Efficient Markerless Motion Classification Using Radar

Eom, Changhyeon; Han, Sooji; Chun, Sabin; Joo, Soyoung; Yoon, Jisu; Kim, Min; Park, Jongchul; Park, Sanghong

doi:10.3390/s25113293

This is an early access version, the complete PDF, HTML, and XML versions will be available soon.

Open AccessArticle

Efficient Markerless Motion Classification Using Radar

by

Changhyeon Eom

¹,

Sooji Han

²,

Sabin Chun

¹

,

Soyoung Joo

¹,

Jisu Yoon

¹,

Min Kim

³

,

Jongchul Park

^4,*

and

Sanghong Park

^5,*

¹

Department of Physical Education, Graduate School, Pukyong National University, Busan 48513, Republic of Korea

²

Industry-Academia Cooperation Foundation, Pukyong National University, Busan 48513, Republic of Korea

³

Department of Maritime ICT & Mobility Research, Korea Institute of Ocean Science & Technology, 385, Haeyang-ro, Yeongdo-gu, Busan 49111, Republic of Korea

⁴

Department of Marine Sports, Pukyong National University, Busan 48513, Republic of Korea

⁵

Department of Electronic Engineering, Pukyong National University, Busan 48513, Republic of Korea

^*

Authors to whom correspondence should be addressed.

Sensors 2025, 25(11), 3293; https://doi.org/10.3390/s25113293

Submission received: 26 March 2025 / Revised: 8 May 2025 / Accepted: 22 May 2025 / Published: 23 May 2025

(This article belongs to the Section Radar Sensors)

Download Versions Notes

Abstract

This study proposes a novel method that uses radar for markerless motion classification by using effective features derived from micro-Doppler signals. The training phase uses three-dimensional marker coordinates captured by a motion-capture system to construct basis functions, which enable modeling of micro-motions of the human body. During the testing phase, motion classification is performed without markers, relying solely on radar signals. The feature vectors are generated by applying cross-correlation between the received radar signal and the basis functions, then compressed using principal component analysis, and classified using a simple nearest-neighbor algorithm. The proposed method achieves nearly 100% classification accuracy with a compact feature set and is accurate even at high signal-to-noise ratios. Experimental results demonstrate that to optimize training data and increase computational efficiency, the sampling duration and sampling interval must be set appropriately.

Keywords: 3-dimensional marker position; PCA; micro-Doppler; human body

Share and Cite

MDPI and ACS Style

Eom, C.; Han, S.; Chun, S.; Joo, S.; Yoon, J.; Kim, M.; Park, J.; Park, S. Efficient Markerless Motion Classification Using Radar. Sensors 2025, 25, 3293. https://doi.org/10.3390/s25113293

AMA Style

Eom C, Han S, Chun S, Joo S, Yoon J, Kim M, Park J, Park S. Efficient Markerless Motion Classification Using Radar. Sensors. 2025; 25(11):3293. https://doi.org/10.3390/s25113293

Chicago/Turabian Style

Eom, Changhyeon, Sooji Han, Sabin Chun, Soyoung Joo, Jisu Yoon, Min Kim, Jongchul Park, and Sanghong Park. 2025. "Efficient Markerless Motion Classification Using Radar" Sensors 25, no. 11: 3293. https://doi.org/10.3390/s25113293

APA Style

Eom, C., Han, S., Chun, S., Joo, S., Yoon, J., Kim, M., Park, J., & Park, S. (2025). Efficient Markerless Motion Classification Using Radar. Sensors, 25(11), 3293. https://doi.org/10.3390/s25113293

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Menu

Efficient Markerless Motion Classification Using Radar

Abstract

Share and Cite

Article Metrics

Article Access Statistics

Further Information

Guidelines

MDPI Initiatives

Follow MDPI