SAR?target variant recognition?method based on multi-information joint dynamic sparse representation

Abstract

The invention discloses an SAR?target variant recognition?method based on multi-information joint dynamic sparse representation. The method comprises steps: (1) a target training?dictionary with respect to?image?domain target amplitude?information represented by the formula, a shadow?training dictionary with respect to?image?domain target shadow?information represented by the formula and a?frequency domain training dictionary with respect to?frequency domain target amplitude?information represented by the formula are built with an original SAR image of a training sample as the basis, and a multi-information training dictionary D is jointed; (2) a normalized test?target vector shown in the description, a normalized test?shadow vector shown in the description and a normalized frequency domain test?target vector shown in the description are built with an SAR image of a test sample as the basis, and a multi-information test matrix Y shown in the description is obtained after jointing; (3) according to the multi-information training dictionary D and the multi-information test matrix Y, a joint sparse formula is built and a joint sparse coefficient matrix X is solved; and (4) the test sample is restructured by using the obtained joint sparse coefficient matrix X and the final classification result is obtained according to the reconstruction error?minimization principle.

Description

technical field [0001] The invention belongs to the field of radar automatic target recognition, and relates to a SAR target variant recognition method based on multi-information joint dynamic sparse representation, which is suitable for classification and recognition of targets in SAR images. Background technique [0002] Radar imaging technology was developed in the 1950s, and has been developed by leaps and bounds in the next 60 years. It has been widely used in military, agriculture, forestry, geology, ocean, disaster, mapping and many other aspects. [0003] Synthetic aperture radar (SAR) has the characteristics of all-weather, all-time, high resolution and strong penetrating power. It has become an important means of earth observation and military reconnaissance. Automatic target recognition of SAR images has attracted more and more attention. [0004] SAR image automatic target recognition methods usually adopt the three-level processing flow proposed by the Lincoln L...

AI Technical Summary

Problems solved by technology

For the identification of target variants, if only one feature or information is used, the recognition rate obtained is not high enough

For example, SRC is only based on the magnitude information of the target region in the image domain, and when the training sample does not contain the target variant, the recognition effect on the target variant is poor in the test phase.

Images