CN-121599874-B - OCT image filtering enhancement method for cardiovascular internal medicine

CN121599874BCN 121599874 BCN121599874 BCN 121599874BCN-121599874-B

Abstract

The invention relates to the technical field of image processing, in particular to an OCT image filtering enhancement method for cardiovascular internal medicine, which comprises the following steps: the method comprises the steps of obtaining an OCT image, and classifying distribution parameters and tissue distribution parameters of the image, wherein the classifying distribution parameters are obtained by classifying an original image based on brightness change of the original image, and the tissue distribution parameters are obtained by dividing the original image based on tissue distribution of the original image. And then determining a noise influence factor of the original image based on the difference of the classification distribution parameters and the tissue distribution parameters, and carrying out non-local mean filtering on the original image according to the noise influence factor to obtain a clear target image with speckle noise suppressed, and finally enhancing the target image, so that the enhanced OCT image is ensured to retain important structural details and reduce unnecessary noise interference.

Inventors

Wang Yunhuan
LI WEI
DENG ZINING

Assignees

辽宁省全无信息技术有限公司

Dates

Publication Date: 20260508
Application Date: 20251201

Claims (8)

1. A cardiovascular OCT image filtering enhancement method, the method comprising: acquiring an original image, wherein the original image is an Optical Coherence Tomography (OCT) image; Acquiring classification distribution parameters of the original image, wherein the classification distribution parameters are obtained by classifying the original image based on the brightness change of the original image; obtaining tissue distribution parameters of the original image, wherein the tissue distribution parameters are obtained by dividing the original image based on tissue distribution of the original image; determining a noise impact factor of the original image based on the classification distribution parameter and the tissue distribution parameter, the noise impact factor being used to represent a difference between the classification distribution parameter and the tissue distribution parameter; According to the noise influence factors, carrying out non-local mean filtering on the original image to obtain a target image; enhancing the target image; the obtaining the classification distribution parameters of the original image comprises the following steps: Acquiring a brightness fluctuation parameter of each pixel point in the original image, wherein the brightness fluctuation parameter is determined based on the change rate of the gray value of the pixel point; Clustering each pixel point based on the brightness fluctuation parameters of each pixel point to obtain a plurality of first classification clusters and first pixel point clusters contained in the plurality of first classification clusters; determining an image distribution factor of each first classification cluster according to the relation between a first pixel point cluster in each first classification cluster and a first classification cluster center point; Determining classification distribution parameters of the original image according to the image distribution factors of each first classification cluster; The obtaining the tissue distribution parameters of the original image comprises the following steps: Dividing the original image subjected to the closed operation to obtain a plurality of second pixel point clusters; determining a snare parameter of each second pixel cluster according to the plurality of second pixel clusters; and determining the tissue distribution parameters of the original image according to the snare parameters of each second pixel point cluster.
2. The method for enhancing filtering of OCT images according to claim 1, wherein the parameters of the fluctuation of brightness and darkness are determined based on the distance between the maximum points in the original image, the gray values of the maximum points, the average gray values of the pixel points between the maximum points, and the number of the pixel points between the maximum points, wherein the maximum points are the pixel points with the largest gray values in the neighborhood of the pixel points.
3. The method of claim 2, wherein the image distribution factor is determined based on an entropy of information of the first pixel cluster in each first classification cluster from the center point angle and a variance of the distance from the center point.
4. A cardiovascular OCT image filtering enhancement method according to claim 3, wherein the classification distribution parameters are determined based on an average value of the image distribution factors of each first classification cluster and an information entropy of an average center distance of the first pixel clusters in each first classification cluster.
5. The method of claim 1, wherein the snare parameter is determined based on a minimum circumscribed circle area of the second cluster of pixels, and a number of pixels belonging to the second cluster of pixels in the minimum circumscribed circle.
6. The method of claim 1, wherein the tissue distribution parameter is determined based on an entropy of a product of a minimum circumscribed circle diameter of the second pixel cluster and the snare parameter, and a variance of a minimum circumscribed circle center of the second pixel cluster.
7. A method of enhancing an OCT image of cardiovascular medicine according to claim 3, wherein said enhancing the target image comprises: the target image is enhanced based on histogram equalization.
8. A method of enhancing OCT image filtering of cardiovascular medicine according to claim 3, wherein the original image is a cardiovascular image.

Description

OCT image filtering enhancement method for cardiovascular internal medicine Technical Field The invention relates to the technical field of image processing, in particular to an OCT image filtering enhancement method for cardiovascular internal medicine. Background Optical coherence tomography (Optical Coherence Tomography, OCT) is a high resolution imaging technique widely used for cardiovascular examinations, such as assessing the microstructure of blood vessels, in particular in assessing atherosclerotic plaques and other cardiovascular diseases. Since some minor or early lesions may be undetectable in the original image, special enhancement processing of the cardiovascular OCT image is required to improve the physician's examination of the patient's cardiovascular OCT image, which may help to find these minor abnormalities. However, due to the speckle noise in the OCT image, the speckle noise is also enhanced during the enhancement process, thereby affecting the sharpness and enhancement effect of the image. Disclosure of Invention In order to solve the technical problem that the OCT image is not clear due to the fact that speckle noise is enhanced when the OCT image is enhanced, the invention aims to provide a cardiovascular OCT image filtering enhancement method, and the technical scheme adopted by the method is as follows: a cardiovascular medical OCT image filtering enhancement method, the method comprising: Acquiring an original image, wherein the original image is an Optical Coherence Tomography (OCT) image; acquiring classification distribution parameters of an original image, wherein the classification distribution parameters are obtained by classifying the original image based on the brightness change of the original image; obtaining tissue distribution parameters of an original image, wherein the tissue distribution parameters are obtained by dividing the original image based on tissue distribution of the original image; Determining a noise influence factor of the original image based on the classification distribution parameter and the tissue distribution parameter, wherein the noise influence factor is used for representing the difference between the classification distribution parameter and the tissue distribution parameter; According to the noise influence factors, carrying out non-local mean filtering on the original image to obtain a target image; The target image is enhanced. Optionally, the step of acquiring the classification distribution parameters of the original image includes: Acquiring a brightness fluctuation parameter of each pixel point in an original image, wherein the brightness fluctuation parameter is determined based on the change rate of the gray value of the pixel point; clustering each pixel point based on the brightness fluctuation parameters of each pixel point to obtain a plurality of first classification clusters and first pixel point clusters contained in the plurality of first classification clusters; Determining an image distribution factor of each first classification cluster according to the relation between a first pixel point cluster in each first classification cluster and a first classification cluster center point; and determining the classification distribution parameters of the original image according to the image distribution factors of each first classification cluster. Optionally, the brightness fluctuation parameter is determined based on a distance between each maximum point, a gray value of each maximum point, an average gray value of pixel points between the maximum points, and the number of pixel points between the maximum points in the original image, where the maximum point is a pixel point with the largest gray value in the neighborhood of each pixel point. Optionally, the image distribution factor is determined based on the information entropy of the first pixel point cluster and the angle of the center point in each first classification cluster and the variance of the distance from the center point. Optionally, the classification distribution parameter is determined based on an average value of the image distribution factors of each first classification cluster and an information entropy of an average center distance of the first pixel point clusters in each first classification cluster. Optionally, the steps of obtaining the tissue distribution parameters of the original image include: Dividing the original image subjected to the closed operation to obtain a plurality of second pixel point clusters; determining a snare parameter of each second pixel cluster according to the plurality of second pixel clusters; And determining the tissue distribution parameters of the original image according to the snare parameters of each second pixel cluster. Optionally, the snare parameter is determined based on a minimum circumscribed circle area of the second pixel cluster, and a number of pixels belonging to the second pixel cluster in the minimum ci