Hand Gesture Recognition Using Ultra-Wideband Radar with Random Forest

Yao Li; Xin Wang; Wentao He; Baodai Shi

doi:10.3788/LOP202158.2207002

Laser & Optoelectronics Progress, Volume. 58, Issue 22, 2207002(2021)

Hand Gesture Recognition Using Ultra-Wideband Radar with Random Forest

Yao Li^*, Xin Wang, Wentao He, and Baodai Shi

Tracking Guidance Teaching and Research Section, Air Defense and Missile Defense College, Air Force Engineering University, Xi’an, Shaanxi 710051, China

show less

Abstract Get PDF(in Chinese)

Figures & Tables(14)

Fig. 1. Flow chart of echo signal processing

Download full size

Fig. 2. Comparison of time domain and frequency domain among different gestures. (a) Gesture 1; (b) gesture 2; (c) gesture 3; (d) gesture 4

Download full size

Fig. 3. Flowchart of system working

Download full size

Fig. 4. Feature map clustering analysis

Download full size

Fig. 5. Clustering results for different K values. (a) Original image; (b) K=2; (c) K =4; (d) K =8

Download full size

Fig. 6. Visualization of clustering results

Download full size

Fig. 7. Random forest visualization

Download full size

Fig. 8. Contribution rate of some features

Download full size

Fig. 9. SNR of maps with different channel numbers

Download full size

Table 1. Comparison of recognition accuracy for different number of accumulated echo pulses
View table
Table 1. Comparison of recognition accuracy for different number of accumulated echo pulses
Number of accumulated echo pulses 48 96 128 256
Average accuracy of classification /% 67.23 81.46 98.67 99.97

Table 2. Comparison of accuracy for different feature maps
View table
Table 2. Comparison of accuracy for different feature maps
Type of feature map Average accuracy of classification /%
RGB map 98.67
R channel map 77.98
G channel map 81.25
B channel map 78.25
Gray-scale map 83.54

Table 3. Results of classification for different parameter settings of random forest
View table
Table 3. Results of classification for different parameter settings of random forest
Type of input CART number Maximum number of features Average accuracy of classification /%
RDM 50 6 79.26
12 85.32
20 86.17
100 6 87.34
12 98.67
20 99.12
200 6 86.54
12 97.12
20 98.31

Table 4. Setting of parameters in gesture recognition algorithm

View table

Table 4. Setting of parameters in gesture recognition algorithm

Parameter	Setting
Number of transmitting antennas	1
Number of receiving antennas	1
Number of integrated radar pulses	128
RDM frame duration /ms	14
Number of sampling points	8000
Type of input	RDM
Size of input	512×512×3
CART number	100
Maximum number of features	12
Number of gesture classes	6

Table 5. Comparison of proposed algorithm with other algorithms
View table
Table 5. Comparison of proposed algorithm with other algorithms
Algorithm Training time /h Recognition speed /(frame·s^-1) Accuracy /%
ShuffleNet V2 4.46 41 90.64
Mobilenet V2 4.52 42 91.53
VGG 16 4.21 46 86.63
Algorithm proposed in this article 2.61 41 98.93

Tools

Get Citation

Copy Citation Text

Yao Li, Xin Wang, Wentao He, Baodai Shi. Hand Gesture Recognition Using Ultra-Wideband Radar with Random Forest[J]. Laser & Optoelectronics Progress, 2021, 58(22): 2207002

Download Citation

EndNote(RIS)BibTex Plain Text

Set citation alerts for article

Save article for my favorites