a g-based 0 Distributed Stochastic Gradient Variational Bayesian SAR Image Segmentation Method

Abstract

The invention discloses a G-based 0 The distributed stochastic gradient variational Bayesian SAR image segmentation method extracts the sketch map of the SAR image according to the initial sketch model; divides the SAR image into mixed pixel subspace, homogeneous pixel subspace and structured pixel subspace according to the area map; Each extremely inhomogeneous region in the pixel subspace estimates its G 0 distribution parameters, using G based on 0 The distributed stochastic gradient variational Bayesian model learns its structural features, thereby realizing the unsupervised segmentation of the mixed pixel subspace; for the homogeneous pixel subspace and the structural pixel subspace, the corresponding segmentation is performed, and the segmentation results of the three subspaces are fused. , and finally get the SAR image segmentation result. Because the present invention assumes both the prior distribution and approximate posterior distribution of latent variables in the model to satisfy the G in the extremely heterogeneous region 0 The corresponding analytical formula is derived for learning, so the accuracy of clustering of extremely inhomogeneous regions in the mixed pixel subspace is improved.

Description

technical field [0001] The invention belongs to the technical field of image processing, in particular to a G-based 0 The distributed stochastic gradient variational Bayesian SAR image segmentation method can be applied to accurately segment different regions of synthetic aperture radar SAR and target detection and recognition in SAR images. Background technique [0002] Synthetic aperture radar (SAR) is an important progress in the field of remote sensing technology, which is used to obtain high-resolution images of the earth's surface. Compared with other types of imaging technologies, SAR has a very important advantage. It is not affected by atmospheric conditions such as clouds, rainfall or heavy fog, and light intensity, and can obtain high-resolution remote sensing data all day and all weather. SAR technology has important guiding significance for military, agriculture, geography and many other fields. [0003] Image segmentation refers to the process of dividing an ...

AI Technical Summary

Problems solved by technology

The disadvantage of this method is that when obtaining the feature vector of the SAR image, the pixel-level features of the SAR image are used, and the unique structural features of the SAR image due to the correlation between pixels are not automatically learned. Insufficient use of structural features that truly represent the characteristics of SAR image features, resulting in inaccurate segmentation results

Images