Recognition Framework in Video Camera |
| title_sort | glbrf group based lightweight human behavior recognition framework in video camera |
| topic | behavior recognition deep learning video-based lightweight learning framework location-based grouping |
| url | https://www.mdpi.com/2076-3417/14/6/2424 |
| work_keys_str_mv | AT youngchanlee glbrfgroupbasedlightweighthumanbehaviorrecognitionframeworkinvideocamera AT soyeonlee glbrfgroupbasedlightweighthumanbehaviorrecognitionframeworkinvideocamera AT byeongchangkim glbrfgroupbasedlightweighthumanbehaviorrecognitionframeworkinvideocamera AT daeyoungkim glbrfgroupbasedlightweighthumanbehaviorrecognitionframeworkinvideocamera |