Method and device for filtering out noise in pixels

Abstract

The embodiment of the invention provides a method and device for filtering out noise in pixels to solve the problem that according to an existing filtering method, when the pixels in an image comprise impulse noise and gaussian noise, the impulse noise and the gaussian noise need to be filtered out respectively one time, the efficiency is low when the filtering method is used for filtering out the noise in the image. The method comprises the steps that at least one similarity degree of a contrast pixel with a corresponding pixel is determined from a local neighbor area of a compare pixel according to the compare pixel; according to the determined similarity degrees, the weight of the local neighbor area of the compare pixel to the local neighbor area of pixels to be processed is determined; the pixel value of filtered to-be-processed pixels is determined according to the weight of the local neighbor area of the at least one compare pixel to the local neighbor area of the pixels to be processed in a sliding window.

Description

technical field [0001] The invention relates to the technical field of image processing, in particular to a method and device for filtering out noise in pixels. Background technique [0002] In the field of image processing, filtering out noise in images is an important step in image processing. The noise in the image comes from the process of image acquisition, transmission and storage. For example, when the image is captured, the light condition is poor, the sensitivity is high, the image acquisition device has a dead pixel, or an error occurs during the conversion of an analog image into a digital image, an error occurs during image storage, or interference from noise during image transmission, etc. These noises bring great difficulties to the post-processing of images. [0003] Although there are various forms of noise, almost all of them can be represented by a combination of the two most basic noise models, Gaussian noise and impulse noise. For impulse noise, there ...

AI Technical Summary

Problems solved by technology

[0011] Embodiments of the present invention provide a method and device for filtering noise in pixels, which are used to solve the problem that when the pixels in the image contain both impulsive noise and Gaussian noise in the existing filtering method, the impulsive noise and Gaussian noise need to be respectively Each filter is performed once, which leads to the problem of low efficiency of these filtering methods in filtering out the noise in the image

Images