CN114972197A - X-ray film imaging quality evaluation method and system - Google Patents
X-ray film imaging quality evaluation method and system Download PDFInfo
- Publication number
- CN114972197A CN114972197A CN202210461714.7A CN202210461714A CN114972197A CN 114972197 A CN114972197 A CN 114972197A CN 202210461714 A CN202210461714 A CN 202210461714A CN 114972197 A CN114972197 A CN 114972197A
- Authority
- CN
- China
- Prior art keywords
- ray image
- abnormal
- imaging quality
- gray level
- histogram
- 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.)
- Pending
Links
Images
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
- G06T7/0012—Biomedical image inspection
-
- 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
- G06T2207/10—Image acquisition modality
- G06T2207/10116—X-ray image
-
- 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
- G06T2207/30168—Image quality inspection
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Radiology & Medical Imaging (AREA)
- Quality & Reliability (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Image Processing (AREA)
Abstract
The invention relates to the field of computer vision, in particular to an X-ray film imaging quality evaluation method and system, which comprises the following steps: acquiring a gray level histogram of an X-ray image to be evaluated, and acquiring a suspected abnormal X-ray image by using the distribution form of the histogram; obtaining an exposure abnormal value of the X-ray image by utilizing a gray level histogram of the suspected abnormal X-ray image; dividing the gray level histogram of the abnormal X-ray image, and obtaining the gray level richness of the abnormal X-ray image by utilizing the distribution of each gray level in each cluster in a division interval; obtaining the steepness of the abnormal X-ray image by using the highest peak and the left trough of the gray level histogram of the abnormal X-ray image; obtaining the contrast of the abnormal X-ray image by utilizing the gray richness and the steepness of the abnormal X-ray image; obtaining a quality abnormal value of the abnormal X-ray image by using the exposure abnormal value and the contrast; and evaluating the imaging quality of the X-ray image according to the quality abnormal value. The method is used for evaluating the imaging quality of the X-ray film, and can improve the evaluation efficiency.
Description
Technical Field
The invention relates to the field of computer vision, in particular to an X-ray film imaging quality evaluation method and system.
Background
The X-ray film is formed by that after X-rays pass through a human body, different tissues of each part of the human body have different absorption degrees on the X-rays, so that different density information of the X-rays can be displayed and distributed on a photosensitive film. When a medical image is taken, overexposure, underexposure and blurring of an X-ray film are often caused by light, shaking of a patient and the like, and a doctor is easy to make a misdiagnosis due to the problem of imaging quality when observing the X-ray film. Therefore, it is necessary to evaluate the imaging quality of the X-ray film.
The main observation currently used for quality assessment is by means of the human eye. However, this method is time-consuming, labor-consuming and inefficient. Therefore, the invention provides a method for improving the efficiency of the imaging quality evaluation of the X-ray film.
Disclosure of Invention
The invention provides an X-ray film imaging quality evaluation method and system, comprising the following steps: acquiring a gray level histogram of an X-ray image to be evaluated, and acquiring a suspected abnormal X-ray image by using the distribution form of the histogram; obtaining an exposure abnormal value of the X-ray image by utilizing a gray level histogram of the suspected abnormal X-ray image; dividing the gray level histogram of the abnormal X-ray image, and obtaining the gray level richness of the abnormal X-ray image by utilizing the distribution of each gray level in each cluster in a division interval; obtaining the steepness of the abnormal X-ray image by using the highest peak and the left trough of the gray level histogram of the abnormal X-ray image; obtaining the contrast of the abnormal X-ray image by utilizing the gray richness and the steepness of the abnormal X-ray image; obtaining a quality abnormal value of the abnormal X-ray image by using the exposure abnormal value and the contrast; compared with the prior art, the X-ray image quality evaluation method based on the image quality abnormal value has the advantages that the X-ray image is analyzed on the basis of computer vision, the exposure abnormal value and the contrast of the X-ray image with abnormal imaging quality are obtained by utilizing the gray scale characteristics of the X-ray image with abnormal imaging quality, the quality parameters of the X-ray image with abnormal imaging quality are further obtained, the imaging quality of the X-ray image is evaluated by utilizing the quality parameters of the X-ray image with abnormal imaging quality, and the imaging quality evaluation efficiency and accuracy of the X-ray film are effectively improved.
In order to achieve the above object, the present invention adopts the following technical solution, a method for evaluating imaging quality of an X-ray film, comprising:
and acquiring an X-ray image to be evaluated and a gray level histogram thereof.
Judging the distribution form of the gray levels in the gray level histogram: when the distribution of the gray levels in the gray level histogram is normal distribution, the X-ray image to be evaluated is an X-ray image with qualified imaging quality; and when the distribution of the gray levels in the gray level histogram is not normal, the X-ray image to be evaluated is an X-ray image with abnormal suspected imaging quality.
Judging whether the X-ray image has an abnormal exposure phenomenon or not by utilizing the gray level in the gray level histogram of the X-ray image with abnormal suspected imaging quality: when the abnormal exposure phenomenon does not exist in the X-ray image, the X-ray image is the X-ray image with qualified imaging quality; when the abnormal exposure phenomenon exists in the X-ray image, the X-ray image is the X-ray image with abnormal imaging quality, and the abnormal exposure value of the X-ray image with abnormal imaging quality is calculated.
And clustering the X-ray images with abnormal imaging quality, and segmenting the gray level of the gray level histogram of the X-ray images by using the obtained clustering cluster to obtain all segmentation intervals.
And calculating the richness of the gray levels of the X-ray image with abnormal imaging quality by using the distribution ratio of each gray level in the segmentation interval in each cluster and the number of the clusters.
Calculating the steepness of the gray level histogram of the X-ray image with abnormal imaging quality by utilizing the gray levels of the highest peak and the left trough in the gray level histogram of the X-ray image with abnormal imaging quality and the number of pixel points.
And calculating the contrast of the X-ray image with abnormal imaging quality by utilizing the richness of the gray level of the X-ray image with abnormal imaging quality and the steepness of the gray level histogram of the X-ray image.
And calculating the quality abnormal parameter of the X-ray image with abnormal imaging quality by using the exposure abnormal value and the contrast of the X-ray image with abnormal imaging quality.
And evaluating the imaging quality of the X-ray image according to the quality abnormity parameters of the X-ray image with abnormal imaging quality.
Further, according to the method for evaluating the imaging quality of the X-ray film, the abnormal exposure value of the X-ray image with abnormal imaging quality is calculated as follows:
and calculating to obtain the sum of all slopes between the gray level of the wave trough on the right side of the highest peak in the histogram and 255 by using the adjacent gray levels between the wave trough on the right side of the highest peak in the gray histogram of the suspected imaging quality abnormal X-ray image and the number of the pixels of the adjacent gray levels.
And (3) judging the sum of all slopes between the gray level of the trough on the right side of the highest peak and the gray level of the trough on the right side of the highest peak in the histogram to 255: when the sum of the slopes is greater than or equal to 0, the X-ray image with the suspected abnormal imaging quality is an X-ray image with the abnormal imaging quality, and an overexposure value of the X-ray image with the abnormal imaging quality is calculated; and when the sum of the slopes is less than 0, judging whether the X-ray image with abnormal suspected imaging quality has an overexposure phenomenon or not and judging whether the X-ray image with abnormal suspected imaging quality has an underexposure phenomenon or not.
The process of calculating the overexposure value of the X-ray image with abnormal imaging quality is as follows: and calculating the ratio of the truncation area of the curve between the gray level of the trough on the right side of the highest peak and the trough of 255 in the histogram to the adaptive rectangle to obtain the overexposure value of the X-ray image with abnormal imaging quality.
The process of judging whether the under-exposure phenomenon exists in the X-ray image with abnormal suspected imaging quality is as follows: and calculating to obtain the sum of all slopes between 0 and the valley gray level on the left side of the first peak in the histogram by utilizing the adjacent gray levels between 0 and the valley gray level on the left side of the first peak in the histogram and the number of pixel points of the adjacent gray levels.
Judging the sum of all slopes between the gray levels of the wave troughs from 0 to the left side of the first wave peak in the histogram: when the sum of the slopes is less than or equal to 0, the X-ray image with the suspected imaging quality abnormality is an X-ray image with the suspected imaging quality abnormality, and the ratio of the truncation area of a curve from 0 to the valley gray level at the left side of the first peak to the adaptive rectangle thereof is calculated to obtain an underexposure value; and when the sum of the slopes is greater than 0, the X-ray image with the suspected abnormal imaging quality is an X-ray image with qualified imaging quality, and an exposure abnormal value of the X-ray image with the abnormal imaging quality is calculated.
Further, in the method for evaluating the imaging quality of the X-ray film, the expression of the overexposure value of the X-ray image with abnormal imaging quality is as follows:
wherein B is the overexposure value of the X-ray image with abnormal imaging quality,
is the cut-off area of the curve between the gray level of the trough on the right side of the highest peak in the histogram and 255, f (I) is the function of the curve between the gray level of the trough on the right side of the highest peak in the histogram and 255, I is the gray level between the gray level of the trough on the right side of the highest peak in the histogram and 255, I T Is the gray level of the trough on the right side of the highest peak in the histogram, G jd The local maximum value of the number of the pixel points in the curve between the gray level of the trough on the right side of the highest peak in the histogram and 255.
Further, in the method for evaluating imaging quality of an X-ray film, the richness of the gray levels of the X-ray image with abnormal imaging quality is obtained as follows:
and obtaining the optimal K value of the X-ray image with abnormal imaging quality by using an elbow rule.
And clustering the X-ray images by using the optimal K value to obtain K clustering clusters of the X-ray images.
And calculating the average gray value of each cluster, and sequencing the K clusters according to the mode that the average gray value is from small to large.
And dividing the gray level of the gray level histogram into K intervals according to the ratio of the pixels in the sequenced K clustering clusters.
And obtaining the distribution ratio of each gray level in each cluster by using the number of the pixel points corresponding to each gray level in each interval of the histogram in each cluster.
And calculating the richness of the gray levels of the X-ray image with abnormal imaging quality by using the distribution ratio of each gray level in each cluster and the number of clusters.
Further, in the method for evaluating imaging quality of an X-ray film, the expression of steepness of a gray level histogram of an X-ray image with abnormal imaging quality is as follows:
wherein DQ is steepness of gray level histogram of X-ray image with abnormal imaging quality, G max Is the highest wave in the histogramNumber of pixels of peak, G min The number of pixels of the left trough of the highest peak in the histogram, I max Is the gray level of the highest peak in the histogram, I min The gray level of the trough on the left of the highest peak in the histogram.
Further, in the method for evaluating the imaging quality of the X-ray film, the expression of the quality abnormality parameter of the X-ray image with abnormal imaging quality is as follows:
wherein ZL is the quality abnormal parameter of the X-ray image with abnormal imaging quality, beta is the exposure abnormal value of the X-ray image with abnormal imaging quality, DB is the contrast of the X-ray image with abnormal imaging quality, and omega 1 、ω 2 Are weights.
Further, in the method for evaluating the imaging quality of the X-ray film, the process of evaluating the imaging quality of the X-ray image is specifically as follows:
setting a threshold value, and judging the quality abnormal parameters of the X-ray image with abnormal imaging quality: when the quality abnormal parameter of the X-ray image with abnormal imaging quality is less than or equal to the threshold value, the imaging quality of the X-ray image is good; when the quality abnormal parameter of the X-ray image with abnormal imaging quality is larger than the threshold value, the imaging quality of the X-ray image is poor.
The invention also provides an imaging quality evaluation system of the X-ray film, which comprises an acquisition unit, a processing unit, a calculation unit and an evaluation unit:
the acquisition unit acquires images of all parts of the human body by using an X-ray machine.
And the processing unit is used for processing the image acquired by the acquisition unit by the computer, acquiring a gray level histogram of the X-ray image to be evaluated, and acquiring the X-ray image with abnormal suspected imaging quality according to the distribution condition of the gray level histogram.
And the computing unit is used for obtaining the X-ray image with abnormal imaging quality by the computer according to the gray scale characteristics of the X-ray image with abnormal imaging quality obtained by the processing unit, and computing the exposure abnormal value and the contrast of the X-ray image with abnormal imaging quality so as to obtain the quality abnormal parameters of the X-ray image with abnormal imaging quality.
And the evaluation unit is used for evaluating the imaging quality of the X-ray image by the computer according to the quality abnormity parameters of the X-ray image with abnormal imaging quality obtained by the calculation unit.
The invention has the beneficial effects that:
according to the X-ray image quality evaluation method, the X-ray image is analyzed on the basis of computer vision, the exposure abnormal value and the contrast of the X-ray image with abnormal imaging quality are obtained by utilizing the gray characteristic of the X-ray image with abnormal imaging quality, the quality parameter of the X-ray image with abnormal imaging quality is further obtained, the imaging quality of the X-ray image is evaluated by utilizing the quality parameter of the X-ray image with abnormal imaging quality, and the imaging quality evaluation efficiency and accuracy of the X-ray film are effectively improved.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the embodiments or the description of the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
Fig. 1 is a schematic flow chart of an X-ray film imaging quality evaluation method according to embodiment 1 of the present invention;
fig. 2 is a schematic flow chart of an X-ray film imaging quality evaluation method according to embodiment 2 of the present invention;
fig. 3 is a schematic diagram of a gray level histogram of an X-ray film with qualified imaging quality according to embodiment 2 of the present invention;
FIG. 4 is a schematic diagram of an overexposed X-ray film provided in embodiment 2 of the present invention;
FIG. 5 is a schematic diagram of an X-ray film with normal brightness according to embodiment 2 of the present invention;
FIG. 6 is a graph showing the relationship between the "sum of distances" and K for K values according to embodiment 2 of the present invention;
fig. 7 is a block diagram of a flow chart of an imaging quality evaluation system for an X-ray film according to embodiment 3 of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Example 1
The embodiment of the invention provides an imaging quality evaluation method of an X-ray film, as shown in fig. 1, comprising the following steps:
s101, obtaining an X-ray image to be evaluated and a gray level histogram of the X-ray image.
The gray level histogram is a function of gray level distribution, and is a statistic of gray level distribution in an image.
S102, judging the distribution form of the gray levels in the gray level histogram: when the distribution of the gray levels in the gray level histogram is normal distribution, the X-ray image to be evaluated is an X-ray image with qualified imaging quality; and when the distribution of the gray levels in the gray level histogram is not normal, the X-ray image to be evaluated is an X-ray image with abnormal suspected imaging quality.
The normal X-ray film histogram should present a normal distribution similar to the middle symmetry except for the background region, and the middle is high and the two sides are low.
S103, judging whether the X-ray image has an abnormal exposure phenomenon or not by utilizing the gray level in the gray level histogram of the X-ray image with abnormal suspected imaging quality: when the abnormal exposure phenomenon does not exist in the X-ray image, the X-ray image is the X-ray image with qualified imaging quality; when the abnormal exposure phenomenon exists in the X-ray image, the X-ray image is the X-ray image with abnormal imaging quality, and the abnormal exposure value of the X-ray image with abnormal imaging quality is calculated.
Wherein the exposure abnormal value comprises an overexposure value and an underexposure value.
And S104, clustering the X-ray images with abnormal imaging quality, and segmenting the gray level of the gray level histogram of the X-ray images by using the obtained clustering cluster to obtain all segmentation intervals.
The process of separating a collection of physical or abstract objects into classes composed of similar objects is referred to herein as clustering.
And S105, calculating the gray level richness of the X-ray image with abnormal imaging quality by using the distribution ratio of each gray level in each cluster in the segmentation interval and the number of the clusters.
Wherein, the better the richness of the image gray level is, the more beneficial the image recognition is.
S106, calculating the steepness of the gray level histogram of the X-ray image with abnormal imaging quality by utilizing the gray levels of the highest peak and the left trough in the gray level histogram of the X-ray image with abnormal imaging quality and the number of pixel points.
The greater the steepness of the histogram, the lower the original image contrast.
S107, calculating the contrast of the X-ray image with abnormal imaging quality by utilizing the richness of the gray level of the X-ray image with abnormal imaging quality and the steepness of the gray level histogram of the X-ray image.
Wherein the contrast is used to calculate a quality anomaly parameter.
And S108, calculating the quality abnormal parameter of the X-ray image with abnormal imaging quality by using the exposure abnormal value and the contrast ratio of the X-ray image with abnormal imaging quality.
Wherein the quality anomaly parameter is used for quality assessment.
And S109, evaluating the imaging quality of the X-ray image according to the quality abnormity parameters of the X-ray image with abnormal imaging quality.
Setting a threshold value, and comparing the relation between the quality abnormal parameter and the threshold value to evaluate the imaging quality of the X-ray image.
The beneficial effect of this embodiment lies in:
according to the method and the device, the X-ray image is analyzed on the basis of computer vision, the abnormal exposure value and the contrast of the X-ray image with abnormal imaging quality are obtained by utilizing the gray scale characteristics of the X-ray image with abnormal imaging quality, the quality parameters of the X-ray image with abnormal imaging quality are further obtained, the imaging quality of the X-ray image is evaluated by utilizing the quality parameters of the X-ray image with abnormal imaging quality, and the imaging quality evaluation efficiency and accuracy of the X-ray film are effectively improved.
Example 2
The embodiment of the invention provides an imaging quality evaluation method of an X-ray film, as shown in FIG. 2, comprising the following steps:
s201, collecting an image and segmenting the image by utilizing a neural network.
Collecting an X-ray film of a human vertebra, and performing semantic segmentation on the X-ray film of the vertebra through a large amount of neural network training:
1. a CNN network is adopted, the network is of an Encoder-Decoder structure, and a data set is divided into a training set and a test set according to the proportion of 7 to 3.
2. And performing edge tracing processing on a target area of each X-ray image, manually marking a target pixel point as 1, and marking other areas as 0.
3. The loss function used by the network is a cross entropy loss function.
For identifying the image with poor screening by a machine, the most important is to calculate the interference degree or abnormal degree of the target pixel region in the whole image, and the embodiment combines the characteristics of the gray histogram to construct an image quality abnormal calculation model from two basic parameters of the brightness and the contrast of the image.
S202, establishing a histogram and calculating an exposure abnormal value.
The image gray level histogram is the distribution of all pixel points of an image based on gray levels, and the richness of the image gray level distribution can be clearly obtained by constructing the histogram. The denser the spacing of the gray levels, the more similar the gray levels are.
FIG. 3 is a gray level histogram of an X-ray film with acceptable imaging quality, connecting the vertices of each term in the histogram, and smoothing to obtain a relief curve. The change in the curve represents the change in the ratio of different gray levels in the image.
Calculating image exposure anomaly:
when a hospital takes an X-ray film, a doctor needs to control ambient light in a room and informs that a patient cannot wear a metal object to avoid reflecting light to damage imaging, and sometimes, because exposure parameters of imaging equipment are set incorrectly, images are overexposed and underexposed.
The overexposure and the underexposure of the image can cause key information loss, which affects the diagnosis of doctors, and fig. 4 is an overexposed X-ray film, and fig. 5 is a normal X-ray film.
Normal X-ray film histograms should exhibit a normal distribution similar to central symmetry, high in the middle and low on both sides, except for the background region. The left image with the middle wave crest is darker overall, which is called underexposure, and the right image is too bright and loses a large amount of information, which is called overexposure. However, in many cases, for example, in an X-ray film of a lung shadow, the distribution of highlight gray scales is more, the peak value naturally deviates to the right, the background black part is too much, and the peak value naturally deviates to the left, so that the peak position cannot be seen.
When the X-ray signal passes through the human body, it will gradually attenuate, and the high gray value presented on the bone will not reach the highest, that is, will not be gathered at 255 on the gray histogram, and this embodiment considers that:
when the highest peak of the histogram curve is deviated to the right, and when the right side of the histogram curve has a trough and does not continuously rise any more, the image is exposed in a normal range; when the right side has no wave trough, the gliding process is directly cut off by I-255, or the curve starts from the wave trough again and sharply increases to the right, and the image has an overexposure phenomenon.
For the histogram of any X-ray film, firstly calculating the sum a of all slopes between the gray level of a trough on the right side of the highest peak in the histogram and 255:
the above formula a represents the right trough gray level I T Sum of all slopes between to 255, I n Representing right trough gray level I T To an nth gray level of 255, I n-1 Representing right trough gray level I T To the (n-1) th gray level between 255, G n Representing right trough gray level I T Number of pixels corresponding to nth gray level between 255, G n-1 Representing right trough gray level I T And the number of pixel points corresponding to the (n-1) th gray level from 255. I denotes the right valley gray level I T To 255 gray levels.
When a is more than or equal to 0, the histogram curve shows incremental increase after the wave trough, namely, the overexposure phenomenon exists in the X-ray film, the overexposure value needs to be calculated, and the overexposure value calculation process is as follows:
the gray scale of the overexposure part can directly cover the original image, and the overexposure part is uniformly decreased from the center to the periphery, so that the gray scale around 255 has almost no interval, the ratio of the cross-sectional area of the curve to the adaptive rectangle is calculated, namely the overexposure value can be expressed as highlight, and the length of the adaptive rectangle is 255-I T Width G of jd ，G jd Is the local maximum of the number of gray scale pixels in the segment. Then the overexposure value is given as:
b is the value of the over-exposure,
is the cross-sectional area of the curve for that segment.
When a is less than 0, the X-ray film may be a normal image or has an underexposure phenomenon, the sum of all slopes between the gray level 0 and the gray level of the trough on the left side of the first peak in the histogram is calculated, and whether the underexposure exists is judged:
the image under-exposure appears as a low overall gray level, and the background area of the X-ray film is black, but its pixel value does not actually reach I-0.
Like overexposure, when the peak of the histogram curve is shifted to the left and there is no valley on the left of the first peak, but the first peak is directly truncated by I ═ 0, at this time, the image is underexposed, and when there is a valley on the left, the image is normal. The algorithm implementation is the same as overexposure.
Therefore, when the sum of all slopes between the gray level 0 and the gray level of the trough on the left side of the first peak is less than or equal to 0, the underexposure phenomenon is detected, and the ratio of the cross-sectional area of the curve between the gray level 0 and the gray level of the trough on the left side of the first peak to the adaptive rectangle thereof is calculated to obtain the underexposure value C'.
I V Gray level of trough on left side of first peak, G jD The local maximum value of the number of the pixel points from 0 to the valley gray level on the left side of the first peak in the histogram.
When the sum of all slopes from the gray level 0 to the gray level of the valley on the left side of the first peak is greater than 0, the phenomenon is normal.
Thus, an exposure abnormal value β is obtained. When overexposed, β ═ B, and when underexposed, β ═ C'.
And S203, calculating the contrast of the image according to the gray level histogram.
In addition to brightness, an X-ray film with good imaging quality must have rich gray levels, and then the gray levels have obvious gray level differences to form vivid image textures.
1. Evaluating the richness of the image, if only the number and probability of all gray levels are calculated are limited, because the visual effects of adjacent gray levels on the image have similarity, calculating all gray levels cannot reflect how much difference exists between the gray levels, and cannot be used for describing the richness of the image.
In this embodiment, based on K-Means clustering, it is proposed to classify the gray levels of the original X-ray image by calculating the K value, that is, similar gray levels are grouped together, and the distribution of the gray levels in these different regions shows the richness of the gray levels.
The K value is achieved by the elbow method. Selecting different K values, and drawing a relation graph of the sum of distances and K of each K value, as shown in FIG. 6. And calculating K to be 2, 3, 4 and … to obtain the functional relation graph, and selecting the K value as the optimal value when the sum of the distances between all points in each cluster and the center point of the cluster is changed from large reduction to stable reduction. And performing K-Means clustering on the X-ray image to obtain K gray scale regions.
And arranging the average gray values of each gray area from small to large.
In the histogram, according to the pixel ratio in the region divided by the K value, the horizontal axis gray level of the histogram is divided into K sections (Q) 1 ，Q 2 ，Q 3 …Q K ) There is a significant difference in gray level between each region divided by K. The gray levels in each divided region are all the same kind (similar) gray levels. All the gray values of the X-ray image are scattered in each interval.
And calculating the ratio of each gray level in each segmentation region, wherein the pixel point set of the X-ray image is W', the pixel points have N gray levels, N is less than or equal to 255, and the pixel points are distributed on the gray level of the histogram and are also divided into K parts by K.
ω q Representing the distribution ratio, H, of each gray level in the qth gray scale region q Expressing the number of pixel points corresponding to each gray level in the qth gray level region, C q And expressing the quantity of all pixel points in the qth gray scale region.
The sum of all omega is 1, each gray scale area is the gray scale of the same type, the more even the distribution of the gray scale areas is in each gray scale area, the better the richness is represented, and the difference between image layers is larger. I.e. the variance is calculated for the set of data:
the smaller σ, the better the image gray level richness. K represents the number of gray areas.
2. The image richness is reflected in the last step, the contrast can be really reflected by combining the difference between the gray levels, and the contrast cannot be really reflected by the X-ray film of the medical image because the threshold is difficult to set only aiming at the gray level difference, and most of the X-ray film has subjectivity.
The present example considers that: in a normal gray level histogram of an X-ray film, a background gray level is removed, a middle aggregation should be presented, normal distribution descending towards two ends is provided, peaks and troughs of main peaks of the normal X-ray film should simultaneously satisfy that a difference value of pixel numbers is small, a difference value of gray levels is large, and a measurement parameter, namely steepness, can be obtained by integrating two points, and a calculation formula of the measurement parameter is as follows:
DQ is a steepness parameter, G max -G min Is from trough to crest, i.e. the difference between the number of pixels of crest and trough on the left side, I max -I min Is the difference in gray level between the peak and the trough.
The shorter the distance of the slope per degree, that is, the smaller the DQ value, the higher the sharpness and the worse the image quality.
The formula represents the meaning: the larger the slope between two points is, the longer the oblique side is, the larger the steepness of the histogram is, and the smaller the original image contrast is, so the steepness can be used as an abnormal parameter to evaluate the image contrast.
3. Good images have rich gray levels, the gray levels have obvious gray level difference, and the contrast ratio abnormal model is constructed by taking the poor images as the contrast ratio abnormal model:
DB＝We -σ e -DQ
DB is a contrast anomaly parameter, sigma represents richness of image gray level, and DQ represents steepness of gray level. The smaller both parameters are, e -σ And e -DQ The larger the value is in the range of 0-1, the larger the value is, W is a constant, and the value is adjusted to be in the range of DB. If W is 1, the DB value approaches 1 as the image quality is worse.
And S204, combining the exposure abnormal value and the contrast to construct a calculation model of quality abnormality.
ZL is a quality abnormal parameter, beta represents exposure abnormal, and DB is a contrast abnormal parameter.
ω 1 、ω 2 For weighting, the formula is set by self, and the relation between contrast and exposure abnormity is constructed, because whether overexposure or underexposure abnormity occurs, the contrast abnormity can be influenced, in short, an image is too bright or too dark, the gray level of the image is higher, the gray level of the image is lower, and the contrast becomes lower. Values of DB and beta are 0-1, and ZL describes a correlation coefficient between the DB and the beta, namely the bigger ZL is, the greater the correlation between the DB and the beta is, the more unstable the image is, and the image quality is difficult to repair by adjusting parameters; the smaller ZL is, the parallel deviation of the two parameters is realized, the more stable the image is, and the image with good quality can be obtained only by adjusting the brightness or the contrast in a single direction.
And S205, evaluating the imaging quality of the X-ray film according to the quality abnormal parameters.
And setting a threshold value &, and when the quality abnormal parameter is greater than the threshold value, the imaging quality of the X-ray film is poor, the X-ray film cannot be used and needs to be deleted.
The beneficial effect of this embodiment lies in:
according to the method and the device, the X-ray image is analyzed on the basis of computer vision, the abnormal exposure value and the contrast of the X-ray image with abnormal imaging quality are obtained by utilizing the gray scale characteristics of the X-ray image with abnormal imaging quality, the quality parameters of the X-ray image with abnormal imaging quality are further obtained, the imaging quality of the X-ray image is evaluated by utilizing the quality parameters of the X-ray image with abnormal imaging quality, and the imaging quality evaluation efficiency and accuracy of the X-ray film are effectively improved.
Example 3
The embodiment of the invention provides an imaging quality evaluation system of an X-ray film, which comprises an acquisition unit, a processing unit, a calculation unit and an evaluation unit, as shown in FIG. 7:
the acquisition unit acquires images of all parts of the human body by using an X-ray machine;
the processing unit is used for constructing a gray level histogram of the X-ray image to be evaluated by the computer according to the gray level value of each pixel point in the image acquired by the acquisition unit and acquiring the X-ray image suspected of abnormal imaging quality according to the distribution condition of the gray level histogram;
the computing unit is used for obtaining the X-ray image with abnormal imaging quality by the computer according to the gray scale characteristics of the X-ray image with abnormal suspected imaging quality obtained by the processing unit, computing the exposure abnormal value and the contrast of the X-ray image with abnormal imaging quality and further obtaining the quality abnormal parameter of the X-ray image with abnormal imaging quality by utilizing the exposure abnormal value and the contrast;
and the evaluation unit is used for finishing the evaluation of the imaging quality of the X-ray image by judging the size relationship between the quality abnormal parameter of the X-ray image with abnormal imaging quality obtained by the calculation unit and the threshold value.
The beneficial effect of this embodiment lies in:
according to the method and the device, the X-ray image is analyzed on the basis of computer vision, the abnormal exposure value and the contrast of the X-ray image with abnormal imaging quality are obtained by utilizing the gray scale characteristics of the X-ray image with abnormal imaging quality, the quality parameters of the X-ray image with abnormal imaging quality are further obtained, the imaging quality of the X-ray image is evaluated by utilizing the quality parameters of the X-ray image with abnormal imaging quality, and the imaging quality evaluation efficiency and accuracy of the X-ray film are effectively improved.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like that fall within the spirit and principle of the present invention are intended to be included therein.
Claims (8)
1. An imaging quality evaluation method for an X-ray film, comprising:
acquiring an X-ray image to be evaluated and a gray level histogram thereof;
judging the distribution form of the gray levels in the gray level histogram: when the distribution of the gray levels in the gray level histogram is normal distribution, the X-ray image to be evaluated is an X-ray image with qualified imaging quality; when the distribution of the gray levels in the gray level histogram is not normal, the X-ray image to be evaluated is an X-ray image with abnormal suspected imaging quality;
judging whether the X-ray image has an abnormal exposure phenomenon or not by utilizing the gray level in the gray level histogram of the X-ray image with abnormal suspected imaging quality: when the abnormal exposure phenomenon does not exist in the X-ray image, the X-ray image is the X-ray image with qualified imaging quality; when the abnormal exposure phenomenon exists in the X-ray image, the X-ray image is the X-ray image with abnormal imaging quality, and the abnormal exposure value of the X-ray image with abnormal imaging quality is calculated;
clustering the X-ray images with abnormal imaging quality, and segmenting the gray level of the gray level histogram of the X-ray images by using the obtained clustering cluster to obtain all segmentation intervals;
calculating the gray level richness of the X-ray image with abnormal imaging quality by using the distribution ratio of each gray level in the segmentation interval in each cluster and the number of the clusters;
calculating the steepness of the gray level histogram of the X-ray image with abnormal imaging quality by utilizing the gray levels of the highest peak and the left trough in the gray level histogram of the X-ray image with abnormal imaging quality and the number of pixel points;
calculating the contrast of the X-ray image with abnormal imaging quality by utilizing the richness of the gray level of the X-ray image with abnormal imaging quality and the steepness of the gray level histogram of the X-ray image;
calculating the quality abnormal parameter of the X-ray image with abnormal imaging quality by using the exposure abnormal value and the contrast of the X-ray image with abnormal imaging quality;
and evaluating the imaging quality of the X-ray image according to the quality abnormity parameters of the X-ray image with abnormal imaging quality.
2. The method as claimed in claim 1, wherein the abnormal exposure value of the X-ray image with abnormal imaging quality is calculated as follows:
calculating the sum of all slopes between the gray level of the wave trough on the right side of the highest peak in the histogram and 255 by using the adjacent gray levels between the wave trough on the right side of the highest peak in the gray level histogram of the suspected imaging quality and the number of the pixels of the adjacent gray levels;
and (3) judging the sum of all slopes between the gray level of the trough on the right side of the highest peak and the gray level of the trough on the right side of the highest peak in the histogram to 255: when the sum of the slopes is greater than or equal to 0, the X-ray image with the suspected abnormal imaging quality is an X-ray image with the abnormal imaging quality, and an overexposure value of the X-ray image with the abnormal imaging quality is calculated; when the sum of the slopes is less than 0, judging whether the X-ray image with abnormal suspected imaging quality has an overexposure phenomenon or not and judging whether the X-ray image with abnormal suspected imaging quality has an underexposure phenomenon or not;
the process of calculating the overexposure value of the X-ray image with abnormal imaging quality is as follows: calculating the ratio of the truncation area of the curve between the gray level of the trough on the right side of the highest peak and the gray level of the trough on the right side of the histogram to the adaptive rectangle of the curve to obtain the overexposure value of the X-ray image with abnormal imaging quality;
the process of judging whether the under-exposure phenomenon exists in the X-ray image with abnormal suspected imaging quality is as follows: calculating the sum of all slopes between 0 and the valley gray level at the left side of the first peak in the histogram by utilizing the adjacent gray levels between 0 and the valley gray level at the left side of the first peak in the histogram and the number of pixel points of the adjacent gray levels;
judging the sum of all slopes between 0 and the valley gray level on the left side of the first peak in the histogram: when the sum of the slopes is less than or equal to 0, the X-ray image with the suspected abnormal imaging quality is an X-ray image with the abnormal imaging quality, and the ratio of the truncation area of the curve between 0 and the valley gray level at the left side of the first peak to the adaptive rectangle thereof is calculated to obtain an underexposure value; and when the sum of the slopes is greater than 0, the X-ray image with the suspected abnormal imaging quality is an X-ray image with qualified imaging quality, and an exposure abnormal value of the X-ray image with the abnormal imaging quality is calculated.
3. The method of claim 2, wherein the expression of the overexposure value of the X-ray image with abnormal imaging quality is as follows:
wherein B is the overexposure value of the X-ray image with abnormal imaging quality,
is the cut-off area of the curve between the gray level of the trough on the right side of the highest peak in the histogram and 255, f (I) is the function of the curve between the gray level of the trough on the right side of the highest peak in the histogram and 255, I is the gray level between the gray level of the trough on the right side of the highest peak in the histogram and 255, I T Is the gray level of the trough on the right side of the highest peak in the histogram, G jd The local maximum value of the number of the pixel points in the curve between the gray level of the trough on the right side of the highest peak in the histogram and 255.
4. The method as claimed in claim 1, wherein the gray-scale richness of the X-ray image with abnormal imaging quality is obtained as follows:
obtaining the optimal K value of the X-ray image with abnormal imaging quality by utilizing an elbow rule;
clustering the X-ray images by using the optimal K value to obtain K clustering clusters of the X-ray images;
calculating the average gray value of each cluster, and sequencing the K clusters according to the mode that the average gray value is from small to large;
dividing the gray level of the gray level histogram into K intervals according to the ratio of the pixels in the sequenced K clustering clusters;
obtaining the distribution ratio of each gray level in each cluster by using the number of the pixel points corresponding to each gray level in each interval of the histogram in each cluster;
and calculating the richness of the gray levels of the X-ray image with abnormal imaging quality by using the distribution ratio of each gray level in each cluster and the number of clusters.
5. The method as claimed in claim 1, wherein the expression of steepness of a gray level histogram of an X-ray image with abnormal imaging quality is as follows:
wherein DQ is steepness of gray level histogram of X-ray image with abnormal imaging quality, G max The number of pixels with the highest peak in the histogram, G min The number of pixels of the left trough of the highest peak in the histogram, I max Is the gray level of the highest peak in the histogram, I min The gray level of the trough on the left of the highest peak in the histogram.
6. The method as claimed in claim 1, wherein the expression of the abnormal quality parameter of the X-ray image with abnormal imaging quality is as follows:
wherein ZL is the quality abnormal parameter of the X-ray image with abnormal imaging quality, beta is the exposure abnormal value of the X-ray image with abnormal imaging quality, DB is the contrast of the X-ray image with abnormal imaging quality, and omega 1 、ω 2 Are weights.
7. The method as claimed in claim 1, wherein the process of evaluating the imaging quality of the X-ray image comprises the following steps:
setting a threshold value, and judging the quality abnormal parameters of the X-ray image with abnormal imaging quality: when the quality abnormal parameter of the X-ray image with abnormal imaging quality is less than or equal to the threshold value, the imaging quality of the X-ray image is good; when the quality abnormal parameter of the X-ray image with abnormal imaging quality is larger than the threshold value, the imaging quality of the X-ray image is poor.
8. An imaging quality evaluation system of an X-ray film is characterized by comprising an acquisition unit, a processing unit, a calculation unit and an evaluation unit:
the acquisition unit acquires images of all parts of the human body by using an X-ray machine;
the processing unit is used for processing the image acquired by the acquisition unit by the computer, acquiring a gray level histogram of the X-ray image to be evaluated, and acquiring the X-ray image with abnormal suspected imaging quality according to the distribution condition of the gray level histogram;
the computing unit is used for obtaining the X-ray image with abnormal imaging quality by the computer according to the gray scale characteristics of the X-ray image with abnormal suspected imaging quality obtained by the processing unit, and computing the exposure abnormal value and the contrast of the X-ray image with abnormal imaging quality so as to obtain the quality abnormal parameters of the X-ray image with abnormal imaging quality;
and the evaluation unit is used for evaluating the imaging quality of the X-ray image by the computer according to the quality abnormity parameters of the X-ray image with abnormal imaging quality obtained by the calculation unit.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210461714.7A CN114972197A (en) | 2022-04-28 | 2022-04-28 | X-ray film imaging quality evaluation method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210461714.7A CN114972197A (en) | 2022-04-28 | 2022-04-28 | X-ray film imaging quality evaluation method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114972197A true CN114972197A (en) | 2022-08-30 |
Family
ID=82978704
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210461714.7A Pending CN114972197A (en) | 2022-04-28 | 2022-04-28 | X-ray film imaging quality evaluation method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114972197A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115311283A (en) * | 2022-10-12 | 2022-11-08 | 山东鲁玻玻璃科技有限公司 | Glass tube drawing defect detection method and system |
| CN116823808A (en) * | 2023-08-23 | 2023-09-29 | 青岛豪迈电缆集团有限公司 | Intelligent detection method for cable stranded wire based on machine vision |
-
2022
- 2022-04-28 CN CN202210461714.7A patent/CN114972197A/en active Pending
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115311283A (en) * | 2022-10-12 | 2022-11-08 | 山东鲁玻玻璃科技有限公司 | Glass tube drawing defect detection method and system |
| CN116823808A (en) * | 2023-08-23 | 2023-09-29 | 青岛豪迈电缆集团有限公司 | Intelligent detection method for cable stranded wire based on machine vision |
| CN116823808B (en) * | 2023-08-23 | 2023-11-17 | 青岛豪迈电缆集团有限公司 | Intelligent detection method for cable stranded wire based on machine vision |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114972197A (en) | X-ray film imaging quality evaluation method and system | |
| US6011862A (en) | Computer-aided method for automated image feature analysis and diagnosis of digitized medical images | |
| DE69635711T2 (en) | NACHWEISSYSTEM AND METHOD FOR THE QUANTATIVE DETECTION OF TOOTH CARIES | |
| EP0757544B1 (en) | Computerized detection of masses and parenchymal distortions | |
| US8873826B2 (en) | Method for brightness level calculation of the digital x-ray image for medical applications | |
| CN111862044B (en) | Ultrasonic image processing method, ultrasonic image processing device, computer equipment and storage medium | |
| CN115661135A (en) | Focus region segmentation method for cardio-cerebral angiography | |
| CN116542966B (en) | Intelligent bone age analysis method for children endocrine abnormality detection | |
| CN110021019B (en) | AI-assisted hair thickness distribution analysis method for AGA clinical image | |
| CN116152505A (en) | Bone target identification and segmentation method based on X-ray data | |
| WO2020066257A1 (en) | Classification device, classification method, program, and information recording medium | |
| CN116246174B (en) | Sweet potato variety identification method based on image processing | |
| CN114693682B (en) | Spine feature identification method based on image processing | |
| Jeevakala et al. | A novel segmentation of cochlear nerve using region growing algorithm | |
| CN115689987A (en) | DR image-based dual-view vertebral fracture characteristic detection method | |
| CN116993764B (en) | Stomach CT intelligent segmentation extraction method | |
| CN115797246A (en) | Pathological image quality evaluation and adjustment method and system, electronic device and medium | |
| CN116485819B (en) | Ear-nose-throat examination image segmentation method and system | |
| CN114742753A (en) | Image evaluation method and device based on neural network | |
| Sankaran | An improved multipath residual CNN-based classification approach for periapical disease prediction and diagnosis in dental radiography | |
| Mashudi et al. | Future Iris Imaging with Advanced Fuzzified Histogram Equalization. | |
| CN117237342B (en) | Intelligent analysis method for respiratory rehabilitation CT image | |
| CN117557587B (en) | Endoscope cold light source brightness automatic regulating system | |
| Kardawi et al. | Image Enhancement for Breast Cancer Detection on Screening Mammography Using Deep Learning | |
| CN116403245B (en) | Poultry breeding data analysis method and system |
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 |