Human Motion Recognition With Limited Radar Micro-Doppler Signatures

Abstract

The performance of deep learning (DL) algorithms for radar-based human motion recognition (HMR) is hindered by the diversity and volume of the available training data. In this article, to tackle the issue of insufficient training data for HMR, we propose an instance-based transfer learning (ITL) method with limited radar micro-Doppler (MD) signatures, alleviating the burden of collecting and annotating a large number of radar samples. ITL is a unique algorithm that consists of three interconnected parts, including DL model pretraining, correlated source data selection, and adaptive collaborative fine-tuning (FT). Any of the three components cannot be excluded; otherwise, the performance of the entire algorithm decreases. The experiments with a radar data set of six human motions show that ITL achieves state-of-the-art performance for HMR with limited training samples, outperforming several existing transfer learning approaches. Especially, when there are only 100 samples per person per class, ITL yields an F1 score of 96.7%. Last but not least, ITL is more generalized to human motion differences. Though adapted to recognize the persons' motions in a small-scale target data set, ITL can also classify the persons' motion data used for pretraining, achieving up to 11.0% F1 score enhancement over the conventional FT method.

Experimental Verification of Micro-Doppler Radar Measurements of Fall-Risk-Related Gait Differences for Community-Dwelling Elderly Adults

“…The MDR system was composed of two antennas and transmitted a continuous sinusoidal wave with a frequency of f 0 = 24.0 GHz to a walking participant. The signals received after synchronous detection s(t) were composed of the Doppler frequencies f d corresponding to the Doppler velocities v d = cf d /(2f 0 ) (c: speed of light) of the scattering centers on the body parts, such as the legs and torso [18,19] (i.e., the Fourier transform of s(t) is expressed as S(f ) = ∑A i δ(f −f di ) where δ(f ) is Dirac's delta function and A i and f di are received amplitude and Doppler frequency of i-th scattering center). We collected the data of the participants in the measurement area shown in Figure 1a.…”

“…Micro-Doppler radar (MDR) is a promising solution to these problems. MDR can remotely measure the velocity of entire human body parts without placing any constraints on participants [18][19][20]. There are no limitations of lighting conditions and clothing.…”

Multi-Feature Transformer-Based Learning for Continuous Human Motion Recognition with High Similarity Using mmWave FMCW Radar

“…To alleviate the burden of collecting and annotating a large number of radar samples, Li et al. [ 8 ] proposed an HMR with an instance-based transfer learning (ITL) method under the limited radar micro-Doppler (MD) signatures. Sakagami et al.…”

“…Machine learning or deep learning-based human motion recognition results recently are widely adopted to improve the recognition accuracy. For instance, the recognition of a human body’s gait using Convolutional Neural Networks (CNN) is investigated in [ 8 , 9 ].…”

Multiple Stationary Human Targets Detection in Through-Wall UWB Radar Based on Convolutional Neural Network

“…Recently, deep learning technology has gained huge advances in the remote sensing community due to its powerful feature extraction capabilities [37][38][39][40][41]. Convolutional neural network (CNN) is originally designed to deal with the image-format datasets, which can also be adjusted to handle radar signals.…”