Human Activity Detection and Coarse Localization Outdoors Using Micro-Doppler Signatures

Abstract

Human activity detection outdoors is emerging as a very important research field due to its potential application in surveillance, assisted living, search and rescue, and military applications. For such applications it is important to have detailed information about the human target, for example, whether the detected target is a single person or a group of people, what activity a target is performing, and the rough location of the target. In this paper, we propose novel usage of machine learning techniques to perform subject classification, human activity classification, people counting, and coarse localization by classifying micro-Doppler signatures obtained from a low-cost and low-power radar system. Our experiments were performed outdoors. For feature extraction of micro-Doppler signatures, we applied a two-directional two-dimensional principle component analysis (2D2PCA). Our results show that by applying 2D2PCA, the accuracy results of Support Vector Machine (SVM) and knearest neighbors (kNN) classifiers were greatly improved. We also designed and implemented a Convolutional Neural Network (CNN) for the target classifications in terms of type, number, activity and coarse localization. Our CNN model obtained very high classification accuracies (97% to 100%), which are superior to the best results obtained by SVM and kNN. Finally, we investigated the effects of the frame length of the sliding window, the angle of the direction of movement, and the number of radars used on the classification performance, providing valuable guidelines for machine learning modeling and experimental setup of micro-Doppler based research and applications.

Temporal Convolutional Neural Networks for Radar Micro-Doppler Based Gait Recognition

“…More recently, some studies propose the combination of micro-Doppler data and deep learning algorithms for the gait-based human recognition [ 13 , 14 , 15 , 16 ]. However, the hierarchical structure of deep learning is more suitable to identify complicated patterns from raw data (i.e., images and signals) without any feature extraction [ 24 , 25 ]. According to this, in [ 13 ], a deep autoencoder is used to perform human gait recognition with micro-Doppler radar.…”

RAMP-CNN: A Novel Neural Network for Enhanced Automotive Radar Object Recognition

“…The proposed RAMP-CNN model achieves significant performance improvement over prior works on object recognition under parking lot, curbside, and on-road scenario, which establishes a new state-of-art baseline. In some hard cases, the radar object recognition functionality of RAMP-CNN might still be poor for supporting autonomous driving presently 22 . However, it can be further improved in the future via incorporating more preprocessing to increase spatial resolution or adopting advanced radar platform with more antennas.…”

Sequential Human Gait Classification With Distributed Radar Sensor Fusion

“…2 illustrates the spectrograms acquired from six different walking gaits. Positive and negative Doppler frequencies [33], [34] are caused by reversal in net direction (towards/away) with respect to the radar. It may be observed that some pairs of gaits (e.g.…”