Image texture filtering method based on weighted median filtering

Abstract

The invention provides an image texture filtering method based on weighted median filtering. The method can be briefly summarized as following five steps: a user provides an image as an input image; according to the input image, average filtering operation is performed to obtain a processed image; according to the input image, an image improved relative to the overall change is obtained; according to the obtained image, the interpolated input image, and the average filtered image, a guide image is obtained; according to the guide image, a weight is calculated, the weighted median filtering is adopted to perform texture filtering operation on the input image, and a texture filtered image is obtained. According to the invention, based on the texture filtering operation performed on the image by the weighted median filtering, a new method for calculating the guide image is provided, and compared with the algorithm in the prior art, the method can preferably process the texture information having extreme change in the image.

Description

Technical field [0001] The invention relates to an image texture filtering method, in particular to an image texture method based on weighted median filtering. Background technique [0002] Image texture filtering algorithm refers to the use of algorithms to remove the complex texture information contained in the image while keeping the significant structural information in the image from being weakened. Image texture filtering algorithms have a wide range of applications in image processing, machine vision, pattern recognition and other fields. For example, the image texture filtering algorithm can be used to remove some densely distributed noise information contained in the image, and can be used as a preprocessing operation to improve the effect of edge detection. Due to the importance of image texture filtering algorithms, many algorithms have been proposed to perform texture filtering operations. [0003] Image filtering algorithms based on boundary preservation, such algori...

AI Technical Summary

Problems solved by technology

Due to the characteristics of the bilateral filtering algorithm, the algorithm of Cho et al. will blur the structural boundary of the image to a certain extent and at the same time cannot handle the extremely changing texture information well (see Cho H, Lee H, Kang H, Lee S.Bilateral texture filtering. ACM Transactions on Graphics, 2014, 33(4): 1-8)

Images