University of Twente Student Theses

Login

Measurement-based rotational hand gesture recognition using mmWave MIMO FMCW radar

Wilbrink, E.J.J. (2022) Measurement-based rotational hand gesture recognition using mmWave MIMO FMCW radar.

[img] PDF
2MB
Abstract:Contact-free gesture recognition can be used to convey commands to computers, allowing for more user-friendly Human-Computer Interactions than traditional input devices, like remote controllers. This paper proposes methods to estimate the radius and period of periodic rotational gestures measured using millimeter-wave (mmWave) radar based on the Doppler-Time Map (DTM), Range-Time Map (RTM) and a combination of the DTM and RTM. Following the literature, a 3D Fast Fourier Transform is applied to the radar data, yielding the DTM and RTM. These are then further processed using existing techniques, like subtracting the mean over time from the RTM to remove static objects and constant false alarm rate detection to find the approximate distance between the center of the gesture and the radar. The accuracy and robustness of these methods, that combine existing processing steps in a novel way, are determined using a set of measurements of various radii and periods of the hand's circular trajectory at different distances and angles with respect to the radar. Virtual Reality glasses are used to establish a ground truth for the measurements. It is found that the combined method using both the DTM and RTM is most robust for radius recognition with 4.4cm average error. Besides, the estimate for period recognition is found accurate in 83% of the scenarios for all methods.
Item Type:Essay (Bachelor)
Faculty:EEMCS: Electrical Engineering, Mathematics and Computer Science
Subject:53 electrotechnology
Programme:Electrical Engineering BSc (56953)
Link to this item:https://purl.utwente.nl/essays/92062
Export this item as:BibTeX
EndNote
HTML Citation
Reference Manager

 

Repository Staff Only: item control page