SAR image target detection method based on multi-source feature migration and false alarm elimination

Abstract

The invention discloses an SAR image target detection method based on multi-source feature migration and false alarm rejection, and the method comprises the steps: obtaining a to-be-detected image, and carrying out the detection of the to-be-detected image according to a preset false alarm rate and a two-parameter constant false alarm rate algorithm, and obtaining a plurality of target pixel point sets; the to-be-detected image is a synthetic aperture radar (SAR) image; inputting each target pixel point set into a region sensing model to enable the region sensing model to detect the target pixel point sets; rejecting or retaining a target pixel point set according to a detection result, and taking the retained target pixel point set as a target detection result; wherein the region sensing model is a pre-trained neural network model. According to the method, preliminary target detection is performed on the to-be-detected image through the two-parameter constant false alarm rate algorithm, and false alarm elimination is performed by using the region sensing model, so that the calculation cost can be reduced, the problem of relatively high false alarm rate of a detection result is solved, and target detection of the pixel level of the SAR image is also realized.

Description

technical field [0001] The invention belongs to the technical field of image processing, and in particular relates to a SAR image target detection method based on multi-source feature migration and false alarm elimination. Background technique [0002] In related technologies, the target detection research on SAR images is mainly carried out by using the statistical characteristics of image gray values. The constant false alarm rate (Constant false alarm rate, CFAR) detection algorithm is a classic SAR image ship target detection algorithm. , the algorithm uses the statistical law of image gray space to distinguish the target from the background clutter through the preset false alarm rate and the given clutter distribution model. [0003] With the successful application of deep learning theory to the field of target detection in photoelectric images, those skilled in the art began to try to use deep learning target detection algorithms on SAR images. When using deep learnin...

AI Technical Summary

Problems solved by technology

For traditional transfer learning, a source domain is usually used to provide auxiliary training data for the training of the target domain model. It may not be able to provide effective assistance to the model training of the target domain, and sometimes even affect the performance of the target domain model, and the hierarchical parameters of the "freeze-adjustment" of the neural network rely on empirical settings, which makes the transfer learning effect of a single source domain Unstable, affecting the accuracy of target detection in SAR images

Images