Visual effects based feature extraction for dynamic radar target echo series

Feature extraction is of great importance in radar target recognition. The extraction of effective and reliable features is fundamental and critical for the construction of target recognition systems. It is difficult to build target scattering models rigorously for regular low-resolution radars abiding by electromagnetic theories. Based on the observation and analysis on a large number of real-life data, and taking the thinking process and the description method of radar operators for reference, a set of feature extraction methods basing on visual effects for dynamic radar target echo series have been introduced. The application of these methods in multiple radar target recognition projects has testified their effectiveness and robustness.