Noise reduction processing method of ultrasound medical image

Abstract

The invention discloses a noise reduction processing method of an ultrasound medical image. The noise reduction processing method of the ultrasound medical image is characterized by comprising the following steps of calculating an edge intensity image E of a gray-scale ultrasound image I; calculating a mask image P in a smooth organization structure area of the gray-scale ultrasound image I; calculating an edge intensity threshold curve Kcurve of the gray-scale ultrasound image I along a depth direction of an ultrasonic probe scanning line; extracting a corresponding edge intensity threshold from the edge intensity threshold curve Kcurve and utilizing a diffusion coefficient function g to obtain a diffusion coefficient for any ith row image Ei of the edge intensity image E; and utilizing a diffusion equation (formula as follows, wherein all portions of the formula respectively stand for an image the gray-scale ultrasound image I evolving to time t to be formed into, an evolution time step length, a divergence operator and a gradient of the gray-scale ultrasound image I) to perform filtering processing on the image according to the generated diffusion coefficient. The noise reduction processing method of the ultrasound medical image has the advantages of enabling the ultrasound medical image to be capable of utilizing different edge intensity thresholds along the depth direction of probe scanning and achieving the effect of speckle noise suppression optimization.

Description

technical field [0001] The invention relates to the technical field of medical image processing, in particular to a noise reduction processing method for gray-scale ultrasonic images by automatically detecting a diffusion threshold. Background technique [0002] There is usually a large amount of speckle noise in medical gray-scale ultrasound images (B mode), which will significantly reduce the quality of ultrasound images and cover up some important tissue lesions. It brings great difficulties, and there is a risk of missed diagnosis and misdiagnosis. [0003] The filtering method is usually used to suppress the high-frequency part of the signal, and the useful edge information will be removed while filtering out the noise. Over-smoothing; As a result, some filtering algorithms based on edge preservation have been developed rapidly, such as anisotropic diffusion, BM3D and nonlocal means and other medical image denoising techniques. This provides a huge space for improving...

AI Technical Summary

Problems solved by technology

[0034] According to the imaging characteristics of ultrasonic gray-scale images, the present invention uses anisotropic filtering as a framework to propose a noise reduction processing method for gray-scale ultrasonic images based on automatic detection of diffusion thresholds, so as to solve the problem of smoothing of current medical ultrasonic image noise reduction processing. Or the occurrence of false structures and other technical problems that lead to the decline in the quality of medical ultrasound images

Images