A vehicle target identification method based on physical feature extraction and SVM

Abstract

The invention discloses a vehicle target identification method based on physical feature extraction and SVM (Support Vector Machine). The vehicle target identification method comprises the following implementation steps of (1) inputting radar echo data; (2) carrying out the echo data imaging; (3) extracting an image maximum value region; (4) constructing a position set and an attribute set; (5) constructing a position dictionary and an attribute dictionary; (6) constructing an optimal value set of scattering center parameters; (7) estimating a frequency dependence factor of the scattering center; (8) obtaining a fusion frequency dependence factor; (9) obtaining fusion polarization parameters; (10) acquiring a scattering center feature vector; (11) extracting a vehicle target feature vector; and (12) predicting the type of the test sample. The method can reflect the physical structure of the vehicle target, has clear physical significance and robust scattering center type distribution law characteristics, obtains high and stable classification accuracy, and can be used for radar target recognition.

Description

technical field [0001] The invention belongs to the technical field of image processing, and further relates to a physical feature extraction and a vehicle target recognition method of a Support Vector Machine (SVM) in the technical field of image classification. The invention can be used to classify and recognize the vehicle target in the synthetic aperture radar SAR (Synthetic Aperture Radar) image. Background technique [0002] In the vehicle target recognition process based on synthetic aperture radar SAR images, the two most important steps are feature extraction and recognition. The characteristics of the target obtained by the traditional synthetic aperture radar SAR image vehicle target recognition algorithm through the learning algorithm depend on the target synthetic aperture radar SAR image, and these features do not have a clear physical meaning and will be affected by factors such as the target, environmental changes, and imaging. impact, making the performance...

AI Technical Summary

Problems solved by technology

The disadvantage of this method is that the features extracted based on the intra-class scatter matrix and inter-class scatter matrix reflect the statistical characteristics of the target, but cannot reflect the physical characteristics of the target such as structure and outline, and do not have clear physical meaning. Lack of robustness, recognition accuracy is easily affected by factors such as targets, environmental changes, and imaging

The disadvantage of this method is that the training data and the test data are required to be precisely matched. When the training data and the test data do not match, the recognition accuracy is low.

Images