Automatic Arm Motion Recognition Based on Radar Micro-Doppler Signature Envelopes

Abstract

In considering human-machine interface (HMI) for smart environment, a simple but effective method is proposed for automatic arm motion recognition with a Doppler radar sensor. Arms, in lieu of hands, have stronger radar cross-section and can be recognized from relatively longer distances. An energy-based thresholding algorithm is applied to the spectrograms to extract the micro-Doppler (MD) signature envelopes. The positive and negative frequency envelopes are concatenated to form a feature vector. The nearest neighbor (NN) classifier with Manhattan distance (L1) is then used to recognize the arm motions. It is shown that this simple method yields classification accuracy above 97 percent for six classes of arm motions. Despite its simplicity, the proposed method is superior to those of handcrafted featurebased classifications and low-dimension representation techniques based on principal component analysis (PCA), and is comparable to convolutional neural network (CNN).

Arm Motion Classification Using Curve Matching of Maximum Instantaneous Doppler Frequency Signatures

“…It is clear that the envelopes, representing the maximum instantaneous Doppler frequencies, can well enclose the local power distributions. It is also evident that the MD characteristics of the spectrograms are in agreement and consistent with each arm motion kinematics [16].…”

“…To further examine the impact of downsampling on the NN classifier, the downsampled features are put into the NN classifier with L1 distance. This resulted in classification accuracy of 97.13% [16], which is nearly the same as when using the entire sequence. The confusion matrix is given in Table I.…”

“…Fig. 6 shows the "attention" motion with the normal speed at different In our previous work [16,17], each of the original extracted envelope feature contained 2000 samples for both positive and negative Doppler frequencies, and was directly fed into the NN classifier with L1 distance measure, achieving an overall accuracy 97.17% [16]. To avoid high computations of DTW dealing with long time series, we downsample the envelopes to 200 samples.…”

“…Recent work in automatic arm motion recognition using the maximum instantaneous Doppler frequencies, i.e., the envelope of the MD signature of the data spectrogram, as features followed by the NN classifier has provided classification rates reaching close to 97% [16]. In this work, the feature vector consists of the augmented positive frequency and negative frequency envelopes.…”