Human Similar Activity Recognition Using Millimeter-Wave Radar Based on CNN-BiLSTM and Class Activation Mapping

Human activity recognition (HAR) is an important field in the application of millimeter-wave radar. Radar-based HARs typically use Doppler signatures as primary data. However, some common similar human activities exhibit similar features that are difficult to distinguish. Therefore, the identificati...

Full description

Saved in:
Bibliographic Details
Main Authors: Xiaochuan Wu, Zengyi Ling, Xin Zhang, Zhanchao Ma, Weibo Deng
Format: Article
Language:English
Published: MDPI AG 2025-02-01
Series:Eng
Subjects:
Online Access:https://www.mdpi.com/2673-4117/6/3/44
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849341461308375040
author Xiaochuan Wu
Zengyi Ling
Xin Zhang
Zhanchao Ma
Weibo Deng
author_facet Xiaochuan Wu
Zengyi Ling
Xin Zhang
Zhanchao Ma
Weibo Deng
author_sort Xiaochuan Wu
collection DOAJ
description Human activity recognition (HAR) is an important field in the application of millimeter-wave radar. Radar-based HARs typically use Doppler signatures as primary data. However, some common similar human activities exhibit similar features that are difficult to distinguish. Therefore, the identification of similar activities is a great challenge. Given this problem, a recognition method based on convolutional neural networks (CNN), bidirectional long short-term memory (BiLSTM), and class activation mapping (CAM) is proposed in this paper. The spectrogram is formed by processing the radar echo signal. The high-dimensional features are extracted by CNN, and then the corresponding feature vectors are fed into the BiLSTM to obtain the recognition results. Finally, the class activation mapping is used to visualize the decision recognition process of the model. Based on the data of four similar activities of different people collected by mm-wave radar, the experimental results show that the recognition accuracy of the proposed model reached 94.63%. Additionally, the output results of this model have strong robustness and generalization ability. It provides a new way to improve the accuracy of human similar posture recognition.
format Article
id doaj-art-a4d0bf58f77b49ac925d90b078246cf4
institution Kabale University
issn 2673-4117
language English
publishDate 2025-02-01
publisher MDPI AG
record_format Article
series Eng
spelling doaj-art-a4d0bf58f77b49ac925d90b078246cf42025-08-20T03:43:37ZengMDPI AGEng2673-41172025-02-01634410.3390/eng6030044Human Similar Activity Recognition Using Millimeter-Wave Radar Based on CNN-BiLSTM and Class Activation MappingXiaochuan Wu0Zengyi Ling1Xin Zhang2Zhanchao Ma3Weibo Deng4School of Electronic Engineering, Harbin Institute of Technology, Harbin 150001, ChinaSchool of Information and Communication Engineering, Harbin Institute of Technology, Harbin 150001, ChinaSchool of Electronic Engineering, Harbin Institute of Technology, Harbin 150001, ChinaSchool of Information and Communication Engineering, Harbin Institute of Technology, Harbin 150001, ChinaSchool of Electronic Engineering, Harbin Institute of Technology, Harbin 150001, ChinaHuman activity recognition (HAR) is an important field in the application of millimeter-wave radar. Radar-based HARs typically use Doppler signatures as primary data. However, some common similar human activities exhibit similar features that are difficult to distinguish. Therefore, the identification of similar activities is a great challenge. Given this problem, a recognition method based on convolutional neural networks (CNN), bidirectional long short-term memory (BiLSTM), and class activation mapping (CAM) is proposed in this paper. The spectrogram is formed by processing the radar echo signal. The high-dimensional features are extracted by CNN, and then the corresponding feature vectors are fed into the BiLSTM to obtain the recognition results. Finally, the class activation mapping is used to visualize the decision recognition process of the model. Based on the data of four similar activities of different people collected by mm-wave radar, the experimental results show that the recognition accuracy of the proposed model reached 94.63%. Additionally, the output results of this model have strong robustness and generalization ability. It provides a new way to improve the accuracy of human similar posture recognition.https://www.mdpi.com/2673-4117/6/3/44mm-wave radarconvolutional neural networks (CNN)human similar activity recognition (HSAR)bidirectional long short-term memory (BiLSTM)class activation mapping (CAM)
spellingShingle Xiaochuan Wu
Zengyi Ling
Xin Zhang
Zhanchao Ma
Weibo Deng
Human Similar Activity Recognition Using Millimeter-Wave Radar Based on CNN-BiLSTM and Class Activation Mapping
Eng
mm-wave radar
convolutional neural networks (CNN)
human similar activity recognition (HSAR)
bidirectional long short-term memory (BiLSTM)
class activation mapping (CAM)
title Human Similar Activity Recognition Using Millimeter-Wave Radar Based on CNN-BiLSTM and Class Activation Mapping
title_full Human Similar Activity Recognition Using Millimeter-Wave Radar Based on CNN-BiLSTM and Class Activation Mapping
title_fullStr Human Similar Activity Recognition Using Millimeter-Wave Radar Based on CNN-BiLSTM and Class Activation Mapping
title_full_unstemmed Human Similar Activity Recognition Using Millimeter-Wave Radar Based on CNN-BiLSTM and Class Activation Mapping
title_short Human Similar Activity Recognition Using Millimeter-Wave Radar Based on CNN-BiLSTM and Class Activation Mapping
title_sort human similar activity recognition using millimeter wave radar based on cnn bilstm and class activation mapping
topic mm-wave radar
convolutional neural networks (CNN)
human similar activity recognition (HSAR)
bidirectional long short-term memory (BiLSTM)
class activation mapping (CAM)
url https://www.mdpi.com/2673-4117/6/3/44
work_keys_str_mv AT xiaochuanwu humansimilaractivityrecognitionusingmillimeterwaveradarbasedoncnnbilstmandclassactivationmapping
AT zengyiling humansimilaractivityrecognitionusingmillimeterwaveradarbasedoncnnbilstmandclassactivationmapping
AT xinzhang humansimilaractivityrecognitionusingmillimeterwaveradarbasedoncnnbilstmandclassactivationmapping
AT zhanchaoma humansimilaractivityrecognitionusingmillimeterwaveradarbasedoncnnbilstmandclassactivationmapping
AT weibodeng humansimilaractivityrecognitionusingmillimeterwaveradarbasedoncnnbilstmandclassactivationmapping