Polarimetric SAR (synthetic aperture radar) image target detection method based on multipolarization features and FCN (full convolutional)-CRF (conditional random field) fusion network

Abstract

The invention discloses a polarimetric SAR (synthetic aperture radar) image target detection method based on multipolarization features and an FCN (fully convolutional network)-CRF (conditional random field) fusion network. The invention aims to solve the problem of low detection accuracy of a polarimetric SAR artificial target of the prior art. The method includes the following steps that: a polarized SAR image to be detected is inputted, and Lee filtering is performed on the polarization coherent matrix T of the polarimetric SAR image; Pauli decomposition is performed on a polarimetric scattering matrix S, so that a pixel point-based feature matrix F1 can be formed; and Yamaguchi decomposition is performed on the filtered coherent matrix T, so that a pixel-based feature matrix F2 can be formed. According to the method of the invention, the multi-polarization feature and the FCN-CRF-based fusion network are applied to the detection of a polarimetric SAR artificial target, and therefore, the detection accuracy of the polarimetric SAR artificial target can be improved; and the method can be applied to multi-target classification tasks.

Description

technical field [0001] The invention belongs to the technical field of image processing, and in particular relates to a convolution network-based polarization SAR image target detection method, which can be used for target recognition. Background technique [0002] Polarization SAR has the advantages of all-weather, all-time, high resolution, side-view imaging, etc. It can obtain the fine features and geometric features of the target. It is very urgent to make fast and accurate detection of man-made targets in military and civilian applications. The efficient use of convolutional network in image feature extraction makes it have important theoretical value and broad application prospects in solving the problem of extremely SAR artificial target detection. [0003] Traditional convolutional networks such as CNN use image blocks as input for training and prediction to implement a pixel category detection task, which not only increases storage space, but also has low computati...

AI Technical Summary

Problems solved by technology

[0003] Traditional convolutional networks such as CNN use image blocks as input for training and prediction to implement a pixel category detection task, which not only increases storage space, but also has low computational efficiency, and the size of image blocks is much smaller than the entire image. Lead to the loss of some features, thus limiting the detection performance

[0004] In 2015, in response to the problems existing in CNN, Jonathan Long et al. proposed Fully Convolutional Networks, referred to as FCN. This network extends the category detection task from the image level to the pixel level, thereby detecting the region of interest, but the FCN detection result is not fine enough. It is easy to ignore the details of the image, and cannot accurately detect artificial targets containing multi-polarization features, and the pixel-level detection does not fully consider the spatial neighborhood information

Images