Mixed block similarity-based polarized SAR (Synthetic Aperture Radar) image speckle reduction method

Abstract

The invention discloses a mixed block similarity-based polarized SAR (Synthetic Aperture Radar) image speckle reduction method, and belongs to the technical field of image processing. The method comprises the following implementation processes: inputting polarized SAR covariance matrix data, and constructing a shape adaptive block of each point of the polarized SAR covariance matrix data; calculating homogeneous similarity weights and structural similarity weights of all points in the search area of each point of the polarized SAR data to obtain a mixed similarity weight; and obtaining the estimation values of the covariance matrix of each point of the polarized SAR data by weighted averaging, and performing Pauli decomposition to obtain a speckle-removed image of the polarized SAR data. By the method, the mixed similarity weight is introduced by combining the homogeneous similarity weights and the structural similarity weights, similar sets can be selected more accurately, and the problem that detail information remaining and speckle noise suppression cannot be balanced very well by a Pretest filter method is solved. The speckle reduction capacity of a polarized SAR image can be improved, and the detail information can be remained well; therefore, the method can be used for the speckle reduction processing of the polarized SAR image.

Description

technical field [0001] The invention belongs to the technical field of image processing, and is mainly aimed at the improvement of the Pretest filter method, in particular to a speckle reduction method for polarimetric SAR images based on mixed block similarity, which can be used for speckle removal processing of polarimetric SAR images. Background technique [0002] Polarization Synthetic Aperture Radar (POLSAR) simultaneously transmits and receives echo signal pulses in the horizontal direction H and vertical direction V, and there is a correlation between each polarization pulse. It is a new imaging radar for measuring echo signals. The polarimetric SAR image can not only measure the amplitude of the signal, but also record the phase difference information of the echo signal between different polarization states. It not only has the characteristics of all-weather, all-time, high resolution and strong penetrating power of SAR images, but also provides more abundant ground ...

AI Technical Summary

Problems solved by technology

This method has a good speckle suppression effect on the homogeneous area of ​​the polarimetric SAR image, but because it only uses the block-based structural similarity measure when selecting similar points, and this structural similarity measure is in some images The area is not accurate, such as measuring the structural similarity between a weak target area and a uniform area. The larger the block size, the less accurate the structural similarity measurement between them, which causes the restored value of the image pixel to deviate from its true value too much. Big

Therefore, the details of the image filtered by the Pretest filter method will become blurred as the block size increases, which will bring difficulties to subsequent applications such as target detection, recognition, and object classification.

[0005] The Pretest filter method has a good speckle suppression effect on the homogeneous area of ​​the polarimetric SAR image, but because it uses a smaller block, the filter effect of this method is poor at the linear structure and the texture structure, because the small image block contains more Less structural information; and if a larger image block is used, it will blur details such as point targets and edges

Images