Feature Extraction of Micro-Motional Targets Via Time-Range Distribution

Abstract

The micro-motional period is an important feature for spatial target discrimination. Firstly, the movement and wideband echo models of wobbling targets with bistatic observations are introduced. Then, the characteristics of the mono-/bi-static high-resolution range profiles (HRRPs) with different polarizations are analyzed based on the simulation echoes of the wobbling target via the electromagnetic calculations. Finally, the observed HRRP sequence is employed to construct a time-range distribution matrix, and an estimation method of the micro-motional period is proposed based on the time-range distribution matrix correlation, which is validated by the measurement experiments in the anechoic chamber.INDEX TERMS Micro-motion, bistatic radar, time-range distribution, feature extraction.

Feature Extraction Method of Piano Performance Technique Based on Recurrent Neural Network

“…The feature extraction of piano playing techniques is based on the hand model to identify the finger direction vector, and the model is constantly verified by the data collected by two ultra-wide angle infrared sensors in real time (Ai et al, 2019); . By collecting the player's hand joint angle parameters, phalange length, palm scale, arm length and other linear parameters, the hand model is established, and the results are shown in Figure 1.…”

Three-Dimensional Absolute Attitude Reconstruction of a Rigid Body Based on Multi-Station HRRP Sequences

“…As the radar bandwidth increases, the projection range of the scattering centers can be resolved in the distance dimension. In [5]- [7], the target micro-motion parameters were estimated based on the HRRP sequences or the inverse synthetic aperture radar (ISAR) images. The basis of these algorithms is to build a reasonable micro-motion model of the target.…”