Ground Target Classification Method Based on Random Forest and Data Rejection

Abstract

The invention discloses a ground target classification method based on random forest and data rejection. The realization steps are: (1) preprocessing the training sample set; (2) extracting the training feature matrix; (3) training the random forest classification (4) preprocess the test sample; (5) extract the test feature vector; (6) calculate the ground target output probability vector; (7) judge whether the test sample is rejected; (8) if rejected, test The sample is used as an echo signal without target micro-motion characteristics; (9) If the judgment is not rejected, the category corresponding to the maximum value in the output probability vector is used as the ground target classification result of the test sample. The invention rejects clutter, deceptive interference and target signals without micro-motion, improves the classification and recognition rate of ground moving targets, and adopts a classifier with parallel processing capability to improve the real-time performance of the method.

Description

technical field [0001] The invention belongs to the technical field of communication, and further relates to a ground target classification method based on random forest and data rejection in the technical field of radar signal processing. The invention can classify different vehicle targets and human targets moving on the ground in real time in the environment with clutter and deceptive interference. Background technique [0002] In radar ground target classification, radar returns may contain clutter, deceptive jamming, and non-jittering ground target signals. For clutter and deceptive interference, the classification method should refuse to classify; and for the target signal without micro-motion, because the effective micro-motion features required for classification cannot be extracted, the classification method cannot effectively and correctly classify. How to reasonably and effectively remove clutter, deceptive interference and target signals without micro-motion is ...

AI Technical Summary

Problems solved by technology

Although this method can process radar signals with high signal-to-noise ratio and identify tracked vehicles and wheeled vehicles, there are still shortcomings in this method: since this method only directly compares the observed Doppler spectrum with the wheeled vehicle Doppler spectrum template, tracked vehicle Doppler spectrum template distance, so in the environment with clutter and deceptive interference, clutter and deceptive interference will still be identified as vehicle targets; It is randomly classified, which causes misjudgment of the classification method and reduces the recognition rate

Although this method can process and classify radar signals of ground targets, there are still shortcomings in this method: since this method uses a support vector machine classifier, the support vector machine itself is a classifier with a serial structure, so it supports Vector machine classifiers require more processing time, which leads to poor real-time performance of the method

Images