CN115526798A - Medical X-ray video sequence image denoising method and system and readable storage medium - Google Patents
Medical X-ray video sequence image denoising method and system and readable storage medium Download PDFInfo
- Publication number
- CN115526798A CN115526798A CN202211183118.3A CN202211183118A CN115526798A CN 115526798 A CN115526798 A CN 115526798A CN 202211183118 A CN202211183118 A CN 202211183118A CN 115526798 A CN115526798 A CN 115526798A
- Authority
- CN
- China
- Prior art keywords
- image
- video sequence
- block
- image block
- matching
- 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
- 238000000034 method Methods 0.000 title claims abstract description 47
- 239000013598 vector Substances 0.000 claims abstract description 69
- 230000009467 reduction Effects 0.000 claims abstract description 54
- 238000006073 displacement reaction Methods 0.000 claims abstract description 52
- 230000011218 segmentation Effects 0.000 claims abstract description 51
- 238000001914 filtration Methods 0.000 claims abstract description 28
- 238000004364 calculation method Methods 0.000 claims description 32
- 230000000737 periodic effect Effects 0.000 claims description 16
- 238000001514 detection method Methods 0.000 claims description 8
- 238000011156 evaluation Methods 0.000 claims description 5
- 230000006870 function Effects 0.000 claims description 5
- 210000000988 bone and bone Anatomy 0.000 description 34
- 210000004872 soft tissue Anatomy 0.000 description 15
- 230000001766 physiological effect Effects 0.000 description 7
- 230000000694 effects Effects 0.000 description 5
- 238000003384 imaging method Methods 0.000 description 5
- 238000012545 processing Methods 0.000 description 5
- 230000029058 respiratory gaseous exchange Effects 0.000 description 4
- 238000001356 surgical procedure Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000000399 orthopedic effect Effects 0.000 description 2
- 210000001519 tissue Anatomy 0.000 description 2
- 238000012935 Averaging Methods 0.000 description 1
- 210000003484 anatomy Anatomy 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 230000004069 differentiation Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000000968 intestinal effect Effects 0.000 description 1
- 210000004197 pelvis Anatomy 0.000 description 1
- 230000008855 peristalsis Effects 0.000 description 1
- 238000012805 post-processing Methods 0.000 description 1
- 238000013139 quantization Methods 0.000 description 1
- 230000000241 respiratory effect Effects 0.000 description 1
- 230000036391 respiratory frequency Effects 0.000 description 1
- 238000010845 search algorithm Methods 0.000 description 1
- 210000000689 upper leg Anatomy 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Images
Classifications
-
- G06T5/70—
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/20—Image enhancement or restoration by the use of local operators
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/13—Edge detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/136—Segmentation; Edge detection involving thresholding
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/20—Analysis of motion
- G06T7/223—Analysis of motion using block-matching
- G06T7/238—Analysis of motion using block-matching using non-full search, e.g. three-step search
-
- 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
- G06T2207/10016—Video; Image sequence
-
- 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
- G06T2207/20—Special algorithmic details
- G06T2207/20021—Dividing image into blocks, subimages or windows
-
- 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
- G06T2207/20—Special algorithmic details
- G06T2207/20172—Image enhancement details
- G06T2207/20192—Edge enhancement; Edge preservation
-
- 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
- G06T2207/30008—Bone
Abstract
The application provides a medical X-ray video sequence image denoising method and system and a readable storage medium, wherein the medical X-ray video sequence image denoising method comprises the following steps: acquiring a video sequence image; filtering the video sequence images through a band-pass filter; carrying out threshold segmentation on the video sequence image; dividing a video sequence image into a plurality of image blocks; matching the image block of the current frame with the image block of the previous frame in a preset search range by a preset matching criterion to obtain a most similar matching block; calculating a final displacement vector according to the image block of the current frame and the most similar matching block; and according to the final displacement vector, performing weighted noise reduction on the video sequence image through a recursive noise reduction algorithm to obtain an output image. By means of the technical scheme, all image frames can be averaged to the maximum extent on the premise that blurring does not occur, noise is reduced, blurring does not occur, and quality of a moving image is effectively improved.
Description
Technical Field
The application relates to the technical field of medical image processing, in particular to a medical X-ray video sequence image denoising method and system and a readable storage medium.
Background
The X-ray image noise reduction is an important link in the X-ray image post-processing, especially in the application of X-ray imaging in an operating room, the X-ray dosage is low, the quantum conversion efficiency of a detector is limited, and a great deal of noise is contained in the final image, so that serious interference is brought to the diagnosis of a doctor, especially the differentiation of an anatomical structure with small attenuation difference. From the source of noise, the noise is mainly classified into quantum noise, electronic noise, quantization noise, and the like, wherein the quantum noise is caused by randomness of poisson distribution of X photons and occupies a dominant position in most imaging.
Under given hardware conditions, noise reduction is usually performed by a certain algorithm, and the conventional processing includes time domain and space domain. The spatial domain is processed on a single-frame image, image edge and detail information are reserved through a certain filtering or iterative calculation mode, noise is reduced, but local characteristics of the image are usually changed, so that the image presents unnatural texture characteristics, time domain noise reduction mainly includes weighted averaging of multiple frames of images in time domain, when an imaged object does not move, a good noise reduction effect can be obtained, and when an object moves in a visual field, serious motion blur can be brought.
Disclosure of Invention
The present application aims to solve or improve the above technical problems.
Therefore, the first objective of the present application is to provide a medical X-ray video sequence image denoising method.
A second object of the present application is to provide a medical X-ray video sequence image noise reduction system.
A third object of the present application is to provide a medical X-ray video sequence image denoising system.
A fourth object of the present application is to provide a readable storage medium.
In order to achieve the first object of the present application, a technical solution of a first aspect of the present application provides a medical X-ray video sequence image denoising method, including: acquiring a video sequence image; filtering the video sequence images through a band-pass filter; carrying out threshold segmentation on the filtered video sequence image; dividing the video sequence image after threshold segmentation into a plurality of image blocks; matching the image block of the current frame with the image block of the previous frame in a preset search range by a preset matching criterion to obtain a most similar matching block; calculating a final displacement vector according to the image block of the current frame and the most similar matching block; and according to the final displacement vector, performing weighted noise reduction on the video sequence image through a recursive noise reduction algorithm to obtain an output image.
According to the medical X-ray video sequence image denoising method provided by the application, the video sequence image is firstly obtained. Designing a band-pass filter according to the X-ray skeleton image, and filtering the video sequence image through the band-pass filter. The specific band-pass filter designed for the X-ray skeleton image can filter high-frequency noise and low-frequency caps and retain image edge details. And then carrying out threshold segmentation on the filtered video sequence image, and segmenting the pixels of the most remarkable bone tissue boundary. Since the active joint motion of the patient does not exist in the operation, the motion is usually the motion of the bone dragged by the passive soft-group motion, and the motion is rigid, the motion of the whole bone can still be replaced by using the segmented pixels for subsequent motion estimation and matching, so that the calculation amount is greatly reduced. And dividing the video sequence image after threshold segmentation into a plurality of image blocks, wherein the displacements of all pixels in the image blocks are the same. And matching the image block of the current frame with the image block of the previous frame in a preset search range by a preset matching criterion to obtain the most similar matching block. The relative displacement between the image block of the current frame and the most similar matching block is the motion vector. After the image block of the previous frame and the most similar matching block are obtained, the displacement vectors of all the image blocks can be averaged to obtain the final displacement vector due to the rigidity of the bone motion. And finally, translating the previous frame of image according to the final displacement vector, and performing weighted noise reduction by using a recursive noise reduction algorithm to obtain an output image. The motion of the X-ray medical image is analyzed, the motion rule and the image characteristics of the X-ray medical image are identified, a band-pass filter is designed for the X-ray video sequence image, the edge characteristics of the X-ray video sequence image are extracted, and then a fast matching search method is utilized to match a moving object, so that all image frames are averaged to the maximum extent on the premise of not generating blur, that is, noise is reduced, the blur is not generated, the quality of a moving image is effectively improved, and the problem of blur distortion of the X-ray equipment in an operating room in a motion scene is solved.
In addition, the technical scheme provided by the application can also have the following additional technical characteristics:
in the above technical solution, the method for denoising medical X-ray video sequence images further comprises: identifying the periodicity of the periodic signal by an extreme point detection method to obtain an extreme point of the periodic signal; and taking the motion vector corresponding to the extreme point of the previous period as the initial value of the motion vector of the current frame.
In the technical scheme, the medical X-ray video sequence image denoising method further comprises the steps of identifying the periodicity of the periodic signal through an extreme point detection method, obtaining an extreme point of the periodic signal, and then taking a motion vector corresponding to the extreme point of the previous period as an initial value of the motion vector of the current frame. It can be understood that, since the physiological activities caused by the soft tissue are usually periodic and regular, such as breathing and heartbeat, the characteristics of the previous motion cycle can be used as the initial value of the next cycle, so as to speed up the calculation, and at the same time, provide additional physiological activity information. For periodic signals, the periodicity can be identified by using an extreme point detection method, the periodicity of amplitude signals of motion vectors is obvious, so that the local maximum of the signals is calculated, and in order to avoid noise interference, the distance between adjacent extreme values is in a reasonable range, and the extreme points with undersized distances are shielded. After the signal period is identified, the distance between the current frame and the extreme point of the previous period can be judged, and the motion vector at the corresponding position of the previous period is used as the initial value of the motion vector estimation of the current frame, so that the calculation speed is accelerated.
In the above technical solution, matching an image block of a current frame with an image block of a previous frame within a preset search range by using a preset matching criterion to obtain a most similar matching block specifically includes: inputting an image block of a current frame and an image block of a previous frame; respectively calculating the similarity of the image block of the current frame and the image block of the previous frame in a first preset search range by taking the estimation of the motion vector in the previous period as an initial value to obtain a first most similar image block; judging whether the similarity of the first most similar image block is smaller than an empirical threshold; if the similarity of the first most similar image block is smaller than the empirical threshold, taking the first most similar image block as a most similar matching block; if the similarity of the most similar image block is larger than or equal to the experience threshold, respectively calculating the similarity of the image block of the current frame and the image block of the previous frame in a second preset search range to obtain a second most similar image block; judging whether the similarity of the second most similar image block is smaller than an empirical threshold; and if the similarity of the second most similar image block is smaller than the empirical threshold, taking the second most similar image block as the most similar matching block.
In the technical scheme, matching is performed on an image block of a current frame and an image block of a previous frame within a preset search range through a preset matching criterion to obtain a most similar matching block, specifically, the image block of the current frame and the image block of the previous frame are firstly input. The image blocks are shifted by 2 pixels. Then, the similarity of the image block of the current frame and the image block of the previous frame in a first preset search range is respectively calculated, and the delta x and delta y of the minimum value of the similarity are searched to obtain a first most similar image block. And judging whether the similarity of the first most similar image block is smaller than an empirical threshold, and if the similarity of the first most similar image block is smaller than the empirical threshold, using the first most similar image block as a most similar matching block. If the similarity of the most similar image block is larger than or equal to the empirical threshold, respectively calculating the similarity of the image block of the current frame and the image block of the previous frame in a second preset search range, namely traversing by 1 pixel at intervals again in 8 neighborhoods of the position of deltax and deltay, and calculating the position of the minimum similarity to obtain a second most similar image block. And finally, judging whether the similarity of the second most similar image block is smaller than an empirical threshold. And if the similarity of the second most similar image block is smaller than the empirical threshold, taking the second most similar image block as the most similar matching block.
In the above technical solution, the evaluation function of the similarity includes an euclidean distance.
In the technical scheme, the matching similarity of the image block of the previous frame and the image block of the previous frame takes the Euclidean distance as an evaluation function, and the problem is changed into searching for delta x and delta y.
In the above technical solution, the threshold segmentation is performed on the filtered video sequence image, and the method specifically includes: respectively calculating the maximum value, the minimum value and the mean value of the video sequence image through a segmentation algorithm; and calculating a segmentation threshold value according to the maximum value, the minimum value and the mean value.
In the technical scheme, threshold segmentation is performed on the filtered video sequence image, specifically, the maximum value, the minimum value and the mean value of the video sequence image are respectively calculated through a segmentation algorithm. And then calculating a segmentation threshold according to the maximum value, the minimum value and the average value. It will be appreciated that since there is no active joint motion of the patient during surgery, typically motion of bone that is pulled by passive soft-group motion, which is rigid, the use of only these segmented pixels for subsequent motion estimation and matching can still replace the motion of all bones, thereby significantly reducing the computational effort.
In the above technical solution, the calculation formula of the segmentation threshold is:
Threshold＝α×Min+β×Mean+γ×Max；
α+β+γ＝1；
wherein Threshold is a segmentation Threshold, min is the minimum value of the image, mean is the Mean value of the image, max is the maximum value of the image, and alpha, beta and gamma are all weighting coefficients.
In this embodiment, the threshold value can be divided by a calculation formula for dividing the threshold value, and the calculation amount can be further reduced. Where α may be 0.4, β may be 0.6, and γ may be 0.
In the above technical solution, the calculation formula of the band pass filter is as follows:
wherein BPF (x, y) is a band-pass filter, D 0 Is the passband radius, BW is the passband width, and D (x, y) isDistance from the front point (x, y) to the center of the image, x is the horizontal axis coordinate, y is the vertical axis coordinate, and n is the filter order;
the calculation formula of the recursive noise reduction algorithm is as follows:
wherein i is a frame number, i =0,1 \8230, x (i) is an input image of a current frame, y (i) is an output image of the current frame, y (i-1) is an output image of a previous frame, and y (0) = x (0).
In the technical scheme, different passband radiuses and widths are selected through the band-pass filter to filter the image, so that different image characteristics can be obtained, and different levels of noise, bone boundaries and soft tissue backgrounds are presented. The noise is mainly concentrated in a high frequency band, although the high frequency band still has an obvious bone boundary, the noise is more and is difficult to distinguish from the bone boundary, the low frequency band also has a large amount of soft tissue information, and the movement of the soft tissue is non-rigid and is difficult to distinguish from the movement of the rigid bone, so the information of the medium frequency band can be extracted, the obvious bone boundary is reserved, and the interference of the noise and the movement of the non-rigid soft tissue is removed. After the final displacement vector is obtained, the previous frame image is translated according to the displacement vector, and the traditional recursive noise reduction algorithm is used for weighted noise reduction, so that the output of the previous frame algorithm can be obtained.
In order to achieve the second object of the present application, a technical solution of a second aspect of the present application provides a medical X-ray video sequence image denoising system, including: the acquisition module is used for acquiring video sequence images; the filtering module is used for filtering the video sequence images through the band-pass filter; the threshold segmentation module is used for carrying out threshold segmentation on the filtered video sequence image; the image block establishing module is used for dividing the video sequence image after threshold segmentation into a plurality of image blocks; the matching module is used for matching the image block of the current frame with the image block of the previous frame in the preset search range through a preset matching criterion to obtain a most similar matching block; the displacement vector calculation module is used for calculating a final displacement vector according to the image block of the current frame and the most similar matching block; and the weighted noise reduction module is used for carrying out weighted noise reduction on the video sequence image through a recursive noise reduction algorithm according to the final displacement vector to obtain an output image.
The medical X-ray video sequence image noise reduction system comprises an acquisition module, a filtering module, a threshold segmentation module, an image block establishing module, a matching module, a displacement vector calculation module and a weighting noise reduction module. The acquisition module is used for acquiring video sequence images. The filtering module is used for filtering the video sequence images through the band-pass filter. And the threshold segmentation module is used for performing threshold segmentation on the filtered video sequence images. The image block establishing module is used for dividing the video sequence image after threshold segmentation into a plurality of image blocks. The matching module is used for matching the image block of the current frame with the image block of the previous frame in the preset search range through a preset matching criterion to obtain the most similar matching block. And the displacement vector calculation module is used for calculating a final displacement vector according to the image block of the current frame and the most similar matching block. And the weighted noise reduction module is used for carrying out weighted noise reduction on the video sequence image through a recursive noise reduction algorithm according to the final displacement vector to obtain an output image. The motion of the X-ray medical image is analyzed, the motion rule and the image characteristics of the X-ray medical image are identified, a band-pass filter is designed for the X-ray video sequence image, the edge characteristics of the X-ray video sequence image are extracted, and then a fast matching search method is utilized to match a moving object, so that all image frames are averaged to the maximum extent on the premise of not generating blur, that is, noise is reduced, the blur is not generated, the quality of a moving image is effectively improved, and the problem of blur distortion of the X-ray equipment in an operating room in a motion scene is solved.
To achieve the third objective of the present application, a technical solution of a third aspect of the present application provides a medical X-ray video sequence image denoising system, including: the image denoising method comprises a memory and a processor, wherein a program or an instruction which can be run on the processor is stored in the memory, and the processor implements the image denoising method according to any one of the first aspect technical solutions when executing the program or the instruction, so that the method has the technical effects of any one of the first aspect technical solutions, and is not described herein again.
In order to achieve the fourth object of the present application, a readable storage medium is provided in a fourth aspect of the present application, where a program or an instruction is stored on the readable storage medium, and the program or the instruction is executed by a processor to implement the steps of the method for denoising medical X-ray video sequence images according to any one of the first aspect, so that the method has the technical effects of any one of the first aspect, and is not described herein again.
Additional aspects and advantages of the present application will be set forth in part in the description which follows, or may be learned by practice of the present application.
Drawings
The above and/or additional aspects and advantages of the present application will become apparent and readily appreciated from the following description of the embodiments, taken in conjunction with the accompanying drawings of which:
fig. 1 is a schematic flowchart illustrating steps of a medical X-ray video sequence image denoising method according to an embodiment of the present application;
FIG. 2 is a flowchart illustrating steps of a method for denoising medical X-ray video sequence images according to an embodiment of the present application;
FIG. 3 is a flowchart illustrating steps of a method for denoising medical X-ray video sequence images according to an embodiment of the present application;
FIG. 4 is a flowchart illustrating steps of a method for denoising medical X-ray video sequence images according to an embodiment of the present application;
FIG. 5 is a block diagram illustrating the structure of a medical X-ray video sequence image denoising system according to an embodiment of the present application;
FIG. 6 is a block diagram schematically illustrating a structure of a medical X-ray video sequence image denoising system according to another embodiment of the present application;
FIG. 7 is a flowchart illustrating the steps of a method for denoising medical X-ray video sequence images according to another embodiment of the present application;
FIG. 8 is a bandpass filter image of one embodiment of the present application;
FIG. 9 is a graph of images before and after processing by a band pass filter according to one embodiment of the present application;
FIG. 10 is an image before thresholding in accordance with an embodiment of the present application;
FIG. 11 is a thresholded segmented image according to one embodiment of the present application;
FIG. 12 is a flowchart illustrating the steps of a method for denoising medical X-ray video sequence images according to an embodiment of the present application;
FIG. 13 is a comparison of post-and pre-denoising illustrations according to an embodiment of the present application;
FIG. 14 is a schematic representation of a bone motion cycle according to an embodiment of the present application.
Wherein, the correspondence between the reference numbers and the component names in fig. 5 to 13 is:
10: a medical X-ray video sequence image noise reduction system; 110: an acquisition module; 120: a filtering module; 130: a threshold segmentation module; 140: an image block establishing module; 150: a matching module; 160; a displacement vector calculation module; 170: a weighted noise reduction module; 200: a band-pass filter image; 210: processing the pre-image by a band-pass filter; 220, 220; filtering the image at high frequency; 230: filtering the image at intermediate frequency; 240: filtering the image at a low frequency; 250: thresholding the pre-segmented image; 260: segmenting the image by using a threshold value; 270: denoising the image; 280: denoising the pre-image; 300: a memory; 400: a processor.
Detailed Description
In order that the above objects, features and advantages of the present application can be more clearly understood, the present application will be described in further detail with reference to the accompanying drawings and detailed description. It should be noted that the embodiments and features of the embodiments of the present application may be combined with each other without conflict.
In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present application, however, the present application may be practiced in other ways than those described herein, and therefore the scope of the present application is not limited by the specific embodiments disclosed below.
The method and system for reducing noise in a medical X-ray video sequence image and a readable storage medium according to some embodiments of the present application are described below with reference to fig. 1 to 14.
As shown in fig. 1, an embodiment of the first aspect of the present application provides a medical X-ray video sequence image denoising method, including the following steps:
step S102: acquiring a video sequence image;
step S104: filtering the video sequence images through a band-pass filter;
step S106: carrying out threshold segmentation on the filtered video sequence image;
step S108: dividing a video sequence image after threshold segmentation into a plurality of image blocks;
step S110: matching the image block of the current frame with the image block of the previous frame in a preset search range by a preset matching criterion to obtain a most similar matching block;
step S112: calculating a final displacement vector according to the image block of the current frame and the most similar matching block;
step S114: and according to the final displacement vector, performing weighted noise reduction on the video sequence image through a recursive noise reduction algorithm to obtain an output image.
According to the method for denoising medical X-ray video sequence images provided by the embodiment, video sequence images are firstly acquired. Designing a band-pass filter according to the X-ray skeleton image, and filtering the video sequence image through the band-pass filter. The specific band-pass filter designed for the X-ray skeleton image can filter high-frequency noise and low-frequency caps and retain image edge details. And then carrying out threshold segmentation on the filtered video sequence images, and segmenting the pixels of the most significant bone tissue boundary. Since the active joint motion of the patient does not exist in the operation, the motion is usually the motion of the bone dragged by the passive soft-group motion, and the motion is rigid, the motion of the whole bone can still be replaced by using the segmented pixels for subsequent motion estimation and matching, so that the calculation amount is greatly reduced. And dividing the video sequence image after threshold segmentation into a plurality of image blocks, wherein the displacements of all pixels in the image blocks are the same. And matching the image block of the current frame with the image block of the previous frame in a preset search range by a preset matching criterion to obtain the most similar matching block. The relative displacement between the image block of the current frame and the most similar matching block is the motion vector. After the image block of the previous frame and the most similar matching block are obtained, the displacement vectors of all the image blocks can be averaged to obtain a final displacement vector due to the rigidity of the bone motion. And finally, translating the previous frame of image according to the final displacement vector, and performing weighted noise reduction by using a recursive noise reduction algorithm to obtain an output image. The motion of the X-ray medical image is analyzed to identify the motion rule and the image characteristics of the X-ray medical image, a band-pass filter is designed aiming at the X-ray video sequence image, the edge characteristics of the X-ray video sequence image are extracted, and then a fast matching search method is utilized to match a moving object, so that all image frames are averaged to the maximum extent on the premise of not generating blur, that is, noise is reduced, the blur is not generated, the quality of a motion image is effectively improved, and the problem of blur distortion of X-ray equipment in an operating room in a motion scene is solved.
As shown in fig. 2, a method for denoising medical X-ray video sequence images according to another embodiment of the present application includes the following steps:
step S202: acquiring a video sequence image;
step S204: filtering the video sequence images through a band-pass filter;
step S206: respectively calculating the maximum value, the minimum value and the mean value of the video sequence image through a segmentation algorithm;
step S208: calculating a segmentation threshold value according to the maximum value, the minimum value and the mean value;
step S210: dividing the video sequence image after threshold segmentation into a plurality of image blocks;
step S212: matching the image block of the current frame with the image block of the previous frame in a preset search range by a preset matching criterion to obtain a most similar matching block;
step S214: calculating a final displacement vector according to the image block of the current frame and the most similar matching block;
step S216: and according to the final displacement vector, performing weighted noise reduction on the video sequence image through a recursive noise reduction algorithm to obtain an output image.
In this embodiment, threshold segmentation is performed on the filtered video sequence image, specifically, a maximum value, a minimum value, and a mean value of the video sequence image are respectively calculated by a segmentation algorithm. And then calculating a segmentation threshold value according to the maximum value, the minimum value and the mean value. It will be appreciated that since there is no active joint motion of the patient during surgery, typically motion of bone that is pulled by passive soft-group motion, which is rigid, the use of only these segmented pixels for subsequent motion estimation and matching can still replace the motion of all bones, thereby significantly reducing the computational effort.
Further, the calculation formula of the segmentation threshold is as follows:
Threshold＝α×Min+β×Mean+γ×Max；
α+β+γ＝1；
wherein Threshold is a segmentation Threshold, min is the minimum value of the image, mean is the average value of the image, max is the maximum value of the image, and alpha, beta and gamma are all weighting coefficients. By the calculation formula of the division threshold value, the threshold value division can be performed, and the calculation amount can be further reduced. Where α may be 0.4, β may be 0.6, and γ may be 0.
As shown in fig. 3, according to the method for denoising a medical X-ray video sequence image according to an embodiment of the present application, an image block of a current frame is matched with an image block of a previous frame in a preset search range by using a preset matching criterion, so as to obtain a most similar matching block, which specifically includes the following steps:
step S302: inputting an image block of a current frame and an image block of a previous frame;
step S304: respectively calculating the similarity of the image block of the current frame and the image block of the previous frame in a first preset search range by taking the estimation of the motion vector in the previous period as an initial value to obtain a first most similar image block;
step S306: judging whether the similarity of the first most similar image block is smaller than an empirical threshold;
step S308: if the similarity of the first most similar image block is smaller than the empirical threshold, taking the first most similar image block as a most similar matching block;
step S310: if the similarity of the most similar image block is larger than or equal to the experience threshold, respectively calculating the similarity of the image block of the current frame and the image block of the previous frame in a second preset search range to obtain a second most similar image block;
step S312: judging whether the similarity of the second most similar image block is smaller than an empirical threshold;
step S314: and if the similarity of the second most similar image block is smaller than the empirical threshold, taking the second most similar image block as the most similar matching block.
In this embodiment, matching the image block of the current frame with the image block of the previous frame within the preset search range by using a preset matching criterion to obtain a most similar matching block, specifically, firstly inputting the image block of the current frame and the image block of the previous frame. The image blocks are shifted by 2 pixels. Then, the similarity of the image block of the current frame and the image block of the previous frame in a first preset search range is respectively calculated, and the delta x and delta y of the minimum value of the similarity are searched to obtain a first most similar image block. And judging whether the similarity of the first most similar image block is smaller than an empirical threshold, and if the similarity of the first most similar image block is smaller than the empirical threshold, using the first most similar image block as a most similar matching block. If the similarity of the most similar image block is larger than or equal to the empirical threshold, respectively calculating the similarity of the image block of the current frame and the image block of the previous frame in a second preset search range, namely traversing by 1 pixel at an interval again in 8 neighborhoods of the positions of delta x and delta y, and calculating the position of the minimum similarity to obtain a second most similar image block. And finally, judging whether the similarity of the second most similar image block is smaller than an empirical threshold. And if the similarity of the second most similar image block is smaller than the empirical threshold, taking the second most similar image block as the most similar matching block.
Further, the matching similarity between the image block of the previous frame and the image block of the previous frame takes the euclidean distance as an evaluation function, and the problem becomes to find Δ x, Δ y. Wherein, the formula of the Euclidean distance is as follows:
the current is a current image of the image acquisition sequence (i, previous is a previous image of the image acquisition sequence, x is a horizontal axis coordinate, y is a vertical axis coordinate, Δ x is a horizontal axis relative displacement, and Δ y is a vertical axis relative displacement).
In the above embodiment, the calculation formula of the band pass filter is:
wherein D is 0 Is the passband radius, BW is the passband width, D is the distance from the current point (x, y) to the image center, and n is the filter order. Different passband radiuses and widths are selected through the bandpass filter to filter the image, so that different image characteristics can be obtained, and different levels of noise, bone boundaries and soft tissue backgrounds are presented. The noise is mainly concentrated in a high frequency band, although the high frequency band still has an obvious bone boundary, the noise is more and is difficult to distinguish from the bone boundary, the low frequency band also has a large amount of soft tissue information, and the movement of the soft tissue is non-rigid and is difficult to distinguish from the movement of the rigid bone, so the information of the medium frequency band can be extracted, the obvious bone boundary is reserved, and the interference of the noise and the movement of the non-rigid soft tissue is removed.
Further, the calculation formula of the recursive noise reduction algorithm is:
where i is a frame number, i =0,1 \8230, x (i) is an input image of a current frame, y (i) is an output image of the current frame, and y (0) = x (0). After the final displacement vector is obtained, the previous frame image is translated according to the displacement vector, and the traditional recursive noise reduction algorithm is used for weighted noise reduction, so that the output of the previous frame algorithm can be obtained.
As shown in fig. 4, the method for denoising medical X-ray video sequence images according to an embodiment of the present application further includes the following steps:
step S402: identifying the periodicity of the periodic signal by an extreme point detection method to obtain an extreme point of the periodic signal;
step S404: and taking the motion vector corresponding to the extreme point of the previous period as the initial value of the motion vector of the current frame.
In this embodiment, the method for denoising a medical X-ray video sequence image further includes identifying periodicity of the periodic signal by an extreme point detection method to obtain an extreme point of the periodic signal, and then using a motion vector corresponding to the extreme point of the previous period as an initial value of the motion vector of the current frame. It can be understood that, since the physiological activities caused by the soft tissue are usually periodic and regular, such as breathing and heartbeat, the characteristic of the previous motion cycle can be used as the initial value of the next cycle, so as to speed up the calculation speed, and at the same time, provide additional physiological activity information. For periodic signals, the periodicity can be identified by using an extreme point detection method, the local maximum of the signals is calculated because the periodicity of amplitude signals of motion vectors is obvious, and in order to avoid noise interference, the distance between adjacent extreme values is in a reasonable range, and the extreme points with too small distance are shielded. After the signal period is identified, the distance between the current frame and the extreme point of the previous period can be judged, and the motion vector at the corresponding position of the previous period is used as the initial value of the motion vector estimation of the current frame, so that the calculation speed is accelerated.
As shown in fig. 5, an embodiment of the second aspect of the present application provides a medical X-ray video sequence image denoising system 10, including: an obtaining module 110, configured to obtain a video sequence image; a filtering module 120, configured to filter the video sequence images through a band-pass filter; a threshold segmentation module 130, configured to perform threshold segmentation on the filtered video sequence images; an image block establishing module 140, configured to divide the video sequence image after threshold segmentation into a plurality of image blocks; the matching module 150 is configured to match an image block of a current frame with an image block of a previous frame within a preset search range according to a preset matching criterion to obtain a most similar matching block; a displacement vector calculation module 160, configured to calculate a final displacement vector according to the image block of the current frame and the most similar matching block; and the weighted noise reduction module 170 is configured to perform weighted noise reduction on the video sequence image through a recursive noise reduction algorithm according to the final displacement vector to obtain an output image.
The medical X-ray video sequence image noise reduction system 10 according to the present embodiment includes an obtaining module 110, a filtering module 120, a threshold segmentation module 130, an image block establishing module 140, a matching module 150, a displacement vector calculating module 160, and a weighted noise reduction module 170. The obtaining module 110 is configured to obtain images of a video sequence. The filtering module 120 is configured to filter the images of the video sequence through a band-pass filter. The threshold segmentation module 130 is configured to threshold the filtered video sequence images. The image block establishing module 140 is configured to divide the threshold-divided video sequence image into a plurality of image blocks. The matching module 150 is configured to match an image block of a current frame with an image block of a previous frame within a preset search range according to a preset matching criterion, so as to obtain a most similar matching block. The displacement vector calculation module 160 is configured to calculate a final displacement vector according to the image block of the current frame and the most similar matching block. And the weighted noise reduction module 170 is configured to perform weighted noise reduction on the video sequence image through a recursive noise reduction algorithm according to the final displacement vector to obtain an output image. The motion of the X-ray medical image is analyzed to identify the motion rule and the image characteristics of the X-ray medical image, a band-pass filter is designed aiming at the X-ray video sequence image, the edge characteristics of the X-ray video sequence image are extracted, and then a fast matching search method is utilized to match a moving object, so that all image frames are averaged to the maximum extent on the premise of not generating blur, that is, noise is reduced, the blur is not generated, the quality of a motion image is effectively improved, and the problem of blur distortion of X-ray equipment in an operating room in a motion scene is solved.
As shown in fig. 6, an embodiment of the third aspect of the present application provides a medical X-ray video sequence image denoising system 10, including: a memory 300 and a processor 400, where the memory 300 stores a program or an instruction that can be executed on the processor 400, and the processor 400 implements the steps of the method for reducing noise in a medical X-ray video sequence image according to any one of the embodiments of the first aspect when executing the program or the instruction, so that the method has the technical effects of any one of the embodiments of the first aspect, which are not described herein again.
An embodiment of the fourth aspect of the present application provides a readable storage medium, on which a program or an instruction is stored, where the program or the instruction is executed by a processor to implement the steps of the method for denoising medical X-ray video sequence images according to any one of the embodiments of the first aspect, so that the method has the technical effects of any one of the embodiments of the first aspect, and is not described herein again.
As shown in fig. 7 to 14, a method for denoising medical X-ray video sequence images according to an embodiment provided by the present application includes filtering, threshold segmentation, motion estimation and matching, recursive denoising, and motion vector estimation.
As shown in fig. 7, the whole algorithm is still based on a recursive noise reduction framework, and the flow chart is as follows:
step S502: the i-1 frame image;
step S504: filtering in a frequency domain;
step S506: filtering in a frequency domain;
step S508: ith frame image
Step S510: filtering in a frequency domain;
step S512: filtering in a frequency domain;
step S514: estimating the motion;
step S516: a sequence of motion vectors;
step S518: recursive noise reduction;
step S520: and outputting the ith frame.
Filtering:
in the orthopedic surgery application, the main scene causing the image noise to be larger is spine or pelvis/femur related operation, such as pain treatment, vertebroplasty, femoral head replacement, etc., the main motion of the patient during the imaging in the operation is caused by breathing or intestinal peristalsis, the motion amplitude is smaller, while the doctor in the operation is concerned with the image of the bone part, and a band-pass filter is designed to filter the image:
wherein D is 0 Is the passband radius, BW is the passband width, D is the distance from the current point (x, y) to the image center, and n is the filter order. As shown in fig. 8, a bandpass filter image 200.
As shown in fig. 9, the image 210 before the band-pass filter processing is filtered by different pass-band radii and widths, so as to obtain a high-frequency filtered image 220, a medium-frequency filtered image 230, and a low-frequency filtered image 240. The image is filtered by selecting different pass band radiuses and widths, so that different image characteristics can be obtained, and different levels of noise, bone boundaries and soft tissue backgrounds are presented. The noise is mainly concentrated in a high frequency band, and although the high frequency band still has an obvious bone boundary, the noise is more and is difficult to distinguish from the bone boundary; and the low-frequency band has a large amount of soft tissue information, and because the movement of the soft tissue is non-rigid and is difficult to distinguish from the movement of rigid bones, the information of the medium-frequency band can be extracted, so that the obvious bone boundary is reserved, and the interference of noise and the movement of the non-rigid soft tissue is removed.
Threshold segmentation:
in order to further reduce the calculation amount in the engineering realization, the image can be simply subjected to threshold segmentation, the pixels of the most remarkable bone tissue boundary are segmented, and because the active joint motion of the patient does not exist in the operation, the motion of the bone is generally drawn by passive soft group motion, and the motion is rigid, the subsequent motion estimation and matching can be carried out by only using the segmented pixels, and the motion of the whole bone can still be replaced, thereby greatly reducing the calculation amount.
As shown in fig. 10 and 11, the segmentation algorithm calculates the maximum value, the minimum value and the mean value of the image, respectively, and the segmentation threshold is: threshold = α × Min + β × Mean + γ × Max; α + β + γ =1; wherein Threshold is a segmentation Threshold, min is the minimum value of the image, mean is the Mean value of the image, max is the maximum value of the image, and alpha, beta and gamma are all weighting coefficients. Alpha is 0.4, beta is 0.6 and gamma is 0. The pre-thresholded image 250 is thresholded to obtain a post-thresholded image 260.
Motion estimation and matching: because the motion amplitude of a patient is small during X-ray imaging in operation and the patient moves in a rigid body, the image is divided into 8-8 non-overlapping blocks according to experience by using motion estimation based on image blocks, the displacements of all pixels in the blocks are considered to be the same, then each block of a current frame (i) and a block in a given search range of a reference frame (i-1) are searched according to a certain matching criterion, a matching block which is most similar to the current block is found, and the relative displacement of the two blocks is a motion vector. The matching similarity of two blocks takes the Euclidean distance as an evaluation function, and the problem is changed into the following steps of searching for delta x, delta y:
wherein current and previous are the current image (i) and the previous image (i-1) of the image acquisition sequence, respectively.
Since the motion amplitude of the patient is small during the operation, the search can be limited to a small range, assuming that the respiratory frequency of the adult is f =18 times/min, the maximum offset of the position of the respiratory skeleton is s =10mm, the pixel size of the x-ray detector is ds =0.2mm, the magnification ratio is m =1.5, and the frame rate of the perspective imaging is fr =30fps, the average displacement of the two adjacent frames of images is: Δ = s/(f/60/2 fr) m/ds ≈ 18 pixels, so for this typical application scenario, while taking into account design margins, the search range may be limited to a ± 35pixel range.
As shown in fig. 12, the block matching search algorithm flow is as follows:
step S602: inputting a current frame and a previous frame of image;
step S604: moving image blocks at intervals of 2 pixels;
step S606: calculating the similarity;
step S608: searching for the minimum value of the similarity, namely delta x and delta y;
step S610: stopping searching if the similarity is smaller than an empirical threshold;
step S612: traversing with 1 pixel at the interval again in 8 neighborhoods of the position of the delta x and the position of the delta y, and calculating the position of the minimum similarity;
step S614: obtaining the final matching result of delta x and delta y;
step S616: and stopping the search if the similarity is less than the empirical threshold.
Recursive noise reduction:
after the front and back two registered images are obtained, the displacement vectors of all the blocks can be averaged to obtain the final displacement vector due to the rigidity of the bone motion. And then, translating the previous frame image according to the displacement vector, and performing weighted noise reduction by using a traditional recursive noise reduction algorithm. The calculation formula of the recursive noise reduction algorithm is as follows:
where i is the frame number, i =0,1 \8230, x (i) is the input image of the current frame, y (i) is the output image of the current frame, and y (0) = x (0).
As shown in fig. 13, a noise-reduced image 270 and a noise-reduced image 280.
And (3) motion vector estimation:
since the physiological activities caused by the soft tissue are usually periodic and regular, such as breathing and heartbeat, the characteristics of the previous motion cycle can be used as the initial value of the next cycle, so that the calculation speed is increased, and meanwhile, additional physiological activity information can be provided. As shown in fig. 14, the horizontal axis represents time, and the vertical axis represents the displacement amplitude and displacement direction of the bone motion caused by the physiological activity.
For periodic signals, the periodicity can be identified by using an extreme point detection method, the local maximum of the signals is calculated because the periodicity of amplitude signals of motion vectors is obvious, and in order to avoid noise interference, the distance between adjacent extreme values is in a reasonable range, and the extreme points with too small distance are shielded. After the signal period is identified, the distance between the current frame and the extreme point of the previous period can be judged, and the motion vector at the corresponding position of the previous period is used as the initial value of the motion vector estimation of the current frame, so that the calculation speed is accelerated.
To sum up, the beneficial effect of this application embodiment is:
1. the problem of fuzzy distortion of the X-ray equipment in the operating room in a motion scene is solved, the image quality is greatly improved, and doctors are effectively helped to identify different human tissues.
2. Can be applied to any X-ray projection equipment and is suitable for solving the problem of image blurring caused by physiological motion of human tissues in orthopedic operations.
In this application, the terms "first," "second," and "third" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance; the term "plurality" means two or more unless expressly limited otherwise. The terms "mounted," "connected," "fixed," and the like are used broadly and should be construed to include, for example, "connected" may be a fixed connection, a detachable connection, or an integral connection; "coupled" may be direct or indirect through an intermediary. The specific meaning of the above terms in the present application can be understood by those of ordinary skill in the art as appropriate.
In the description of the present application, it is to be understood that the terms "upper", "lower", "front", "rear", and the like indicate orientations or positional relationships based on those shown in the drawings, and are only for convenience of description and simplicity of description, but do not indicate or imply that the referred device or module must have a specific direction, be constructed in a specific orientation, and be operated, and thus, should not be construed as limiting the present application.
In the description herein, the description of the terms "one embodiment," "some embodiments," "specific embodiments," etc., means that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the application. In this specification, the schematic representations of the terms used above do not necessarily refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.
The above description is only a preferred embodiment of the present application and is not intended to limit the present application, and various modifications and changes may be made to the present application by those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present application shall be included in the protection scope of the present application.
Claims (10)
1. A medical X-ray video sequence image noise reduction method is characterized by comprising the following steps:
acquiring a video sequence image;
filtering the video sequence images through a band-pass filter;
performing threshold segmentation on the filtered video sequence image;
dividing the video sequence image after threshold segmentation into a plurality of image blocks;
matching the image block of the current frame with the image block of the previous frame in a preset search range by a preset matching criterion to obtain a most similar matching block;
calculating a final displacement vector according to the image block of the current frame and the most similar matching block;
and according to the final displacement vector, performing weighted noise reduction on the video sequence image through a recursive noise reduction algorithm to obtain an output image.
2. The method for denoising medical X-ray video sequence images according to claim 1, further comprising:
identifying the periodicity of the periodic signal by an extreme point detection method to obtain an extreme point of the periodic signal;
and taking the motion vector corresponding to the extreme point of the previous period as the initial value of the motion vector of the current frame.
3. The method for denoising a medical X-ray video sequence image according to claim 1, wherein the matching of an image block of a current frame with an image block of a previous frame within a preset search range by a preset matching criterion to obtain a most similar matching block specifically comprises:
inputting an image block of a current frame and an image block of a previous frame;
respectively calculating the similarity of the image block of the current frame and the image block of the previous frame in a first preset search range by taking the estimation of the motion vector in the previous period as an initial value to obtain a first most similar image block;
judging whether the similarity of the first most similar image block is smaller than an empirical threshold;
if the similarity of the first most similar image block is smaller than an empirical threshold, taking the first most similar image block as a most similar matching block;
if the similarity of the most similar image block is larger than or equal to the empirical threshold, respectively calculating the similarity of the image block of the current frame and the image block of the previous frame in a second preset search range to obtain a second most similar image block;
judging whether the similarity of the second most similar image block is smaller than the empirical threshold;
and if the similarity of the second most similar image block is smaller than the empirical threshold, taking the second most similar image block as a most similar matching block.
4. The medical X-ray video sequence image denoising method according to claim 3,
the evaluation function of the similarity includes the euclidean distance.
5. The method for denoising medical X-ray video sequence images according to any of claims 1 to 4, wherein the performing threshold segmentation on the filtered video sequence images specifically comprises:
respectively calculating the maximum value, the minimum value and the mean value of the video sequence images through a segmentation algorithm;
and calculating a segmentation threshold value according to the maximum value, the minimum value and the mean value.
6. The medical X-ray video sequence image denoising method of claim 5,
the calculation formula of the segmentation threshold is as follows:
Threshold＝α×Min+β×Mean+γ×Max；
α+β+γ＝1；
wherein Threshold is a segmentation Threshold, min is the minimum value of the image, mean is the Mean value of the image, max is the maximum value of the image, and alpha, beta and gamma are all weighting coefficients.
7. The method for denoising medical X-ray video sequence images according to any of claims 1 to 4, wherein the band-pass filter has a calculation formula of:
wherein BPF (x, y) is a band-pass filter, D 0 Is the radius of the passband, BW is the width of the passband, D (x, y) is the distance from the current point (x, y) to the center of the image, x is the coordinate of the horizontal axis, y is the coordinate of the vertical axis, and n is the order of the filter;
the calculation formula of the recursive noise reduction algorithm is as follows:
wherein i is a frame number, i =0,1 \ 8230, x (i) is an input image of a current frame, y (i) is an output image of the current frame, y (i-1) is an output image of a previous frame, and y (0) = x (0).
8. A medical X-ray video sequence image denoising system, comprising:
an acquisition module (110) for acquiring images of a video sequence;
a filtering module (120) for filtering the video sequence images by means of a band-pass filter;
a thresholding module (130) for thresholding the filtered images of the video sequence;
an image block establishing module (140) for dividing the threshold-divided video sequence image into a plurality of image blocks;
the matching module (150) is used for matching the image block of the current frame with the image block of the previous frame in the preset search range through a preset matching criterion to obtain a most similar matching block;
a displacement vector calculation module (160) for calculating a final displacement vector according to the image block of the current frame and the most similar matching block;
and the weighted noise reduction module (170) is used for carrying out weighted noise reduction on the video sequence image through a recursive noise reduction algorithm according to the final displacement vector to obtain an output image.
9. A medical X-ray video sequence image denoising system, comprising:
a memory (300) and a processor (400), wherein the memory (300) has stored thereon a program or instructions executable on the processor (400), the processor (400) when executing the program or instructions implementing the steps of the method for denoising medical X-ray video sequence images according to any of claims 1 to 7.
10. A readable storage medium, on which a program or instructions are stored, which, when being executed by a processor, carry out the steps of the method for denoising medical X-ray video sequence images according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211183118.3A CN115526798A (en) | 2022-09-27 | 2022-09-27 | Medical X-ray video sequence image denoising method and system and readable storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211183118.3A CN115526798A (en) | 2022-09-27 | 2022-09-27 | Medical X-ray video sequence image denoising method and system and readable storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115526798A true CN115526798A (en) | 2022-12-27 |
Family
ID=84699984
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211183118.3A Pending CN115526798A (en) | 2022-09-27 | 2022-09-27 | Medical X-ray video sequence image denoising method and system and readable storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115526798A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116016807A (en) * | 2022-12-30 | 2023-04-25 | 广东中星电子有限公司 | Video processing method, system, storable medium and electronic equipment |
-
2022
- 2022-09-27 CN CN202211183118.3A patent/CN115526798A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116016807A (en) * | 2022-12-30 | 2023-04-25 | 广东中星电子有限公司 | Video processing method, system, storable medium and electronic equipment |
| CN116016807B (en) * | 2022-12-30 | 2024-04-19 | 广东中星电子有限公司 | Video processing method, system, storable medium and electronic equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP5331797B2 (en) | Medical diagnostic device and method for improving image quality of medical diagnostic device | |
| US20100266188A1 (en) | Chest x-ray registration, subtraction and display | |
| JP4104054B2 (en) | Image alignment apparatus and image processing apparatus | |
| WO2008024992A2 (en) | Medical image enhancement system | |
| Patil et al. | Preprocessing to be considered for MR and CT images containing tumors | |
| JP2002032735A (en) | Method and device for aligning image | |
| JP2002324238A (en) | Method and device for positioning image | |
| Baltruschat et al. | When does bone suppression and lung field segmentation improve chest x-ray disease classification? | |
| CN115526798A (en) | Medical X-ray video sequence image denoising method and system and readable storage medium | |
| WO2013191061A1 (en) | Image processing device | |
| Buzug et al. | Using an entropy similarity measure to enhance the quality of DSA images with an algorithm based on template matching | |
| CN106910193B (en) | Scanning image processing method | |
| US20070140582A1 (en) | Systems and Methods For Reducing Noise In Image Sequences | |
| CN109544673B (en) | Three-dimensional imaging method and system for cartilage segmentation of medical image | |
| WO2014136415A1 (en) | Body-movement detection device and method | |
| US20050169509A1 (en) | Method for improving the image quality | |
| JP2002125153A (en) | Noise suppression processing unit and recording medium | |
| JP2001283215A (en) | Image processor | |
| Na'am et al. | Filter technique of medical image on multiple morphological gradient (MMG) method | |
| JP4095140B2 (en) | Image processing apparatus and method | |
| Abbas et al. | Perceptual method for MRI medical images improvement in presence of impulse noise | |
| JP5305687B2 (en) | X-ray video imaging system | |
| Taleb et al. | A 3D space–time motion evaluation for image registration in digital subtraction angiography | |
| JP2009054013A (en) | Image processor | |
| KR102492949B1 (en) | Processing apparatus and method for medical image |
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 |