Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2602.21128 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866917292741230592 |
|---|---|
| author | Trinh, Huy Liu, Davis Humaira, Munia Lee, Peter Wang, Zhou |
| author_facet | Trinh, Huy Liu, Davis Humaira, Munia Lee, Peter Wang, Zhou |
| contents | Radar-based human activity recognition has gained attention as a privacy-preserving alternative to vision and wearable sensors, especially in sensitive environments like long-term care facilities. Micro-Doppler spectrograms derived from FMCW radar signals are central to recognizing dynamic activities, but their effectiveness is limited by noise and clutter. In this work, we use a benchmark radar dataset to reimplement and assess three recent denoising and preprocessing techniques: adaptive preprocessing, adaptive thresholding, and entropy-based denoising. To illustrate the shortcomings of conventional metrics in low-SNR regimes, we evaluate performance using both perceptual image quality measures and standard error-based metrics. We additionally propose a novel framework for static activity recognition using range-angle feature maps to expand HAR beyond dynamic activities. We present two important contributions: a temporal tracking algorithm to enforce consistency and a no-reference quality scoring algorithm to assess RA-map fidelity. According to experimental findings, our suggested techniques enhance classification performance and interpretability for both dynamic and static activities, opening the door for more reliable radar-based HAR systems. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_21128 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Vision-Inspired Image Quality Assessment for Radar-Based Human Activity Representations Trinh, Huy Liu, Davis Humaira, Munia Lee, Peter Wang, Zhou Image and Video Processing Radar-based human activity recognition has gained attention as a privacy-preserving alternative to vision and wearable sensors, especially in sensitive environments like long-term care facilities. Micro-Doppler spectrograms derived from FMCW radar signals are central to recognizing dynamic activities, but their effectiveness is limited by noise and clutter. In this work, we use a benchmark radar dataset to reimplement and assess three recent denoising and preprocessing techniques: adaptive preprocessing, adaptive thresholding, and entropy-based denoising. To illustrate the shortcomings of conventional metrics in low-SNR regimes, we evaluate performance using both perceptual image quality measures and standard error-based metrics. We additionally propose a novel framework for static activity recognition using range-angle feature maps to expand HAR beyond dynamic activities. We present two important contributions: a temporal tracking algorithm to enforce consistency and a no-reference quality scoring algorithm to assess RA-map fidelity. According to experimental findings, our suggested techniques enhance classification performance and interpretability for both dynamic and static activities, opening the door for more reliable radar-based HAR systems. |
| title | Vision-Inspired Image Quality Assessment for Radar-Based Human Activity Representations |
| topic | Image and Video Processing |
| url | https://arxiv.org/abs/2602.21128 |