Micro-Doppler radar classification of humans and animals in an operational environment
作者：
Highlights：
• A Gaussian mixture Hidden Markov model is used to classify ground moving targets.
• An improvement in performance over popular state-of-the-art method is demonstrated.
• Targets can be accurately classified at ranges exceeding 1 km.
• The model can distinguish between different classes of animals and humans.
• The accuracy ranges from 75% to 90% for times on target up to 1.25 s.
摘要
•A Gaussian mixture Hidden Markov model is used to classify ground moving targets.•An improvement in performance over popular state-of-the-art method is demonstrated.•Targets can be accurately classified at ranges exceeding 1 km.•The model can distinguish between different classes of animals and humans.•The accuracy ranges from 75% to 90% for times on target up to 1.25 s.
论文关键词：Human classification,Doppler,Animal classification,Hidden Markov model,Gaussian Mixture Model
论文评审过程：Received 25 October 2017, Revised 19 January 2018, Accepted 8 February 2018, Available online 9 February 2018, Version of Record 19 March 2018.
论文官网地址：https://doi.org/10.1016/j.eswa.2018.02.019
