Multiscale Residual Weighted Classification Network for Human Activity Recognition in Microwave Radar
Human activity recognition by radar sensors plays an important role in healthcare and smart homes. However, labeling a large number of radar datasets is difficult and time-consuming, and it is difficult for models trained on insufficient labeled data to obtain exact classification results. In this p...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-01-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/25/1/197 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841548957162930176 |
|---|---|
| author | Yukun Gao Lin Cao Zongmin Zhao Dongfeng Wang Chong Fu Yanan Guo |
| author_facet | Yukun Gao Lin Cao Zongmin Zhao Dongfeng Wang Chong Fu Yanan Guo |
| author_sort | Yukun Gao |
| collection | DOAJ |
| description | Human activity recognition by radar sensors plays an important role in healthcare and smart homes. However, labeling a large number of radar datasets is difficult and time-consuming, and it is difficult for models trained on insufficient labeled data to obtain exact classification results. In this paper, we propose a multiscale residual weighted classification network with large-scale, medium-scale, and small-scale residual networks. Firstly, an MRW image encoder is used to extract salient feature representations from all time-Doppler images through contrastive learning. This can extract the representative vector of each image and also obtain the pre-training parameters of the MRW image encoder. During the pre-training process, large-scale residual networks, medium-scale residual networks, and small-scale residual networks are used to extract global information, texture information, and semantic information, respectively. Moreover, the time–channel weighting mechanism can allocate weights to important time and channel dimensions to achieve more effective extraction of feature information. The model parameters obtained from pre-training are frozen, and the classifier is added to the backend. Finally, the classifier is fine-tuned using a small amount of labeled data. In addition, we constructed a new dataset with eight dangerous activities. The proposed MRW-CN model was trained on this dataset and achieved a classification accuracy of 96.9%. We demonstrated that our method achieves state-of-the-art performance. The ablation analysis also demonstrated the role of multi-scale convolutional kernels and time–channel weighting mechanisms in classification. |
| format | Article |
| id | doaj-art-ed9dd4a820bc415e99c5cd72a4b9f6c8 |
| institution | Kabale University |
| issn | 1424-8220 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj-art-ed9dd4a820bc415e99c5cd72a4b9f6c82025-01-10T13:21:13ZengMDPI AGSensors1424-82202025-01-0125119710.3390/s25010197Multiscale Residual Weighted Classification Network for Human Activity Recognition in Microwave RadarYukun Gao0Lin Cao1Zongmin Zhao2Dongfeng Wang3Chong Fu4Yanan Guo5School of Information and Communication Engineering, Beijing Information Science and Technology University, Beijing 100101, ChinaSchool of Information and Communication Engineering, Beijing Information Science and Technology University, Beijing 100101, ChinaSchool of Information and Communication Engineering, Beijing Information Science and Technology University, Beijing 100101, ChinaBeijing TransMicrowave Technology Company, Beijing 100080, ChinaSchool of Computer Science and Engineering, Northeastern University, Shenyang 110169, ChinaSchool of Information and Communication Engineering, Beijing Information Science and Technology University, Beijing 100101, ChinaHuman activity recognition by radar sensors plays an important role in healthcare and smart homes. However, labeling a large number of radar datasets is difficult and time-consuming, and it is difficult for models trained on insufficient labeled data to obtain exact classification results. In this paper, we propose a multiscale residual weighted classification network with large-scale, medium-scale, and small-scale residual networks. Firstly, an MRW image encoder is used to extract salient feature representations from all time-Doppler images through contrastive learning. This can extract the representative vector of each image and also obtain the pre-training parameters of the MRW image encoder. During the pre-training process, large-scale residual networks, medium-scale residual networks, and small-scale residual networks are used to extract global information, texture information, and semantic information, respectively. Moreover, the time–channel weighting mechanism can allocate weights to important time and channel dimensions to achieve more effective extraction of feature information. The model parameters obtained from pre-training are frozen, and the classifier is added to the backend. Finally, the classifier is fine-tuned using a small amount of labeled data. In addition, we constructed a new dataset with eight dangerous activities. The proposed MRW-CN model was trained on this dataset and achieved a classification accuracy of 96.9%. We demonstrated that our method achieves state-of-the-art performance. The ablation analysis also demonstrated the role of multi-scale convolutional kernels and time–channel weighting mechanisms in classification.https://www.mdpi.com/1424-8220/25/1/197deep learning (DL)human activity recognition (HAR)contrastive learningradar micro-Doppler signaturestime-Doppler images |
| spellingShingle | Yukun Gao Lin Cao Zongmin Zhao Dongfeng Wang Chong Fu Yanan Guo Multiscale Residual Weighted Classification Network for Human Activity Recognition in Microwave Radar Sensors deep learning (DL) human activity recognition (HAR) contrastive learning radar micro-Doppler signatures time-Doppler images |
| title | Multiscale Residual Weighted Classification Network for Human Activity Recognition in Microwave Radar |
| title_full | Multiscale Residual Weighted Classification Network for Human Activity Recognition in Microwave Radar |
| title_fullStr | Multiscale Residual Weighted Classification Network for Human Activity Recognition in Microwave Radar |
| title_full_unstemmed | Multiscale Residual Weighted Classification Network for Human Activity Recognition in Microwave Radar |
| title_short | Multiscale Residual Weighted Classification Network for Human Activity Recognition in Microwave Radar |
| title_sort | multiscale residual weighted classification network for human activity recognition in microwave radar |
| topic | deep learning (DL) human activity recognition (HAR) contrastive learning radar micro-Doppler signatures time-Doppler images |
| url | https://www.mdpi.com/1424-8220/25/1/197 |
| work_keys_str_mv | AT yukungao multiscaleresidualweightedclassificationnetworkforhumanactivityrecognitioninmicrowaveradar AT lincao multiscaleresidualweightedclassificationnetworkforhumanactivityrecognitioninmicrowaveradar AT zongminzhao multiscaleresidualweightedclassificationnetworkforhumanactivityrecognitioninmicrowaveradar AT dongfengwang multiscaleresidualweightedclassificationnetworkforhumanactivityrecognitioninmicrowaveradar AT chongfu multiscaleresidualweightedclassificationnetworkforhumanactivityrecognitioninmicrowaveradar AT yananguo multiscaleresidualweightedclassificationnetworkforhumanactivityrecognitioninmicrowaveradar |