Noise reduction method and computer-readable storage medium for X-ray perspective image

Abstract

The invention discloses a noise reduction method for X-ray fluoroscopic images and a computer-readable storage medium. The method includes: performing Anscombe transformation on an X-ray fluoroscopic image to be de-noised to obtain a first image; Filter the first image to obtain a second image; perform non-local mean filtering on the second image to obtain a third image; perform linear weighting on the X-ray fluoroscopic image and the third image to obtain a fourth image; The fourth image is subjected to inverse Anscombe transform to obtain a denoised X-ray fluoroscopic image. The invention has good noise reduction effect and can retain image details.

Description

technical field [0001] The invention relates to the technical field of image processing, in particular to a noise reduction method for X-ray fluoroscopic images and a computer-readable storage medium. Background technique [0002] DR (Digital Radiography), direct digital X-ray photography system, as the development trend of the current mainstream X-ray filming, has become an essential equipment in the radiology department of the hospital. Its fluoroscopy function is of great value in clinical auxiliary diagnosis, but due to various reasons of imaging systems such as hardware and software, the Poisson distribution noise of X-ray fluoroscopy images is very obvious, which affects clinical auxiliary diagnosis. However, the classic noise reduction algorithms based on Gaussian additive noise failed to achieve good results in clinical applications, such as Gaussian filtering method, bilateral filtering method, NLM (Non-Local Means, non-local mean) filtering method and so on. [00...

AI Technical Summary

Problems solved by technology

Its fluoroscopy function is of great value in clinical auxiliary diagnosis, but due to various reasons of imaging systems such as hardware and software, the Poisson distribution noise of X-ray fluoroscopy images is very obvious, which affects clinical auxiliary diagnosis

However, the classic noise reduction algorithm based on Gaussian additive noise failed to achieve good results in clinical applications, such as Gaussian filtering method, bilateral filtering method, NLM (Non-Local Means, non-local mean) filtering method, etc.

[0003] Although these methods can reduce noise well and have fast processing speed, they will have different degrees of defects.

The Gaussian filtering method can better reduce noise, but it will make the image blurred and lose image details; the bilateral filtering method will make the image blurred; NLM filtering can better retain image details, but the noise reduction effect is not good.

Therefore, the X-ray fluoroscopy images processed by these methods cannot meet the requirements of clinical auxiliary diagnosis.

Images