Motion Classification Using Kinematically Sifted ACGAN-Synthesized Radar Micro-Doppler Signatures

Erol, Barış; Gürbüz, Sevgi Zübeyde; Amin, Moeness G.

doi:10.1109/taes.2020.2969579

Cited by 78 publications

(31 citation statements)

References 40 publications

(41 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To maximize the potential of our data augmentation, a selection criteria needs to be in place for filtering out deviant samples. 30 Towards this aim, principal component analysis (PCA) is used to extract the feature space for each signature.…”

Section: Data Synthesis With Auxilliary Conditional Gans (Acgans)mentioning

confidence: 99%

“…An alternative approach for generating synthetic data, which can potentially bridge this deficiency, is adversarial learning. [27][28][29] Recent studies 27,30 show that auxiliary conditional GANs (ACGANs) are effective in modeling both sensor imperfections and clutter in a through-the-wall sensing scenario. However, the main challenge in application of GANs has been the lack of kinematic fidelity in a significant percentage of signatures generated.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Cross-frequency training with adversarial learning for radar micro-Doppler signature classification (Rising Researcher)

Gürbüz

Rahman

Kurtoğlu

et al. 2020

Radar Sensor Technology XXIV

View full text Add to dashboard Cite

Deep neural networks have become increasingly popular in radar micro-Doppler classification; yet, a key challenge, which has limited potential gains, is the lack of large amounts of measured data that can facilitate the design of deeper networks with greater robustness and performance. Several approaches have been proposed in the literature to address this problem, such as unsupervised pre-training and transfer learning from optical imagery or synthetic RF data. This work investigates an alternative approach to training which involves exploitation of "datasets of opportunity"-micro-Doppler datasets collected using other RF sensors, which may be of a different frequency, bandwidth or waveform-for the purposes of training. Specifically, this work compares in detail the cross-frequency training degradation incurred for several different training approaches and deep neural network (DNN) architectures. Results show a 70% drop in classification accuracy when the RF sensors for pre-training, fine-tuning, and testing are different, and a 15% degradation when only the pre-training data is different, but the fine-tuning and test data are from the same sensor. By using generative adversarial networks (GANs), a large amount of synthetic data is generated for pre-training. Results show that cross-frequency performance degradation is reduced by 50% when kinematically-sifted GAN-synthesized signatures are used in pre-training.

show abstract

Section: Data Synthesis With Auxilliary Conditional Gans (Acgans)mentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Cross-frequency training with adversarial learning for radar micro-Doppler signature classification (Rising Researcher)

Gürbüz

Rahman

Kurtoğlu

et al. 2020

Radar Sensor Technology XXIV

View full text Add to dashboard Cite

show abstract

“…The work in [23] used GANs to generate synthetic radar signatures for walking gaits at different speed, and [24] applied a similar approach to data of six human actions, including movements other than simply walking. Notably, the work in [25] proposed a novel approach to use the adversarial learning of GANs combined with a PCA-based (Principal Component Analysis) kinematic sifting approach to reject the synthetic radar samples that present unrealistic data, i.e. data with artefacts that would not be realistically present in experimental data.…”

Section: Introductionmentioning

confidence: 99%

Continuous Human Activity Classification From FMCW Radar With Bi-LSTM Networks

et al. 2020

View full text Add to dashboard Cite

Recognition of human movements with radar for ambient activity monitoring is a developed area of research that yet presents outstanding challenges to address. In real environments, activities and movements are performed with seamless motion, with continuous transitions between activities of different duration and a large range of dynamic motions, compared with discrete activities of fixed-time lengths which are typically analysed in the literature. This paper proposes a novel approach based on recurrent LSTM and Bi-LSTM network architectures for continuous activity monitoring and classification. This approach uses radar data in the form of a continuous temporal sequence of micro-Doppler or range-time information, differently from from other conventional approaches based on convolutional networks that interpret the radar data as images. Experimental radar data involving 15 participants and different sequences of 6 actions are used to validate the proposed approach. It is demonstrated that using the Dopplerdomain data together with the Bi-LSTM network and an optimal learning rate can achieve over 90% mean accuracy, whereas range-domain data only achieved approximately 76%. The details of the network architectures, insights in their behaviour as a function of key hyper-parameters such as the learning rate, and a discussion on their performance across are provided in the paper.

show abstract

“…The first category is to build classifiers robust to limited training data, such as the models in [10]- [12]. The second category is labeled data augmentation with synthetic data [13]- [15]. Transfer learning, which can take advantage of prior knowledge from an existing large-scale data set (source domain) as a supplement for the tasks on a different but related small-scale data set (target domain), is the third category.…”

Section: Introductionmentioning

confidence: 99%

Human Motion Recognition With Limited Radar Micro-Doppler Signatures

Fioranelli

et al. 2021

IEEE Trans. Geosci. Remote Sensing

View full text Add to dashboard Cite

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.

show abstract

Motion Classification Using Kinematically Sifted ACGAN-Synthesized Radar Micro-Doppler Signatures

Cited by 78 publications

References 40 publications

Cross-frequency training with adversarial learning for radar micro-Doppler signature classification (Rising Researcher)

Cross-frequency training with adversarial learning for radar micro-Doppler signature classification (Rising Researcher)

Continuous Human Activity Classification From FMCW Radar With Bi-LSTM Networks

Human Motion Recognition With Limited Radar Micro-Doppler Signatures

Contact Info

Product

Resources

About