CN118247277B - Self-adaptive enhancement method for lung CT image - Google Patents
Self-adaptive enhancement method for lung CT image Download PDFInfo
- Publication number
- CN118247277B CN118247277B CN202410666336.5A CN202410666336A CN118247277B CN 118247277 B CN118247277 B CN 118247277B CN 202410666336 A CN202410666336 A CN 202410666336A CN 118247277 B CN118247277 B CN 118247277B
- Authority
- CN
- China
- Prior art keywords
- pixel point
- gray
- lung
- pixel
- gray value
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/762—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using clustering, e.g. of similar faces in social networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
- G06T2207/30061—Lung
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Multimedia (AREA)
- Computing Systems (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Evolutionary Computation (AREA)
- Databases & Information Systems (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Quality & Reliability (AREA)
- Apparatus For Radiation Diagnosis (AREA)
Abstract
The invention relates to the technical field of image enhancement, in particular to a lung CT image self-adaptive enhancement method. The method comprises the following steps: acquiring a lung region in a lung CT image; according to the gray scale change condition of each pixel point in the lung area and the pixel points in the neighborhood of the pixel point, determining a gray scale change trend vector and an extension evaluation value of each pixel point; according to the difference between the gray level change trend vector of each pixel point and the gray level change trend vector of the pixel point in the preset nearest neighbor, obtaining a possibility index that each pixel point is a blood vessel pixel point; determining the probability that each pixel belongs to a blood vessel by combining the extension evaluation value, the possibility index and the gray value; determining the density degree of each gray value based on the number of the pixel points with the same gray value in the neighborhood of each pixel point corresponding to each gray value; the probability and the intensity of each pixel belonging to the blood vessel are combined to strengthen the lung area. The invention improves the enhancement effect of lung CT images.
Description
Technical Field
The invention relates to the technical field of image enhancement, in particular to a lung CT image self-adaptive enhancement method.
Background
Lung CT is a conventional diagnosis and treatment means, specifically, X-rays are used to obtain images of the lungs of a human body, and then CT images are used to assist a doctor in checking related diseases of the lungs. However, due to limitations of technical problems or occurrence of accidental factors, the quality of the lung CT image is insufficient, for example, the boundary of the image is not clear enough, and the distribution of bronchi in the lung is blurred, and at this time, the effect of segmenting the image by using the image segmentation algorithm is not good, so that the image processing is required to be performed on the lung CT image to obtain the clear lung CT image so as to obtain the optimal segmentation effect.
When the lung CT image is segmented, because the contrast ratio of important tissues such as blood vessels and other tissues of the lung is low, the lung CT image is required to be enhanced in practice, and the lung tissues are more intuitively and clearly represented for a doctor to observe, but the enhancement of the CT image in the prior art is mainly directly enhanced based on the frequency relation of pixel level, and the degree of representation of image details by different pixel levels is not combined, namely the enhancement process is not targeted, so that the enhancement result cannot represent the detail characteristics in the lung CT image.
Disclosure of Invention
In order to solve the problem that the result of the enhancement processing of the lung CT image by the existing method can not express the detail characteristics in the lung CT image, the invention aims to provide a self-adaptive enhancement method for the lung CT image, which adopts the following technical scheme:
The invention provides a lung CT image self-adaptive enhancement method, which comprises the following steps:
Acquiring a lung region in a lung CT image to be processed;
analyzing the gray scale change condition of each pixel point in the lung area and the pixel points in the neighborhood of the pixel point, and determining the gray scale change trend vector of each pixel point and the extension evaluation value of each pixel point; according to the difference condition of the gray level change trend vector of each pixel point in the lung area and the gray level change trend vector of the pixel point in the preset nearest neighbor, obtaining a possibility index that each pixel point is a vascular pixel point;
Determining the probability that each pixel belongs to a blood vessel by combining the extension evaluation value, the possibility index and the gray value; determining the density degree of each gray value based on the number of the pixel points with the same gray value in the neighborhood of each pixel point corresponding to each gray value in the lung region;
And reinforcing the lung region by combining the probability that each pixel point corresponding to each gray value belongs to a blood vessel and the corresponding density degree to obtain a reinforced lung image.
Preferably, the obtaining the gray scale variation trend vector of each pixel point includes:
for candidate pixel points, respectively calculating gray difference values between the candidate pixel points and each pixel point in the neighborhood of the candidate pixel points, determining the direction of a feature vector based on the sign of the gray difference values, and obtaining the feature vector of each pixel point in the neighborhood based on the absolute value of the gray difference values as the size of the feature vector;
The distribution of the feature vectors of all the pixel points in the neighborhood of the candidate pixel point is synthesized to determine the gray level change trend vector of the candidate pixel point; the candidate pixel points are any one pixel point in the lung area.
Preferably, the extending evaluation value of the pixel point includes:
For candidate pixel points:
The two pixel points which are the closest to the candidate pixel point in the vertical direction of the gray level change trend vector of the candidate pixel point and pass through the straight line of the candidate pixel point are marked as characteristic points of the candidate pixel point; marking the pixel points except the characteristic points in the eight neighborhood of the candidate pixel point as reference points of the candidate pixel point;
And obtaining the extension evaluation value of the candidate pixel point according to the gray level difference between the candidate pixel point and the characteristic point and the reference point.
Preferably, the extension evaluation value of the i-th pixel point is calculated using the following formula:
Wherein, An extended evaluation value representing the ith pixel point of the lung area,A gray value representing the ith pixel point of the lung field,A gray value representing a first feature point of an ith pixel point of the lung region,A gray value representing a second feature point of the ith pixel point of the lung area, U representing the number of reference points of the ith pixel point of the lung area,A gray value representing the ith reference point of the ith pixel point of the lung area,The sign of the absolute value is taken as the representation,Indicating that the first adjustment parameter is preset,Greater than 0; the feature points include a first feature point and a second feature point.
Preferably, the obtaining the probability index that each pixel point is a vascular pixel point according to the difference condition of the gray level change trend vector of each pixel point in the lung area and the gray level change trend vector of the pixel point in the preset nearest neighbor includes:
According to the difference between the gray level change trend vector of the candidate pixel point and the gray level change trend vector of the pixel point in the preset nearest neighbor of the candidate pixel point, a possibility index of the candidate pixel point being a blood vessel pixel point is obtained, and the difference of the gray level change trend vector and the possibility index are in a negative correlation relationship.
Preferably, the determining the probability that each pixel belongs to a blood vessel by combining the extension evaluation value, the likelihood index and the gray value includes:
And obtaining the probability that the candidate pixel belongs to the blood vessel according to the extension evaluation value, the possibility index and the gray value of the candidate pixel, wherein the extension evaluation value, the possibility index and the gray value are in positive correlation with the probability.
Preferably, the determining the intensity of each gray value based on the number of pixels with the same gray value in the neighborhood of each pixel corresponding to each gray value in the lung area includes:
For the candidate gray values, determining the density degree of the candidate gray values based on the number proportion of the pixel points of the candidate gray values in the neighborhood of each pixel point corresponding to the candidate gray values, wherein the number proportion and the density degree are in positive correlation;
the candidate gray value is any gray value in the lung area.
Preferably, the enhancing the lung region by combining the probability that each pixel point corresponding to each gray value belongs to a blood vessel and the corresponding intensity degree to obtain an enhanced lung image includes:
Combining the probability that all pixel points corresponding to each gray value belong to blood vessels and the corresponding intensity, and obtaining the enhancement coefficient of each gray value, wherein the probability and the intensity are in positive correlation with the enhancement coefficient;
and reinforcing each gray value in the lung region based on the reinforcing coefficient to obtain a reinforced lung image.
Preferably, the enhancing each gray value in the lung region based on the enhancement coefficient to obtain an enhanced lung image includes:
marking the product of each gray value and the enhancement coefficient of the gray value in the lung area as a characteristic index of each gray value, and taking the minimum value of the characteristic index of each gray value and 255 as a target gray value corresponding to each gray value;
And replacing the original gray value with the target gray value to obtain the enhanced lung image.
Preferably, the acquiring the lung region in the CT image of the lung to be processed includes:
The pixel points in the CT image of the lung to be processed are divided into two types by adopting a K-means clustering algorithm, and the lung region is determined by combining the gray distribution condition of each type of pixel points.
The invention has at least the following beneficial effects:
The invention analyzes the gray scale condition of surrounding pixel points of each pixel point in the lung region, and evaluates the possibility that the pixel point belongs to the pixel point on the blood vessel by combining the gray scale value of the single pixel point, namely, the pixel point is primarily evaluated from the structural characteristics and the gray scale characteristics of the blood vessel in the lung region; because the lung is formed by mutually matching a plurality of tissues, the respiratory function of the lung is finished together, the gray values of the pixel points of the same tissue are close, the method analyzes the density degree of the pixel points of each gray value, further carries out enhancement treatment on the pixel points of different gray values in different degrees by combining the probability that the pixel points belong to blood vessels and the density degree of each gray value, has better enhancement effect, enables the enhanced image to be clearer, can represent the detail characteristics in the CT image of the lung, and can ensure the accuracy of the analysis effect by analyzing the lung condition by using the enhanced image.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions and advantages of the prior art, the following description will briefly explain the drawings used in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are only some embodiments of the invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flowchart of a method for adaptively enhancing CT images of a lung according to an embodiment of the present invention;
FIG. 2 is a schematic view of a lung region;
FIG. 3 is a flow chart of a method of obtaining a likelihood indicator;
fig. 4 is a block diagram of a lung CT image adaptive enhancement system.
Detailed Description
In order to further describe the technical means and effects adopted by the present invention to achieve the preset purpose, the following detailed description is given to a lung CT image adaptive enhancement method according to the present invention with reference to the accompanying drawings and preferred embodiments.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
The following specifically describes a specific scheme of the lung CT image adaptive enhancement method provided by the invention with reference to the accompanying drawings.
An embodiment of a lung CT image self-adaptive enhancement method:
The specific scene aimed at by this embodiment is: when analyzing the specific condition of the lung, the CT image of the lung is usually collected first, but the CT image is often interfered by the collecting equipment and environmental factors in the collecting process, the quality of the CT image of the lung collected by the image further influences the subsequent analysis result, and the embodiment combines the specific characteristics of the CT image of the lung to enhance the collected CT image of the lung, so as to improve the quality of the CT image of the lung and ensure the accuracy of the subsequent analysis result.
The embodiment provides a lung CT image adaptive enhancement method, as shown in fig. 1, which includes the following steps:
step S1, acquiring a lung region in a lung CT image to be processed.
The method comprises the steps of firstly, collecting a lung CT image of an object to be analyzed by using CT scanning equipment, carrying out gray-scale treatment on the collected lung CT image of the object to be analyzed, and recording the image after the gray-scale treatment as the lung CT image to be processed, wherein the collected image contains bronchi, blood vessels, lung parenchyma and alveoli and possibly also contains some irrelevant information, as shown in fig. 2, the image is a schematic diagram of a lung area, namely, the area between two blackish lung positions and a brightly white outer ring area, so that the lung area in the collected image can be extracted by using a semantic segmentation network, the lung area can be extracted based on gray values and positions of pixel points in the lung CT image to be processed, the interference of irrelevant factors is eliminated, and the subsequent independent analysis is facilitated and the pertinence is enhanced.
In the embodiment, the pixel points in the lung CT image to be processed are divided into two types by adopting a K-means clustering algorithm, and the value of K is 2,K-means clustering algorithm in the clustering process by adopting the K-means clustering algorithm is the prior art and is not repeated here; and respectively calculating the gray average value of each type of pixel point, and taking the cluster with the maximum gray average value as the lung region.
Thus far, the present embodiment acquires the lung region in the CT image of the lung to be processed.
S2, analyzing the gray scale change condition of each pixel point in the lung area and the pixel points in the neighborhood of the pixel point, and determining the gray scale change trend vector of each pixel point and the extension evaluation value of each pixel point; and obtaining a possibility index of each pixel point as a blood vessel pixel point according to the difference condition of the gray level change trend vector of each pixel point in the lung region and the gray level change trend vector of the pixel point in the preset nearest neighbor.
Tissues which can be displayed by normal lung tissues in CT images are bronchi, blood vessels, lung parenchyma, alveoli and the like; the bronchus presents as a bright gray edge, a strip or ellipse inside a dark color in the image, and the blood vessels are in bright gray dendrites or blocks with different sizes, and the two are brighter due to higher density and color; lung parenchyma appears as a fog, alveoli appear as sacs, and are not recognized in CT images under normal conditions. Even some finer vascular human eyes are difficult to identify when the contrast of the CT image is low.
Because the dendritic blood vessel has a certain extensibility, that is, for the pixel points located on the blood vessel, a series of pixel points with smaller gray value difference exist in a certain direction in the neighborhood of the pixel points, the direction is determined according to the gray distribution condition of each pixel point and the surrounding pixel points, and the gray change trend vector of each pixel point is obtained.
Step S21, according to the gray scale change condition of each pixel point and the pixel points in the neighborhood, a gray scale change trend vector and an extension evaluation value of each pixel point are obtained.
In this embodiment, analysis is performed by taking one pixel point in the lung area as an example, and the method provided in this embodiment can be used to process other pixel points.
Any pixel point in the lung area is marked as a candidate pixel point, and gray level difference values between the candidate pixel point and each pixel point in the neighborhood of the candidate pixel point are calculated respectively for the candidate pixel point, and the following needs to be described: the specific acquisition method of the gray level difference value between the candidate pixel point and each pixel point in the neighborhood comprises the following steps: and subtracting the gray value of each pixel in the neighborhood of the candidate pixel from the gray value of the candidate pixel, wherein the obtained difference is used as the gray difference between the candidate pixel and each pixel in the neighborhood of the candidate pixel. Obtaining a characteristic vector of each pixel point in the neighborhood of the candidate pixel point based on the gray difference value, wherein the size of the characteristic vector is an absolute value of the corresponding gray difference value, and if the gray difference value is a positive number, the direction of the characteristic vector is the direction from the corresponding pixel point in the neighborhood to the candidate pixel point; if the gray difference value is a negative number, the direction of the feature vector is the direction from the candidate pixel point to the corresponding pixel point in the neighborhood; if the gray difference is 0, the feature vector is zero vector. Each pixel point in the neighborhood of the candidate pixel point corresponds to a feature vector. And taking the accumulated sum of the feature vectors of all the pixel points in the neighborhood of the candidate pixel point as a gray level change trend vector of the candidate pixel point, and reflecting the gray level change condition of surrounding pixel points of the candidate pixel point relative to the candidate pixel point.
For candidate pixel points: the method comprises the steps of marking two pixel points which are closest to a candidate pixel point and pass through a straight line of the candidate pixel point in the vertical direction of a gray level change trend vector of the candidate pixel point as feature points of the candidate pixel point, namely obtaining two feature points of the candidate pixel point, and marking the two feature points as a first feature point and a second feature point respectively; and marking the pixel points except the two characteristic points in the eight adjacent areas of the candidate pixel point as reference points of the candidate pixel point. And obtaining the extension evaluation value of the candidate pixel point according to the gray level difference between the candidate pixel point and the characteristic point and the reference point. And (5) extending the gray scale change condition of the evaluation value to the pixel points in the neighborhood of the pixel points. The present embodiment marks the eight neighbors here as the first neighbors.
The gray scale difference may be represented by the square of the difference between the two gray scale values, or by the absolute value of the difference between the two gray scale values.
In this embodiment, a specific calculation formula of the extension evaluation value is given, and the extension evaluation value of the i-th pixel point may be specifically expressed as:
Wherein, An extended evaluation value representing the ith pixel point of the lung area,A gray value representing the ith pixel point of the lung field,A gray value representing a first feature point of an ith pixel point of the lung region,A gray value representing a second feature point of the ith pixel point of the lung area, U representing the number of reference points of the ith pixel point of the lung area,A gray value representing the ith reference point of the ith pixel point of the lung area,The sign of the absolute value is taken as the representation,Indicating that the first adjustment parameter is preset,Greater than 0; the feature points include a first feature point and a second feature point.
In this embodiment, the preset first adjustment parameter is introduced into the calculation formula of the extended evaluation value to prevent the denominator from being 0, and in this embodiment, the preset first adjustment parameter is 0.01, and in a specific application, an implementer can set according to specific situations.For reflecting the gray scale difference between the i-th pixel point and its first feature point,For reflecting the gray scale difference between the i-th pixel point and its second feature point,The gray scale difference between the candidate pixel point and the characteristic point and the reference point is reflected, and the larger the value is, the larger the gray scale difference between the candidate pixel point and the characteristic point is. When the gray scale difference between the candidate pixel point and the characteristic point thereof is larger, and the gray scale difference between the candidate pixel point and the characteristic point thereof and the reference point is larger, the gray scale extension degree of the ith pixel point is higher, namely the extension evaluation value of the ith pixel point is larger.
By adopting the method, the gray scale change trend vector and the extension evaluation value of each pixel point in the lung region can be obtained.
Step S22, according to the difference condition of the gray level change trend vector of each pixel point in the lung area and the gray level change trend vector of the pixel point in the preset nearest neighbor, the possibility index that each pixel point is a blood vessel pixel point is obtained.
According to the characteristic of combining the dendritic blood vessel with the CT image, the gray value of the pixel point closer to the center is larger, which shows that the density of the internal substance is higher, and the gray value is smaller as compared with the gray value of the pixel point in the whole blood vessel is closer to the edge of the blood vessel until obvious difference appears in the gray value of the pixel point in the lung area and the gray change trend vector of the pixel point around the pixel point, so that the probability that each pixel point belongs to the blood vessel can be primarily evaluated.
Specifically, according to the difference between the gray level change trend vector of the candidate pixel point and the gray level change trend vector of the pixel point in the preset nearest neighbor of the candidate pixel point, a possibility index of the candidate pixel point being a blood vessel pixel point is obtained, and the difference of the gray level change trend vector and the possibility index are in a negative correlation. The negative correlation means that the dependent variable decreases with increasing independent variable, and the dependent variable increases with decreasing independent variable, and may be a subtraction relationship, a division relationship, or the like. The difference in the gradation change trend vector may be reflected by the square of the difference value or by an index such as an absolute value.
As a specific embodiment, a specific calculation formula of the likelihood index is given, and in this embodiment, the likelihood index that the ith pixel point in the lung area is a vascular pixel point may be expressed as:
Wherein, The probability index that the ith pixel point in the lung area is a vascular pixel point is represented, R represents the number of the pixel points in the preset nearest neighbor of the ith pixel point on the gray level change trend vector of the ith pixel point,The gray scale variation trend vector of the ith pixel point in the preset nearest neighbor of the ith pixel point is represented,The module length of the gray scale variation trend vector of the r pixel point in the preset nearest neighbor on the gray scale variation trend vector of the i pixel point is represented,A gray scale variation trend vector representing the i-th pixel point,The modulus of the gray scale variation trend vector representing the i-th pixel point,The included angle value between the gray scale change trend vector of the ith pixel point and the gray scale change trend vector of the r pixel point in the preset nearest neighbor of the gray scale change trend vector of the ith pixel point is represented,The circumference ratio is indicated.
The method is used for reflecting the difference situation between the module length of the gray level change trend vector of the ith pixel point and the r pixel point in the preset nearest neighbor of the ith pixel point on the gray level change trend vector of the ith pixel point, and the larger the value is, the larger the difference between the module length of the gray level change trend vector of the ith pixel point and the r pixel point in the preset nearest neighbor of the ith pixel point on the gray level change trend vector of the ith pixel point is.For reflecting the difference of the included angle between the gray scale variation trend vector of the ith pixel point and the gray scale variation trend vector of the ith pixel point in the preset nearest neighbor of the ith pixel point, and subtractingThe result is set to be on the monotonically increasing function, and the addition of 1 is set to be 0 or more. In this embodiment, the preset nearest neighbor is 6, that is, the value of R is 6, and in a specific application, the practitioner may set according to the specific situation. It should be noted that: in this embodiment, the pixel point in the preset nearest neighbor is the preset nearest neighbor pixel point on the line passing the current pixel point on the change trend vector of the current pixel point.
When the difference of the module length of the pixel point in the preset neighborhood on the ith pixel point and the gray change trend vector is smaller and the difference of the included angles is smaller, the more likely that the ith pixel point is the pixel point on the blood vessel is indicated, namely the greater the possibility index that the ith pixel point is the blood vessel pixel point is.
By adopting the method, the probability index of each pixel point in the lung region as the blood vessel pixel point can be obtained and is used for preliminarily reflecting the probability that the pixel point belongs to the pixel point on the blood vessel, as shown in fig. 3, which is a flow chart of the method for obtaining the probability index.
Step S3, determining the probability that each pixel belongs to a blood vessel by combining the extension evaluation value, the probability index and the gray value; the intensity of each gray value is determined based on the number of pixels of the same gray value in the neighborhood of each pixel corresponding to each gray value in the lung region.
The probability index only analyzes the gray level change trend of each pixel point and surrounding pixel points, so as to perform preliminary analysis on whether the pixel points belong to blood vessels, the extension evaluation value of the pixel points combines the characteristics of the blood vessels, the evaluation is performed from the angles of the blood vessel structure and gray level distribution, and the gray level value of the pixel points on the blood vessels is larger, so that the extension evaluation value, the probability index and the gray level value of the pixel points are combined to further evaluate the probability that the pixel points belong to the blood vessels in order to improve the accuracy of the judgment result of the pixel points of the blood vessels.
Specifically, according to the extension evaluation value, the probability index and the gray value of the candidate pixel point, the probability that the candidate pixel point belongs to a blood vessel is obtained, wherein the extension evaluation value, the probability index and the gray value are in positive correlation with the probability. The positive correlation relationship indicates that the dependent variable increases with the increase of the independent variable, and the dependent variable decreases with the decrease of the independent variable, and the specific relationship may be a multiplication relationship, an addition relationship, an idempotent of an exponential function, or the like.
In this embodiment, a specific calculation method for the probability that a pixel belongs to a blood vessel is provided, where the probability that an i-th pixel belongs to a blood vessel in a lung area may be expressed as:
Wherein, Representing the probability that the ith pixel point in the lung field belongs to a blood vessel,An extended evaluation value representing the ith pixel point of the lung area,A likelihood indicator indicating that the ith pixel in the lung field is a vascular pixel,The gray value representing the ith pixel point of the lung region, norm () represents the normalization function.
When the extension evaluation value of the ith pixel point is larger, the probability index that the ith pixel point is a blood vessel pixel point is larger, and the gray value of the ith pixel point is also larger, the ith pixel point is more likely to be a pixel point on a blood vessel, namely the probability that the ith pixel point in a lung area belongs to the blood vessel is larger.
By adopting the method, the probability that each pixel point in the lung area belongs to a blood vessel can be obtained.
Because the gray distribution condition of pixel points in the acquired lung CT image shows a certain rule, the density of air is lower, and the absorption of X-rays is also lower, the gray value of the air in the CT image is usually lower and is close to black, and the gray value is close to 0; soft tissues in the lungs, such as lung parenchyma, pulmonary blood vessels and lobes, have higher density and absorption capacity and therefore appear as higher gray values in CT images; the gray scale value of blood in CT images is typically higher than air, but lower relative to other tissues such as lung parenchyma, and the density and absorption capacity of blood is higher, but still lower than other tissues. If a lesion occurs in the lung, the tissue density of the lesion area is generally higher than that of normal lung tissue, and therefore, the lesion area is represented as a higher gray value in a lung CT image. Based on the above, the aggregation degree of the pixel points of each gray value in the lung region can be analyzed, so that different enhancement coefficients can be given to regions of different tissues, and further enhancement processing can be performed.
Specifically, this embodiment is described by taking a gray value of the lung area as an example, and other gray values can be processed by the method provided in this embodiment.
Any gray value in the lung area is recorded as a candidate gray value, and the following needs to be described: the same gray value is a gray value. For the candidate gray values, determining the density degree of the candidate gray values based on the number proportion of the pixel points of the candidate gray values in the neighborhood of each pixel point corresponding to the candidate gray values, wherein the number proportion and the density degree are in positive correlation; the positive correlation relationship indicates that the dependent variable increases with the increase of the independent variable, and the dependent variable decreases with the decrease of the independent variable, and the specific relationship may be a multiplication relationship, an addition relationship, an idempotent of an exponential function, and is determined by practical application.
In this embodiment, a specific calculation method of the intensity level is provided, where the intensity level of the jth gray scale value in the lung area may be expressed as:
Wherein, Represents the intensity of the j-th gray value in the lung field,The number of pixels representing the j-th gray value in the lung field,Representing the number of pixels with the j-th gray value in the preset second neighborhood of the n-th pixel corresponding to the j-th gray value in the lung region,And representing the total number of the pixel points in the preset second neighborhood of the nth pixel point corresponding to the jth gray value in the lung region.
In this embodiment, for any pixel, the preset second neighborhood of the pixel is a circle with the pixel as a center and the preset length as a radius, and the area except the pixel in the circular area is the preset second neighborhood of the pixel, where the preset length is 10 in this embodiment, and in specific applications, the implementer can set according to specific situations.
The number of the pixel points used for representing the j-th gray value in the neighborhood of the n-th pixel point corresponding to the j-th gray value is larger, which means that the more the pixel points in the neighborhood are consistent with the gray of the central pixel point. When the number of the pixel points of the j-th gray value in the neighborhood of each pixel point corresponding to the j-th gray value is larger, the pixel points of the j-th gray value are distributed more densely in the lung area, namely the more densely the j-th gray value in the lung area is.
By adopting the method, the probability that each pixel point in the lung area belongs to a blood vessel and the density degree of each gray value can be obtained.
And S4, reinforcing the lung region by combining the probability that each pixel point corresponding to each gray value belongs to a blood vessel and the corresponding density degree to obtain a reinforced lung image.
The probability that the pixel points belong to the blood vessel is evaluated from the structural characteristics of the blood vessel in the lung region and the gray distribution conditions of different structures, the class of the pixel points in the lung region is evaluated from the position distribution angle of the pixel points with the same gray level, the class of the pixel points is analyzed, and in order to ensure the enhancement effect of the lung CT image, the enhancement coefficient of each gray level is determined by combining the probability that the pixel points belong to the blood vessel and the intensity of each gray level, and then enhancement processing is carried out on the enhancement coefficient.
Specifically, combining the probability that all pixel points corresponding to each gray value belong to blood vessels and the corresponding intensity, and obtaining the enhancement coefficient of each gray value, wherein the probability and the intensity are in positive correlation with the enhancement coefficient. The positive correlation relationship indicates that the dependent variable increases with the increase of the independent variable, and the dependent variable decreases with the decrease of the independent variable, and the specific relationship may be a multiplication relationship, an addition relationship, an idempotent of an exponential function, or the like.
In this embodiment, the enhancement coefficient of the jth gray value in the lung field can be expressed as:
Wherein, An enhancement factor representing the jth gray value in the lung region,The number of pixels representing the j-th gray value in the lung field,Representing the probability that the nth pixel point corresponding to the jth gray value in the lung region belongs to a blood vessel,Representing the intensity of the j-th gray value in the lung field.
When the probability that the pixel points with the j-th gray level belong to blood vessels is larger and the pixel points with the same gray level are denser, the amplitude of the corresponding pixel points to be enhanced is larger, namely the enhancement coefficient of the j-th gray level in the lung area is larger.
By adopting the method, the enhancement coefficient of each gray value in the lung region can be obtained, and then the enhancement processing is respectively carried out on each gray value based on the enhancement coefficient.
Specifically, the product of each gray value and its enhancement coefficient in the lung region is recorded as the characteristic index of each gray value, and the minimum value of the characteristic index of each gray value and 255 is used as the target gray value corresponding to each gray value because the gray value of the pixel point takes the value of [0, 255 ]. By adopting the method, the target gray value corresponding to each gray value in the lung region can be obtained, the original gray value is replaced by the target gray value corresponding to each gray value, the replaced image is marked as the enhanced lung image, and the clear lung region image is obtained.
The enhanced image eliminates the interference of external factors such as equipment, environment and the like, and when the specific condition of the lung is analyzed later, the enhanced image is directly utilized to analyze the lung region, so that the accuracy of an analysis result is improved.
The gray scale condition of surrounding pixel points of each pixel point in the lung region is analyzed, and the possibility that the pixel point belongs to the pixel point on the blood vessel is evaluated by combining the gray scale value of the single pixel point, namely, the pixel point is primarily evaluated from the structural characteristics and the gray scale characteristics of the blood vessel in the lung region; because the lung is mutually matched by a plurality of tissues, the respiratory function of the lung is finished together, the gray values of the pixel points of the same tissue are close, the embodiment analyzes the density degree of the pixel points of each gray value, and then carries out enhancement treatment of different degrees on the pixel points of different gray values by combining the probability that the pixel points belong to blood vessels and the density degree of each gray value, the enhancement effect is better, the enhanced image is clearer, and the accuracy of the analysis effect can be ensured by analyzing the lung condition by using the enhanced image.
In other embodiments, a lung CT image adaptive enhancement system is provided, which includes a lung region extraction module, a feature analysis module, a feature calculation module, and an image enhancement module, as shown in fig. 4, which is a structural block diagram of the lung CT image adaptive enhancement system, where the lung region extraction module is configured to obtain a lung region in a lung CT image to be processed. The feature analysis module is used for analyzing the gray level change condition of each pixel point in the lung area and the pixel points in the neighborhood of the pixel point and determining the gray level change trend vector of each pixel point and the extension evaluation value of each pixel point; and obtaining a possibility index of each pixel point as a blood vessel pixel point according to the difference condition of the gray level change trend vector of each pixel point in the lung region and the gray level change trend vector of the pixel point in the preset nearest neighbor. The feature calculation module is used for determining the probability that each pixel belongs to a blood vessel by combining the extension evaluation value, the probability index and the gray value; the intensity of each gray value is determined based on the number of pixels of the same gray value in the neighborhood of each pixel corresponding to each gray value in the lung region. And the image enhancement module is used for enhancing the lung area by combining the probability that each pixel point corresponding to each gray value belongs to a blood vessel and the corresponding density degree to obtain an enhanced lung image. When the system is operated, the computer is caused to execute the related steps so as to realize the lung CT image self-adaptive enhancement method provided by the embodiment.
In other embodiments, an apparatus is also provided that includes a memory and a processor. The memory is used for storing executable program codes, and the processor is used for calling and running the executable program codes from the memory, so that the device executes the lung CT image self-adaption enhancement method. The apparatus may be embodied as a chip, component or module, which may include a processor and memory coupled together; the memory is used for storing instructions, and when the processor calls and executes the instructions, the chip can be made to execute the lung CT image self-adaptive enhancement method provided by the embodiment.
In other embodiments, a computer program product is provided, which when run on a computer causes the computer to perform the above-mentioned related steps to implement a lung CT image adaptive enhancement method provided by the above-mentioned embodiments.
In other embodiments, a computer readable storage medium is provided, in which a computer program code is stored, which when run on a computer causes the computer to perform the above-mentioned related method steps for implementing a lung CT image adaptive enhancement method provided in the above-mentioned embodiments.
The system, the apparatus, the computer program product, and the computer readable storage medium are all configured to perform the corresponding methods provided above, and therefore, the advantages achieved by the system, the apparatus, the computer program product, and the computer readable storage medium are referred to as the advantages of the corresponding methods provided above, and are not described herein.
It should be noted that: the foregoing description of the preferred embodiments of the present invention is not intended to be limiting, but rather, any modifications, equivalents, improvements, etc. that fall within the principles of the present invention are intended to be included within the scope of the present invention.
Claims (4)
1. A method for adaptively enhancing a lung CT image, comprising the steps of:
Acquiring a lung region in a lung CT image to be processed;
analyzing the gray scale change condition of each pixel point in the lung area and the pixel points in the neighborhood of the pixel point, and determining the gray scale change trend vector of each pixel point and the extension evaluation value of each pixel point; according to the difference condition of the gray level change trend vector of each pixel point in the lung area and the gray level change trend vector of the pixel point in the preset nearest neighbor, obtaining a possibility index that each pixel point is a vascular pixel point;
Determining the probability that each pixel belongs to a blood vessel by combining the extension evaluation value, the possibility index and the gray value; determining the density degree of each gray value based on the number of the pixel points with the same gray value in the neighborhood of each pixel point corresponding to each gray value in the lung region;
The probability that each pixel point corresponding to each gray value belongs to a blood vessel and the corresponding density degree are combined to strengthen the lung region so as to obtain an enhanced lung image;
The obtaining of the gray scale variation trend vector of each pixel point comprises the following steps:
for candidate pixel points, respectively calculating gray difference values between the candidate pixel points and each pixel point in the neighborhood of the candidate pixel points, determining the direction of a feature vector based on the sign of the gray difference values, and obtaining the feature vector of each pixel point in the neighborhood based on the absolute value of the gray difference values as the size of the feature vector;
The distribution of the feature vectors of all the pixel points in the neighborhood of the candidate pixel point is synthesized to determine the gray level change trend vector of the candidate pixel point; the candidate pixel points are any pixel point in the lung area;
the pixel extending evaluation value includes:
For candidate pixel points:
The two pixel points which are the closest to the candidate pixel point in the vertical direction of the gray level change trend vector of the candidate pixel point and pass through the straight line of the candidate pixel point are marked as characteristic points of the candidate pixel point; marking the pixel points except the characteristic points in the eight neighborhood of the candidate pixel point as reference points of the candidate pixel point;
Obtaining an extension evaluation value of the candidate pixel point according to the gray level difference between the candidate pixel point and the characteristic point and the reference point;
The method is characterized in that the extension evaluation value of the ith pixel point is calculated by adopting the following formula:
Wherein, An extended evaluation value representing the ith pixel point of the lung area,A gray value representing the ith pixel point of the lung field,A gray value representing a first feature point of an ith pixel point of the lung region,A gray value representing a second feature point of the ith pixel point of the lung area, U representing the number of reference points of the ith pixel point of the lung area,A gray value representing the ith reference point of the ith pixel point of the lung area,The sign of the absolute value is taken as the representation,Indicating that the first adjustment parameter is preset,Greater than 0; the feature points comprise a first feature point and a second feature point;
according to the difference condition of the gray level change trend vector of each pixel point in the lung region and the gray level change trend vector of the pixel point in the preset nearest neighbor, the obtaining of the probability index that each pixel point is a vascular pixel point comprises the following steps:
According to the difference between the gray level change trend vector of the candidate pixel point and the gray level change trend vector of the pixel point in the preset nearest neighbor of the candidate pixel point, obtaining a possibility index of the candidate pixel point as a blood vessel pixel point, wherein the difference of the gray level change trend vector and the possibility index are in a negative correlation;
the index of the possibility that the ith pixel point in the lung area is a vascular pixel point is expressed as follows:
Wherein, The probability index that the ith pixel point in the lung area is a vascular pixel point is represented, R represents the number of the pixel points in the preset nearest neighbor of the ith pixel point on the gray level change trend vector of the ith pixel point,The gray scale variation trend vector of the ith pixel point in the preset nearest neighbor of the ith pixel point is represented,The module length of the gray scale variation trend vector of the r pixel point in the preset nearest neighbor on the gray scale variation trend vector of the i pixel point is represented,A gray scale variation trend vector representing the i-th pixel point,The modulus of the gray scale variation trend vector representing the i-th pixel point,The included angle value between the gray scale change trend vector of the ith pixel point and the gray scale change trend vector of the r pixel point in the preset nearest neighbor of the gray scale change trend vector of the ith pixel point is represented,Representing the circumference ratio;
the step of obtaining the enhanced lung image by combining the probability that each pixel point corresponding to each gray value belongs to a blood vessel and the corresponding density degree to enhance the lung region comprises the following steps:
Combining the probability that all pixel points corresponding to each gray value belong to blood vessels and the corresponding intensity, and obtaining the enhancement coefficient of each gray value, wherein the probability and the intensity are in positive correlation with the enhancement coefficient;
each gray value in the lung region is enhanced based on the enhancement coefficient to obtain an enhanced lung image;
The step of enhancing each gray value in the lung region based on the enhancement coefficient to obtain an enhanced lung image comprises the following steps:
marking the product of each gray value and the enhancement coefficient of the gray value in the lung area as a characteristic index of each gray value, and taking the minimum value of the characteristic index of each gray value and 255 as a target gray value corresponding to each gray value;
And replacing the original gray value with the target gray value to obtain the enhanced lung image.
2. The method according to claim 1, wherein determining the probability that each pixel belongs to a blood vessel by combining the extension evaluation value, the likelihood index and the gray value comprises:
And obtaining the probability that the candidate pixel belongs to the blood vessel according to the extension evaluation value, the possibility index and the gray value of the candidate pixel, wherein the extension evaluation value, the possibility index and the gray value are in positive correlation with the probability.
3. The method according to claim 1, wherein determining the intensity of each gray value based on the number of pixels of the same gray value in the neighborhood of each pixel corresponding to each gray value in the lung region comprises:
For the candidate gray values, determining the density degree of the candidate gray values based on the number proportion of the pixel points of the candidate gray values in the neighborhood of each pixel point corresponding to the candidate gray values, wherein the number proportion and the density degree are in positive correlation;
the candidate gray value is any gray value in the lung area;
The intensity of the j-th gray value in the lung field is expressed as:
Wherein, Represents the intensity of the j-th gray value in the lung field,The number of pixels representing the j-th gray value in the lung field,Representing the number of pixels with the j-th gray value in the preset second neighborhood of the n-th pixel corresponding to the j-th gray value in the lung region,And representing the total number of the pixel points in the preset second neighborhood of the nth pixel point corresponding to the jth gray value in the lung region.
4. The method of claim 1, wherein the acquiring the lung region in the CT image of the lung to be processed comprises:
The pixel points in the CT image of the lung to be processed are divided into two types by adopting a K-means clustering algorithm, and the lung region is determined by combining the gray distribution condition of each type of pixel points.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410666336.5A CN118247277B (en) | 2024-05-28 | 2024-05-28 | Self-adaptive enhancement method for lung CT image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410666336.5A CN118247277B (en) | 2024-05-28 | 2024-05-28 | Self-adaptive enhancement method for lung CT image |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN118247277A CN118247277A (en) | 2024-06-25 |
| CN118247277B true CN118247277B (en) | 2024-07-26 |
Family
ID=91556613
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410666336.5A Active CN118247277B (en) | 2024-05-28 | 2024-05-28 | Self-adaptive enhancement method for lung CT image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118247277B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN119131039B (en) * | 2024-11-14 | 2025-02-11 | 武汉大学人民医院（湖北省人民医院） | Method and device for identifying crypt branches of inflammatory bowel disease |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115359052A (en) * | 2022-10-19 | 2022-11-18 | 南通鼎顺生物科技有限公司 | Medical image enhancement method based on clustering algorithm |
| CN117408929A (en) * | 2023-12-12 | 2024-01-16 | 日照天一生物医疗科技有限公司 | Tumor CT image area dynamic enhancement method based on image characteristics |
Family Cites Families (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2021217642A1 (en) * | 2020-04-30 | 2021-11-04 | 深圳市大疆创新科技有限公司 | Infrared image processing method and apparatus, and movable platform |
| CN111583157B (en) * | 2020-05-13 | 2023-06-02 | 杭州睿琪软件有限公司 | Image processing method, system and computer readable storage medium |
| WO2022021287A1 (en) * | 2020-07-31 | 2022-02-03 | 华为技术有限公司 | Data enhancement method and training method for instance segmentation model, and related apparatus |
| CN112184659B (en) * | 2020-09-24 | 2023-08-25 | 上海健康医学院 | Lung image processing method, device and equipment |
| CN117058033A (en) * | 2023-08-17 | 2023-11-14 | 吉林省珵康医疗科技有限公司 | Glomerular pathology image enhancement method based on kidney biopsy slice |
| CN117409004B (en) * | 2023-12-14 | 2024-03-15 | 自贡市第一人民医院 | An intelligent assistance system for pulmonary rehabilitation based on medical image analysis |
| CN117474822B (en) * | 2023-12-27 | 2024-03-08 | 江南大学附属医院 | Cardiovascular surgery image intelligent enhancement method |
| CN117474823B (en) * | 2023-12-28 | 2024-03-08 | 大连清东科技有限公司 | CT data processing system for pediatric infectious inflammation detection assistance |
| CN117853386B (en) * | 2024-03-08 | 2024-05-28 | 陕西省人民医院(陕西省临床医学研究院) | Tumor image enhancement method |
| CN118071754B (en) * | 2024-04-24 | 2024-07-19 | 胜利油田中心医院 | Thyroid ultrasound image data processing method |
-
2024
- 2024-05-28 CN CN202410666336.5A patent/CN118247277B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115359052A (en) * | 2022-10-19 | 2022-11-18 | 南通鼎顺生物科技有限公司 | Medical image enhancement method based on clustering algorithm |
| CN117408929A (en) * | 2023-12-12 | 2024-01-16 | 日照天一生物医疗科技有限公司 | Tumor CT image area dynamic enhancement method based on image characteristics |
Also Published As
| Publication number | Publication date |
|---|---|
| CN118247277A (en) | 2024-06-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US7474775B2 (en) | Automatic detection of red lesions in digital color fundus photographs | |
| JP5926937B2 (en) | Image processing apparatus, image processing method, and image processing program | |
| CN115359052B (en) | Medical image enhancement method based on clustering algorithm | |
| CN116109663B (en) | Segmentation method of gastric CT image based on multi-threshold segmentation | |
| CN117764864B (en) | Nuclear magnetic resonance tumor visual detection method based on image denoising | |
| CN118247277B (en) | Self-adaptive enhancement method for lung CT image | |
| CN117408988B (en) | Artificial intelligence-based focus image analysis method and apparatus | |
| CN117557460B (en) | Angiography image enhancement method | |
| CN117745751B (en) | Pulmonary tuberculosis CT image segmentation method based on feature extraction | |
| CN118197610B (en) | Auxiliary identification method and system for herpetic neuralgia based on structural magnetic resonance | |
| CN118172380B (en) | Orthopedics leg bone intelligent recognition segmentation method based on local threshold | |
| CN117974692B (en) | Ophthalmic medical image processing method based on region growing | |
| CN113940702B (en) | Thyroid nodule echo analysis device | |
| CN118314067B (en) | Tumor radio frequency accurate ablation system based on CT image | |
| CN108447066B (en) | Biliary tract image segmentation method, terminal and storage medium | |
| CN117541800B (en) | Laryngoscope image-based laryngeal anomaly segmentation method | |
| CN118447014B (en) | Barium meal contrast image focus identification system for auxiliary diagnosis of digestive system department | |
| US20050141758A1 (en) | Method, apparatus, and program for discriminating calcification patterns | |
| CN118212235B (en) | Capsule endoscope image screening method and system | |
| CN118134919B (en) | Rapid extraction method of hand bones for bone age identification | |
| CN119180822B (en) | Intelligent detection method of meniscus injury based on image processing | |
| CN117974645B (en) | Bronchoscope image processing method based on computer vision | |
| CN119006479B (en) | MRI image analysis method for lumbar vertebra lesion detection | |
| Gómez et al. | Finding regions of interest in pathological images: An attentional model approach | |
| Khan et al. | A Discriminative Color-to-Grayscale Representation for Retinal Vessel Segmentation. |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |