In this letter, we propose an empirical-mode decomposition (EMD)-based method for automatic multicategory mini-unmanned aerial vehicle (UAV) classification. The radar echo signal is first decomposed into a set of oscillating waveforms by EMD. Then, eight statistical and geometrical features are extracted from the oscillating waveforms to capture the phenomenon of blade flashes. After feature normalization and fusion, a nonlinear support vector machine is trained for target class-label prediction. Our empirical results on real measurement of radar signals show encouraging mini-UAV classification accuracy performance.

In this paper, an unmanned aerial vehicles detection system is proposed for an X-band ground-based surveillance FMCW Doppler radar. Two novel features are directly extracted from the Doppler processing results without a time-frequency analysis. Experimental results on measured radar echo signals show that our system consistently outperforms the state-of-thearts in terms of detection accuracy and computational efficiency

In this Correspondence we propose to use an Iterative Adaptive Approach (IAA) to perform the Doppler processing in ground-based surveillance radar for drone detection. It overcomes the limited dwell time and improves the Doppler resolution, and therefore significantly enhances the discrimination of micro-Doppler signature and the correct classification of drones. In addition, the target detectability is also improved. This approach is validated in the field experiments conducted with a commercial portable ground-based surveillance radar.

In this paper, we propose a novel cross-spectral matching system for identity verification based on the palm-vein and the palmprint acquired from the visible (RGB) and the near infrared (NIR) image spectral bands. Considering the vast availability of the visible library, the red and the blue spectrums are treated as sources of gallery samples and the NIR spectral band is utilized as the probe source without loss of generality. Apart from the extraction of palm-vein and palmprint features, the discriminative power of the palmprint templates is enhanced using a simplified Local Binary Pattern (LBP) encoding scheme. The similarity scores obtained by matching the NIR palm-vein templates against the registered RGB palm-vein templates is finally fused with scores obtained from matching the NIR palmprint codes against the registered RGB palmprint codes. Our empirical results on two publicly available multi-spectral palm databases show that the proposed system consistently achieves promising verification performance.

Gabor magnitude is known to be among the most discriminative representations for face images due to its space-frequency co-localization property. However, such property causes adverse effects even when the images are acquired under moderate head pose variations. To address this pose sensitivity issue as well as other moderate imaging variations, we propose an analytic Gabor feedforward network which can absorb such moderate changes. Essentially, the network works directly on the raw face images and produces directionally projected Gabor magnitude features at the hidden layer. Subsequently, several sets of magnitude features obtained from various orientations and scales are fused at the output layer for final classification decision. The network model is analytically trained using a single sample per identity. The obtained solution is globally optimal with respect to the classification total error rate. Our empirical experiments conducted on five face datasets (six subsets) from the public domain show encouraging results in terms of identification accuracy and computational efficiency.