Textural feature extraction and classification for humans and vehicles

Extraction of features and subsequent classification of ground moving targets, especially humans and vehicles, are topics of great relevance for the theoretical research and practical application in the signal processing of the ground surveillance radar. Based on time-frequency (TF) spectrograms, a novel method for the feature extraction of micro-motion targets is proposed in this paper. The experimental results under different signal-to-noise ratio (SNR) and training sample number conditions verify the validity and robustness of the proposed method.