CN111861929B - Ultrasonic image optimization processing method, system and device - Google Patents
Ultrasonic image optimization processing method, system and device Download PDFInfo
- Publication number
- CN111861929B CN111861929B CN202010724086.8A CN202010724086A CN111861929B CN 111861929 B CN111861929 B CN 111861929B CN 202010724086 A CN202010724086 A CN 202010724086A CN 111861929 B CN111861929 B CN 111861929B
- Authority
- CN
- China
- Prior art keywords
- image
- ultrasonic
- layer
- sampling
- edge
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000005457 optimization Methods 0.000 title claims abstract description 34
- 238000003672 processing method Methods 0.000 title claims abstract description 17
- 238000005070 sampling Methods 0.000 claims abstract description 192
- 230000004927 fusion Effects 0.000 claims abstract description 115
- 238000000034 method Methods 0.000 claims abstract description 69
- 238000002604 ultrasonography Methods 0.000 claims description 79
- 230000004044 response Effects 0.000 claims description 62
- 230000008569 process Effects 0.000 claims description 36
- 238000009792 diffusion process Methods 0.000 claims description 23
- 238000001914 filtration Methods 0.000 claims description 11
- 238000009499 grossing Methods 0.000 claims description 7
- 238000004590 computer program Methods 0.000 claims description 6
- 230000002708 enhancing effect Effects 0.000 claims description 5
- 238000012285 ultrasound imaging Methods 0.000 claims description 5
- 238000003384 imaging method Methods 0.000 claims description 2
- 230000001276 controlling effect Effects 0.000 description 8
- 230000008859 change Effects 0.000 description 7
- 230000000694 effects Effects 0.000 description 7
- 238000010586 diagram Methods 0.000 description 4
- 238000003708 edge detection Methods 0.000 description 4
- 230000009471 action Effects 0.000 description 3
- 230000002411 adverse Effects 0.000 description 2
- 238000004422 calculation algorithm Methods 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000002596 correlated effect Effects 0.000 description 1
- 230000000875 corresponding effect Effects 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 238000007499 fusion processing Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 238000011084 recovery Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
-
- 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
- G06T2207/10132—Ultrasound 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
- G06T2207/20172—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
- G06T2207/20—Special algorithmic details
- G06T2207/20212—Image combination
- G06T2207/20221—Image fusion; Image merging
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Ultra Sonic Daignosis Equipment (AREA)
Abstract
The application discloses an ultrasonic image optimization processing method, an ultrasonic image optimization processing system and an ultrasonic image optimization processing device, which are used for downsampling an original ultrasonic image to obtain ultrasonic sampling images with different resolutions, determining the original ultrasonic image and a pre-edge-enhanced ultrasonic image, carrying out edge enhancement processing on the pre-edge-enhanced ultrasonic image according to a preset image edge enhancement strategy to obtain an ultrasonic enhancement image, fusing the pre-edge-enhanced ultrasonic image and the ultrasonic enhancement image to obtain an ultrasonic fusion image, upsampling the ultrasonic fusion image to obtain an upsampled image, and fusing and reconstructing the upsampled image and the original ultrasonic image to obtain a final ultrasonic enhancement image. Therefore, the application not only can strengthen the image edge, but also can make the processed image more natural, and can meet the real-time requirement of clinic on the method.
Description
Technical Field
The present invention relates to the field of ultrasound image processing, and in particular, to an ultrasound image optimization processing method, system and device.
Background
At present, in order to improve the quality of an ultrasonic image, the technical means adopted generally are that the ultrasonic image is divided into a structural area and a non-structural area according to a tissue structure contained in the ultrasonic image, and the structural area is subjected to anisotropic treatment and the non-structural area is subjected to isotropic treatment so as to achieve the purpose of enhancing the edge of the ultrasonic image, thereby improving the quality of the ultrasonic image. However, different areas obtained by dividing the ultrasonic image adopt different image processing modes, so that the difference of the different areas is obvious, and the ultrasonic image with improved quality is not natural to a certain extent.
Therefore, how to provide a solution for quality optimization processing of ultrasound images is a problem that needs to be solved by those skilled in the art.
Disclosure of Invention
The invention aims to provide an ultrasonic image optimization processing method, an ultrasonic image optimization processing system and an ultrasonic image optimization processing device, which not only can strengthen the edge of an image, but also can make the processed image more natural, and can meet the real-time requirement of clinic on the method.
In order to solve the technical problems, the invention provides an ultrasonic image optimization processing method, which comprises the following steps:
Downsampling an original ultrasonic image to obtain ultrasonic sampling images with different resolutions, and determining the original ultrasonic image and the ultrasonic image with the pre-edge enhancement;
Performing edge enhancement processing on the ultrasonic image with the pre-edge enhancement according to a preset image edge enhancement strategy to obtain an ultrasonic enhancement image;
Fusing the pre-edge enhanced ultrasonic image and the ultrasonic enhanced image to obtain an ultrasonic fused image;
And carrying out up-sampling on the ultrasonic fusion image to obtain an up-sampling image, and carrying out fusion reconstruction on the up-sampling image and the original ultrasonic image to obtain a final ultrasonic enhanced image.
Preferably, the process of downsampling the original ultrasound image to obtain ultrasound sampling images with different resolutions, and determining the original ultrasound image and the pre-edge enhanced ultrasound image includes:
three layers of downsampling are carried out on an original ultrasonic image to obtain three layers of ultrasonic sampling images with different resolutions, wherein the resolution of a first layer of ultrasonic sampling image is greater than the resolution of a second layer of ultrasonic sampling image and greater than the resolution of a third layer of ultrasonic sampling image;
taking the first layer ultrasonic sampling image as an original ultrasonic image, and taking the second layer ultrasonic sampling image and the third layer ultrasonic sampling image as pre-edge enhanced ultrasonic images;
Correspondingly, the process of performing edge enhancement processing on the pre-edge enhanced ultrasonic image according to a preset image edge enhancement strategy to obtain an ultrasonic enhanced image comprises the following steps:
Performing edge enhancement processing on the second-layer ultrasonic sampling image according to a preset image edge enhancement strategy to obtain a second-layer ultrasonic enhancement image;
performing edge enhancement processing on the third-layer ultrasonic sampling image according to a preset image edge enhancement strategy to obtain a third-layer ultrasonic enhancement image;
The process of carrying out up-sampling on the ultrasonic fusion image to obtain an up-sampling image, and carrying out fusion reconstruction on the up-sampling image and the original ultrasonic image to obtain a final ultrasonic enhancement image comprises the following steps:
Fusing the third-layer ultrasonic sampling image with the third-layer ultrasonic enhancement image, and upsampling the fused ultrasonic image to obtain a third-layer ultrasonic upsampling image;
Performing fusion reconstruction on the second-layer ultrasonic sampling image, the second-layer ultrasonic enhancement image and the third-layer ultrasonic upsampling image, and upsampling the fused reconstructed ultrasonic image to obtain a second-layer ultrasonic upsampling image;
and carrying out fusion reconstruction on the first layer ultrasonic sampling image and the second layer ultrasonic up-sampling image to obtain a final ultrasonic enhancement image.
Preferably, the process of performing edge enhancement processing on the third layer of ultrasonic sampling image according to a preset image edge enhancement policy to obtain a third layer of ultrasonic enhancement image includes:
and carrying out multi-scale edge enhancement processing on the third layer ultrasonic sampling image to obtain a third layer ultrasonic enhancement image.
Preferably, the process of performing multi-scale edge enhancement processing on the third layer of ultrasonic sampling image to obtain a third layer of ultrasonic enhancement image includes:
Carrying out isotropic filtering denoising treatment on the ultrasonic image subjected to multi-scale edge enhancement treatment to obtain a denoised ultrasonic image The method comprises the steps of carrying out multi-scale edge enhancement in an iterative mode, wherein n is the number of iterations, and the maximum value of n is equal to a preset iteration threshold value;
Ultrasound imaging Respectively inputting the filter to a pre-constructed odd filter and an even filter to obtain odd filter response and even filter response on multiple scales, and calculating the difference between the odd filter response and the even filter response to obtain edge response on multiple scales;
Fusing the edge responses of the multiple scales to obtain an edge image, and performing image pair ultrasonic imaging based on the edge image Performing multi-scale edge enhancement processing and image denoising processing to obtain an ultrasonic image
When the iteration number of the multi-scale edge enhancement is smaller than a preset iteration threshold, the ultrasonic image is obtainedAnd when the iteration number of the multi-scale edge enhancement is equal to a preset iteration threshold value, stopping the iteration.
Preferably, the difference between the odd and even filter responses is calculated to obtain multi-scale edge responses, the multi-scale edge responses are fused to obtain an edge image, and the ultrasonic image is paired based on the edge imagePerforming multi-scale edge enhancement processing and image denoising processing to obtain an ultrasonic imageComprises the following steps:
According to the edge response relation Calculating the difference between the odd and even filter responses to obtain an edge response on the scale sWherein, For the even filter response on scale s; T s is a noise threshold value related to the scale s, x is a pixel point of the image;
based on the response fusion relationship The edge responses of a plurality of scales are fused to obtain an edge image Eg 3, wherein,The value range of (1) is 0,1, m is the number of filter scales;
Enhancing relationships from multiple scales of images For ultrasound imagesPerforming multi-scale edge enhancement processing and image denoising processing to obtain an ultrasonic imageWhere Δt is the time step and δ 3 is the scale factor.
Preferably, the process of performing edge enhancement processing on the second-layer ultrasound sampling image according to a preset image edge enhancement policy to obtain a second-layer ultrasound enhancement image includes:
And performing multidirectional edge enhancement processing on the second-layer ultrasonic sampling image to obtain a second-layer ultrasonic enhancement image.
Preferably, the process of performing multidirectional edge enhancement processing on the second-layer ultrasonic sampling image to obtain a second-layer ultrasonic enhancement image includes:
multidirectional enhancement of relationships according to preset images Performing multidirectional edge enhancement processing and image denoising processing on the second-layer ultrasonic sampling image within a preset edge enhancement time T1 to obtain a second-layer ultrasonic enhancement image;
Wherein, The method comprises the steps of obtaining a gradient operator, wherein I is a gradient value, t is a time operator, k 1 is a parameter for controlling smoothing, u 0 is an initial iteration image, u is an ultrasonic image obtained after diffusion treatment for the last time, beta is a super parameter, beta is epsilon [0,1], eg is an edge operator based on phase, the value range is [0,1], x is a pixel point of the image, g (x) divides the ultrasonic image into two areas, omega 1 = { x epsilon omega, g (x)/(1 } and omega 2＝{x∈Ω,0≤g(x)<1},Ω＝Ω1+Ω2, epsilon represents a normal constant, s is a certain scale of a filter, m is the number of the filter scale, and F s is a phase-based filter characteristic on an s scale; for the even filter response on scale s; Is the odd filter response at scale s, and T s is a noise threshold associated with scale s.
Preferably, the process of fusing the third layer ultrasonic sampling image with the third layer ultrasonic enhancement image and upsampling the fused ultrasonic image to obtain a third layer ultrasonic upsampling image includes:
According to the preset image coarse fusion relation Performing coarse fusion on the third-layer ultrasonic sampling image I 3 and the third-layer ultrasonic enhancement image Eh 3 to obtain a third-layer ultrasonic coarse fusion image Fu 3, wherein Th 3 is a given threshold value, mu is a parameter for controlling the image edge enhancement degree, alpha is a parameter for controlling the image enhancement degree, eg 3 is an edge image of the third-layer ultrasonic sampling image, and the value range is [0,1];
up-sampling the third-layer ultrasonic rough fusion image Fu 3 to obtain a third-layer ultrasonic up-sampling image U 3;
according to the preset image fine fusion relation Performing diffusion treatment on the third-layer ultrasonic up-sampling image U 3 within a preset fine fusion time T2 to obtain a third-layer ultrasonic fusion reconstruction image R 3, wherein U is an ultrasonic image obtained after the previous diffusion treatment, and div is a divergence operator; Is a gradient operator, gradient value, time operator, initial iteration image U 3 and diffusion controlling factor k 2.
Preferably, the process of performing fusion reconstruction on the second-layer ultrasound sampling image, the second-layer ultrasound enhancement image and the third-layer ultrasound up-sampling image, and up-sampling the fused and reconstructed ultrasound image to obtain the second-layer ultrasound up-sampling image includes:
Carrying out fusion reconstruction on the second-layer ultrasonic sampling image I 2, the second-layer ultrasonic enhancement image Fu 2 and the third-layer ultrasonic fusion reconstruction image R 3 according to a preset image weighted fusion relation F2＝w1Fu2+w2R3-sign(w1Fu2+w2R3-Th2)w3I2, to obtain a second-layer ultrasonic weighted fusion image F 2, wherein Th 2 is a given threshold value, and w 1、w2 and w 3 are predefined coefficients;
And up-sampling the second-layer ultrasonic weighted fusion image F 2 to obtain a second-layer ultrasonic up-sampling image U 2.
Preferably, the process of performing fusion reconstruction on the first layer ultrasonic sampling image and the second layer ultrasonic up-sampling image to obtain a final ultrasonic enhancement image includes:
And according to a preset image fusion relation E=gammaU 2+(1-γ)I1, carrying out fusion reconstruction on the first-layer ultrasonic sampling image I 1 and the second-layer ultrasonic up-sampling image U 2 to obtain a final ultrasonic enhanced image E, wherein gammais a control factor.
In order to solve the technical problem, the invention also provides an ultrasonic image optimization processing system, which comprises:
the downsampling module is used for downsampling the original ultrasonic image to obtain ultrasonic sampling images with different resolutions;
An image determining module for determining an original ultrasonic image and a pre-edge enhanced ultrasonic image from the ultrasonic sampling image;
the image enhancement module is used for carrying out edge enhancement processing on the ultrasonic image with the pre-edge enhancement according to a preset image edge enhancement strategy to obtain an ultrasonic enhancement image;
The image fusion module is used for fusing the ultrasonic image with the pre-edge enhancement and the ultrasonic enhancement image to obtain an ultrasonic fusion image;
And the image reconstruction module is used for upsampling the ultrasonic fusion image to obtain an upsampled image, and carrying out fusion reconstruction on the upsampled image and the original ultrasonic image to obtain a final ultrasonic enhanced image.
In order to solve the technical problem, the invention also provides an ultrasonic image optimization processing device, which comprises:
A memory for storing a computer program;
a processor for implementing the steps of any of the ultrasound image optimization processing methods described above when executing the computer program.
The application provides an ultrasonic image optimization processing method, which comprises the steps of downsampling an original ultrasonic image to obtain ultrasonic sampling images with different resolutions, determining the original ultrasonic image and a pre-edge-enhanced ultrasonic image, carrying out edge enhancement processing on the pre-edge-enhanced ultrasonic image according to a preset image edge enhancement strategy to obtain an ultrasonic enhancement image, fusing the pre-edge-enhanced ultrasonic image and the ultrasonic enhancement image to obtain an ultrasonic fusion image, upsampling the ultrasonic fusion image to obtain an upsampled image, and fusing and reconstructing the upsampled image and the original ultrasonic image to obtain a final ultrasonic enhancement image. Therefore, the application not only can strengthen the image edge, but also can make the processed image more natural, and can meet the real-time requirement of clinic on the method.
The invention also provides an ultrasonic image optimization processing system and device, which have the same beneficial effects as the optimization processing method.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required in the prior art and the embodiments will be briefly described below, and it is apparent that the drawings in the following description are only some embodiments of the present invention, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flowchart of an ultrasound image optimization processing method provided in an embodiment of the present invention;
FIG. 2 is a schematic diagram of an ultrasonic image optimization process according to an embodiment of the present invention;
fig. 3 is a schematic structural diagram of an ultrasound image optimization processing system according to an embodiment of the present invention.
Detailed Description
The core of the invention is to provide an ultrasonic image optimization processing method, an ultrasonic image optimization processing system and an ultrasonic image optimization processing device, which not only can strengthen the edge of an image, but also can make the processed image more natural, and can meet the real-time requirement of clinic on the method.
For the purpose of making the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention, and it is apparent that the described embodiments are some embodiments of the present invention, but not all embodiments of the present invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Referring to fig. 1, fig. 1 is a flowchart of an ultrasound image optimization processing method according to an embodiment of the present invention.
The ultrasonic image optimization processing method comprises the following steps:
step S1, downsampling an original ultrasonic image to obtain ultrasonic sampling images with different resolutions, and determining the original ultrasonic image and the ultrasonic image with the pre-edge enhancement.
Specifically, firstly, an original ultrasonic image to be optimized is acquired, then, the acquired original ultrasonic image is downsampled, the main purpose is to obtain ultrasonic sampling images with different resolutions, and the secondary purpose is to reduce the preprocessing data volume of the ultrasonic image so as to reduce the subsequent image processing time.
More specifically, the down-sampling of the original ultrasound image may be performed by 1) down-sampling the original ultrasound image based on a spatial domain (direct extraction), specifically, extracting "rows and columns" of the original image data by a certain number of units to obtain an ultrasound sampling image, e.g., 1,2, k, extracting a sample point every 2 units in one-dimensional data, where the extracted sample is 1,3, until the complete data is extracted. 2) And based on the spatial domain downsampling (post-filtering extraction mode), firstly, filtering the image original data to smooth the image and inhibit noise, and then extracting the filtered image original data in rows and columns at certain unit intervals to obtain an ultrasonic sampling image. Of course, the application can also adopt other image downsampling modes to obtain ultrasonic sampling images with different resolutions, and the application is not limited in particular.
Based on this, after obtaining ultrasound sampling images of different resolutions, the present application determines an original ultrasound image (i.e., an ultrasound sampling image without edge enhancement) and a pre-edge-enhanced ultrasound image from the ultrasound sampling images for subsequent image edge enhancement processing. More specifically, the edge enhancement of the ultrasonic sampling images with different resolutions can be set as that the ultrasonic sampling images with higher resolutions do not need to be enhanced, and the ultrasonic sampling images with lower resolutions need to be enhanced, so that the subsequent image fusion effect can be improved, and the application is not particularly limited herein and is set according to practical situations.
And S2, performing edge enhancement processing on the ultrasonic image with the pre-edge enhancement according to a preset image edge enhancement strategy to obtain an ultrasonic enhanced image.
Specifically, an image edge enhancement strategy for guiding the edge enhancement operation of an ultrasonic image is required to be set in advance, and after the ultrasonic image with the pre-edge enhancement is determined, the ultrasonic image with the pre-edge enhancement is subjected to edge enhancement processing according to the set image edge enhancement strategy, so that an ultrasonic enhancement image is obtained.
And step S3, fusing the ultrasonic image with the pre-edge enhanced and the ultrasonic enhanced image to obtain an ultrasonic fused image.
Specifically, after an ultrasonic enhanced image is obtained, the ultrasonic image before edge enhancement processing and the ultrasonic enhanced image obtained after the edge enhancement processing are fused to obtain an ultrasonic fusion image for subsequent image reconstruction.
And S4, up-sampling the ultrasonic fusion image to obtain an up-sampled image, and carrying out fusion reconstruction on the up-sampled image and the original ultrasonic image to obtain a final ultrasonic enhanced image.
Specifically, after an ultrasonic fusion image is obtained, the application carries out up-sampling on the ultrasonic fusion image to obtain an up-sampling image, and then carries out fusion reconstruction on the up-sampling image and an original ultrasonic image to obtain a final ultrasonic enhancement image. Therefore, the application synthesizes the ultrasonic images with different resolutions to reconstruct the ultrasonic image, not only can strengthen the edge of the image, but also can make the processed image more natural, and can meet the real-time requirement of clinic on the method.
The application provides a quality improvement method of an ultrasonic image, which comprises the steps of downsampling an original ultrasonic image to obtain ultrasonic sampling images with different resolutions, determining the original ultrasonic image and a pre-edge-enhanced ultrasonic image, carrying out edge enhancement treatment on the pre-edge-enhanced ultrasonic image according to a preset image edge enhancement strategy to obtain the ultrasonic enhancement image, fusing the pre-edge-enhanced ultrasonic image and the ultrasonic enhancement image to obtain an ultrasonic fusion image, upsampling the ultrasonic fusion image to obtain an upsampled image, and fusing and reconstructing the upsampled image and the original ultrasonic image to obtain a final ultrasonic enhancement image. Therefore, the application not only can strengthen the image edge, but also can make the processed image more natural, and can meet the real-time requirement of clinic on the method.
Based on the above embodiments:
referring to fig. 2, fig. 2 is a schematic diagram of an ultrasonic image optimization processing according to an embodiment of the present invention.
As an alternative embodiment, the process of downsampling an original ultrasound image to obtain ultrasound sampled images with different resolutions, and determining the original ultrasound image and the pre-edge enhanced ultrasound image includes:
three layers of downsampling are carried out on an original ultrasonic image to obtain three layers of ultrasonic sampling images with different resolutions, wherein the resolution of a first layer of ultrasonic sampling image is greater than the resolution of a second layer of ultrasonic sampling image and greater than the resolution of a third layer of ultrasonic sampling image;
Taking the first layer ultrasonic sampling image as an original ultrasonic image, and taking the second layer ultrasonic sampling image and the third layer ultrasonic sampling image as pre-edge enhanced ultrasonic images;
correspondingly, the process of performing edge enhancement processing on the pre-edge enhanced ultrasonic image according to a preset image edge enhancement strategy to obtain an ultrasonic enhanced image comprises the following steps:
performing edge enhancement processing on the second-layer ultrasonic sampling image according to a preset image edge enhancement strategy to obtain a second-layer ultrasonic enhancement image;
Performing edge enhancement processing on the third-layer ultrasonic sampling image according to a preset image edge enhancement strategy to obtain a third-layer ultrasonic enhancement image;
the process of carrying out up-sampling on the ultrasonic fusion image to obtain an up-sampling image, and carrying out fusion reconstruction on the up-sampling image and the original ultrasonic image to obtain a final ultrasonic enhancement image comprises the following steps:
Fusing the third-layer ultrasonic sampling image with the third-layer ultrasonic enhancement image, and upsampling the fused ultrasonic image to obtain a third-layer ultrasonic upsampling image;
Carrying out fusion reconstruction on the second-layer ultrasonic sampling image, the second-layer ultrasonic enhancement image and the third-layer ultrasonic upsampling image, and upsampling the fused reconstructed ultrasonic image to obtain a second-layer ultrasonic upsampling image;
and carrying out fusion reconstruction on the first layer ultrasonic sampling image and the second layer ultrasonic up-sampling image to obtain a final ultrasonic enhanced image.
Specifically, the application performs three-layer downsampling on the original ultrasonic image to correspondingly obtain three-layer ultrasonic sampling images with different resolutions, wherein the resolutions are a first-layer ultrasonic sampling image I 1, a second-layer ultrasonic sampling image I 2 and a third-layer ultrasonic sampling image I 3 in sequence from high to low.
In consideration of the fact that the higher-resolution ultrasonic sampling image does not need edge enhancement, the lower-resolution ultrasonic sampling image does need edge enhancement, and the subsequent image fusion effect is improved, the method specifically takes the first-layer ultrasonic sampling image as an original ultrasonic image, namely, the ultrasonic image which does not need edge enhancement, and takes the second-layer ultrasonic sampling image and the third-layer ultrasonic sampling image as pre-edge enhancement ultrasonic images. The method comprises the steps of carrying out edge enhancement processing on a second-layer ultrasonic sampling image to obtain a second-layer ultrasonic enhancement image, and carrying out edge enhancement processing on a third-layer ultrasonic sampling image to obtain a third-layer ultrasonic enhancement image.
Based on the above, as shown in fig. 2, the ultrasonic image processing and reconstructing process specifically comprises 1) fusing a third layer ultrasonic sampling image before edge enhancement of the third layer image with a third layer ultrasonic enhancement image obtained after edge enhancement to obtain a fused ultrasonic image, and upsampling the fused ultrasonic image to obtain a third layer ultrasonic upsampling image, 2) fusing and reconstructing the second layer ultrasonic sampling image before edge enhancement of the second layer image and the second layer ultrasonic enhancement image obtained after edge enhancement together with the third layer ultrasonic upsampling image to obtain a fused reconstructed ultrasonic image, upsampling the fused reconstructed ultrasonic image to obtain a second layer ultrasonic upsampling image, and 3) fusing and reconstructing the first layer ultrasonic sampling image and the second layer ultrasonic upsampling image to obtain a final ultrasonic enhancement image.
As an optional embodiment, the process of performing edge enhancement processing on the third layer of ultrasonic sampling image according to a preset image edge enhancement policy to obtain a third layer of ultrasonic enhancement image includes:
and carrying out multi-scale edge enhancement processing on the third layer ultrasonic sampling image to obtain a third layer ultrasonic enhancement image.
In particular, considering that the ultrasonic sampling image with lower resolution is more suitable for a multi-scale edge enhancement processing mode, the application particularly carries out multi-scale edge enhancement processing on the ultrasonic sampling image I 3 of the third layer to obtain the ultrasonic enhancement image of the third layer.
As an alternative embodiment, the process of performing multi-scale edge enhancement processing on the third layer ultrasonic sampling image to obtain the third layer ultrasonic enhancement image includes:
Carrying out isotropic filtering denoising treatment on the ultrasonic image subjected to multi-scale edge enhancement treatment to obtain a denoised ultrasonic image The method comprises the steps of carrying out multi-scale edge enhancement in an iterative mode, wherein n is the number of iterations, and the maximum value of n is equal to a preset iteration threshold value;
Ultrasound imaging Respectively inputting the filter to a pre-constructed odd filter and an even filter to obtain odd filter response and even filter response on multiple scales, and calculating the difference between the odd filter response and the even filter response to obtain edge response on multiple scales;
Fusing the edge responses of the multiple scales to obtain an edge image, and performing image pair ultrasonic image based on the edge image Performing multi-scale edge enhancement processing and image denoising processing to obtain an ultrasonic image
When the iteration number of the multi-scale edge enhancement is smaller than a preset iteration threshold, the ultrasonic image is obtainedAnd when the iteration number of the multi-scale edge enhancement is equal to a preset iteration threshold value, stopping the iteration.
Specifically, the present application will not be described in detail with reference to the following embodiments.
As an alternative embodiment, the difference between the odd and even filter responses is calculated to obtain multi-scale edge responses, the multi-scale edge responses are fused to obtain an edge image, and the ultrasonic image is imaged based on the edge imagePerforming multi-scale edge enhancement processing and image denoising processing to obtain an ultrasonic imageComprises the following steps:
According to the edge response relation Calculating the difference between the odd and even filter responses to obtain an edge response on the scale sWherein, For the even filter response on scale s; T s is a noise threshold value related to the scale s, x is a pixel point of the image;
based on the response fusion relationship The edge responses of a plurality of scales are fused to obtain an edge image Eg 3, wherein,The value range of (1) is 0,1, m is the number of filter scales;
Enhancing relationships from multiple scales of images For ultrasound imagesPerforming multi-scale edge enhancement processing and image denoising processing to obtain an ultrasonic imageWhere Δt is the time step and δ 3 is the scale factor.
Specifically, the present embodiment introduces the principle of multi-scale image edge enhancement:
at present, the traditional image edge enhancement method generally detects the image edge through the change of gray information, namely gradient, but in an ultrasonic image, the image edge detection based on gray can fail due to the influence of adverse factors such as uneven gray, large noise and the like, so the application adopts a phase consistency method to detect the image edge, namely utilizes the phase information of the ultrasonic image, thus the gray characteristics such as the step characteristics of the ultrasonic image can be detected, the Mach-number phenomenon of the ultrasonic image can be detected, and the reliability of the image edge detection is higher.
The specific principle of the phase-based image edge detection method is that an image is subjected to filtering processing with an odd filter and an even filter respectively, and when the response difference of the odd filter and the even filter is large, the pixel point is often positioned on the edge of the image. More specifically, the odd filter and the even filter each have a plurality of scales, and applying the odd filter and the even filter to the image edge detection can obtain multi-scale edge responses, and then fusing the multi-scale edge responses can obtain the edge of the image, namely, an edge image.
Based on the method, the application sets the image multi-scale enhancement relation in advance The method is based on the setting principle that multi-scale image edge enhancement is carried out in an iteration mode, and when the number of iterations is equal to a preset iteration threshold value, iteration is stopped. More specifically, when the third layer of ultrasonic sampling image is subjected to the nth iteration, the multi-scale image edge enhancement process comprises 1) performing isotropic filtering denoising treatment on the ultrasonic image obtained from the last iteration, for example, performing image filtering treatment by isotropic filtering modes such as mean filtering and median filtering, so as to remove noise in the ultrasonic image and obtain a denoised ultrasonic image2) Ultrasound imagingRespectively inputting to a pre-constructed odd filter and even filter to obtain odd filter response and even filter response on multiple scales, 3) according to the edge response relation Calculating the difference between the odd filter response and the even filter response to obtain an edge response at the s-th scale at the nth iteration4) Based on the response fusion relationshipFusing edge responses of multiple scales to obtain an edge image corresponding to the third layer ultrasonic sampling image in the nth iteration5) Enhancing relationships from multiple scales of images Performing ultrasound imagingPerforming multi-scale edge enhancement processing and image denoising processing to obtain an ultrasonic image after the nth iterationThereby enabling multi-scale image edge enhancement.
As an optional embodiment, the process of performing edge enhancement processing on the second layer of ultrasonic sampling image according to a preset image edge enhancement policy to obtain a second layer of ultrasonic enhancement image includes:
And performing multidirectional edge enhancement processing on the second-layer ultrasonic sampling image to obtain a second-layer ultrasonic enhancement image.
Specifically, considering that the ultrasonic sampling image with higher resolution is more suitable for a multidirectional edge enhancement processing mode, the application specifically carries out multidirectional edge enhancement processing on the second-layer ultrasonic sampling image I 2 to obtain a second-layer ultrasonic enhancement image.
In summary, the application provides different edge enhancement processing modes for the ultrasonic images with different layers of pre-edge enhancement, and specifically comprises a multidirectional edge enhancement processing mode and a multi-scale edge enhancement processing mode.
As an alternative embodiment, the process of performing multidirectional edge enhancement processing on the second-layer ultrasonic sampling image to obtain a second-layer ultrasonic enhancement image includes:
multidirectional enhancement of relationships according to preset images Performing multidirectional edge enhancement processing and image denoising processing on the second-layer ultrasonic sampling image within a preset edge enhancement time T1 to obtain a second-layer ultrasonic enhancement image;
Wherein, The method comprises the steps of obtaining a gradient operator, wherein I is a gradient value, t is a time operator, k 1 is a parameter for controlling smoothing, u 0 is an initial iteration image, u is an ultrasonic image obtained after diffusion treatment for the last time, beta is a super parameter, beta is epsilon [0,1], eg is an edge operator based on phase, the value range is [0,1], x is a pixel point of the image, g (x) divides the ultrasonic image into two areas, omega 1 = { x epsilon omega, g (x)/(1 } and omega 2＝{x∈Ω,0≤g(x)<1},Ω＝Ω1+Ω2, epsilon represents a normal constant, s is a certain scale of a filter, m is the number of the filter scale, and F s is a phase-based filter characteristic on an s scale; for the even filter response on scale s; Is the odd filter response at scale s, and T s is a noise threshold associated with scale s.
Specifically, the present embodiment introduces the principle of multi-directional image edge enhancement:
Currently, multi-directional image edge enhancement is generally performed using an anisotropic diffusion model represented by PM (Perona-Malik algorithm), and image edges can be preserved while smoothing the image. Specifically, the basic principle of the PM model is to control diffusion smoothing by using a gradient operator, when the gradient value is large, the image is made to realize weaker smoothing in the area by using a smaller diffusion coefficient in a diffusion equation, so that the effect of maintaining the edge is achieved, and conversely, when the gradient value is small, the image is made to realize larger smoothing in the area by using a larger diffusion coefficient in the diffusion equation. However, the gradient operator of the PM model is obtained through the calculation of the change of the gray information of the image, and the ultrasonic image is very sensitive to the gradient operator based on the change of the gray due to the influences of adverse factors such as uneven gray, large noise and the like, so that the image edge enhancement effect of the PM model is often poor for the ultrasonic image.
In order to solve the above problem, the anisotropic model proposed by the present application adopts a phase-based edge operator to construct a diffusion equation, which is mathematically defined as: where Eg is a phase-based edge operator, expressed mathematically as: (ε plays a role of avoiding division by zero) g (x) divides the ultrasound image into two regions, Ω 1 = { x εΩ, g (x) ≡1} and Ω 2＝{x∈Ω,0≤g(x)<1},Ω＝Ω1+Ω2.
Meanwhile, considering that the PM model only considers the influence of local gradient, so that the overall effect of image enhancement is poor, the anisotropic model provided by the application comprehensively considers the influence of local gradient and all image changes, and the mathematical definition is as follows: thereby improving the overall effect of image enhancement.
In addition, in the image edge enhancement scheme, the enhancement of the image edge is well realized, the image noise is removed while the image edge is enhanced, and the effect is good.
As an alternative embodiment, the process of fusing the third layer ultrasound sampling image with the third layer ultrasound enhancement image, and upsampling the fused ultrasound image to obtain a third layer ultrasound upsampling image includes:
According to the preset image coarse fusion relation Performing coarse fusion on the third-layer ultrasonic sampling image I 3 and the third-layer ultrasonic enhancement image Eh 3 to obtain a third-layer ultrasonic coarse fusion image Fu 3, wherein Th 3 is a given threshold value, mu is a parameter for controlling the image edge enhancement degree, alpha is a parameter for controlling the image enhancement degree, eg 3 is an edge image of the third-layer ultrasonic sampling image, and the value range is [0,1];
Up-sampling the third-layer ultrasonic rough fusion image Fu 3 to obtain a third-layer ultrasonic up-sampling image U 3;
according to the preset image fine fusion relation Performing diffusion treatment on the third-layer ultrasonic up-sampling image U 3 within a preset fine fusion time T2 to obtain a third-layer ultrasonic fusion reconstruction image R 3, wherein U is an ultrasonic image obtained after the previous diffusion treatment; Is a gradient operator, gradient value, time operator, initial iteration image U 3 and diffusion controlling factor k 2.
Specifically, the process of reconstructing the third layer ultrasonic image is specifically completed by adopting the modes of image coarse fusion, image up-sampling and image fine fusion:
1) Image coarse fusion the image coarse fusion process of the application not only considers the third layer ultrasonic enhanced image Eh 3 and the third layer ultrasonic sampling image I 3 after diffusion treatment, but also considers the edge image Eg 3 of the third layer ultrasonic image. When the value of the edge image is larger than a certain threshold value, the importance of the edge image needs to be enhanced, namely the third layer ultrasonic enhancement image Eh 3 is fused with the third layer ultrasonic sampling image I 3 with larger weight, whereas when the value of the edge image is smaller than a certain threshold value, the importance of the edge image needs to be weakened, namely the third layer ultrasonic enhancement image Eh 3 is fused with the third layer ultrasonic sampling image I 3 with smaller weight. Specifically, the fusion weight of the third layer ultrasound enhancement image Eh 3 is designed to be positively correlated with the edge image Eg 3, and when the value of the edge image Eg 3 is larger, the fusion weight of the third layer ultrasound enhancement image Eh 3 is also larger.
Based on the above, according to the edge image Eg 3, the third layer ultrasonic enhancement image Eh 3 and the third layer ultrasonic sampling image I 3 are subjected to coarse fusion, so as to obtain a third layer ultrasonic coarse fusion image Fu 3: Where sign is a sign function, th 3 is a given threshold, when Eg 3-Th3>0,sign(Eg3-Th3) takes positive values, otherwise takes negative values.
2) And (3) image upsampling, namely upsampling the third-layer ultrasonic rough fusion image Fu 3 to obtain a third-layer ultrasonic upsampling image U 3, wherein upsampling modes such as cubic, bilinear can be adopted.
3) Image fine fusion, wherein the image up-sampling mode belongs to a linear model, and is simple and easy to realize, but the details of the high-frequency recovery capability are limited. In the medical image field, doctors need more image details to make accurate judgment. The detail of the image belongs to the region with larger gray value change, but noise also appears in the region, so the application utilizes a diffusion model, adopts detail information such as gradient information to achieve the purpose of improving the resolution of the image, does not amplify the noise, and reduces the influence of the noise on the reconstructed image. Meanwhile, considering that the change of the global image plays a very important role in the integrity of the reconstructed image, the diffusion model of the application not only considers the change of the local gradient, but also considers the change of the global gray information, and is mathematically defined as: thereby, the third layer ultrasonic up-sampling image U 3 is diffused to obtain a third layer ultrasonic fusion reconstructed image R 3.
As an optional embodiment, the process of performing fusion reconstruction on the second layer of ultrasonic sampling image, the second layer of ultrasonic enhancement image and the third layer of ultrasonic upsampling image, and upsampling the fused reconstructed ultrasonic image to obtain the second layer of ultrasonic upsampling image includes:
Carrying out fusion reconstruction on the second-layer ultrasonic sampling image I 2, the second-layer ultrasonic enhancement image Fu 2 and the third-layer ultrasonic fusion reconstruction image R 3 according to a preset image weighted fusion relation F2＝w1Fu2+w2R3-sign(w1Fu2+w2R3-Th2)w3I2, to obtain a second-layer ultrasonic weighted fusion image F 2, wherein Th 2 is a given threshold value, and w 1、w2 and w 3 are predefined coefficients;
And up-sampling the second-layer ultrasonic weighted fusion image F 2 to obtain a second-layer ultrasonic up-sampling image U 2.
Specifically, the process of reconstructing the second-layer ultrasonic image is specifically completed by adopting an image weighted fusion and image upsampling mode:
1) The image weighted fusion comprises the steps of carrying out weighted fusion reconstruction on a second-layer ultrasonic enhanced image Fu 2 and a third-layer ultrasonic fusion reconstructed image R 3, reducing the influence of a second-layer ultrasonic sampling image I 2 if the image after weighted fusion reconstruction is larger than a certain specified threshold value, otherwise, increasing the influence of a second-layer ultrasonic sampling image I 2, wherein the mathematical definition of the fused second-layer ultrasonic weighted fusion image F 2 is ：F2＝w1Fu2+w2R3-sign(w1Fu2+w2R3-Th2)w3I2,, sign is a sign function, when w 1Fu2+w2R3-Th2 is larger than 0, sign takes a positive value, otherwise, sign takes a negative value, both w 1、w2 and w 3 can be defined by adopting the ratio of the mean value to the variance of the image, specifically, w 1 is defined by adopting the ratio of the mean value to the variance of the second-layer ultrasonic enhanced image Fu 2, w 2 is defined by adopting the ratio of the mean value to the variance of the third-layer ultrasonic fusion reconstructed image R 3, and w 3 is defined by adopting the ratio of the mean value to the variance of the second-layer ultrasonic sampling image I 2, or w 1、w2 and w 3 can be specified by a user.
2) And (3) image upsampling, namely upsampling the second-layer ultrasonic weighted fusion image F 2 to obtain a second-layer ultrasonic upsampling image U 2, wherein the upsampling mode can be adopted specifically as the upsampling mode of the third-layer ultrasonic image.
As an alternative embodiment, the process of performing fusion reconstruction on the first layer ultrasonic sampling image and the second layer ultrasonic up-sampling image to obtain a final ultrasonic enhancement image includes:
And according to a preset image fusion relation E=gammaU 2+(1-γ)I1, carrying out fusion reconstruction on the first layer ultrasonic sampling image I 1 and the second layer ultrasonic up-sampling image U 2 to obtain a final ultrasonic enhanced image E, wherein gammais a control factor.
Specifically, the first layer ultrasonic sampling image I 1 and the second layer ultrasonic up-sampling image U 2 are fused and reconstructed to obtain a final ultrasonic enhancement image E, and the mathematical definition of the ultrasonic enhancement image E obtained after fusion is specifically that E=γU 2+(1-γ)I1.
In summary, the application adopts a mechanism from bottom to top, synthesizes images with three layers of resolutions to reconstruct a high-resolution image, can well remove image noise, enhance image edges, improve image signal to noise ratio and increase image contrast, has higher algorithm realizability of a multi-resolution strategy, and in addition, images with different layers of resolutions contain different image details, so that the images can be amplified without losing the image details, and the diagnosis of doctors is facilitated.
Referring to fig. 3, fig. 3 is a schematic structural diagram of an ultrasound image optimization processing system according to an embodiment of the present invention.
The ultrasonic image optimization processing system comprises:
the downsampling module 1 is used for downsampling the original ultrasonic image to obtain ultrasonic sampling images with different resolutions;
An image determining module 2 for determining an original ultrasound image and a pre-edge enhanced ultrasound image from the ultrasound sampling image;
The image enhancement module 3 is used for carrying out edge enhancement processing on the ultrasonic image with the pre-edge enhancement according to a preset image edge enhancement strategy to obtain an ultrasonic enhancement image;
The image fusion module 4 is used for fusing the ultrasonic image with the pre-edge enhanced and the ultrasonic enhanced image to obtain an ultrasonic fusion image;
and the image reconstruction module 5 is used for upsampling the ultrasonic fusion image to obtain an upsampled image, and performing fusion reconstruction on the upsampled image and the original ultrasonic image to obtain a final ultrasonic enhanced image.
The description of the optimizing processing system provided by the present application refers to the embodiment of the optimizing processing method, and the present application is not repeated here.
The application also provides an ultrasonic image optimization processing device, which comprises:
A memory for storing a computer program;
A processor for implementing the steps of any of the ultrasound image optimization processing methods described above when executing a computer program.
The description of the optimizing device provided by the present application refers to the embodiment of the optimizing method, and the disclosure is not repeated here.
It should also be noted that in this specification, relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises an element.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (12)
1. An ultrasonic image optimization processing method is characterized by comprising the following steps:
Downsampling an original ultrasonic image to obtain ultrasonic sampling images with different resolutions, and determining the original ultrasonic image and the ultrasonic image with the pre-edge enhancement;
the process of down-sampling the original ultrasonic image to obtain ultrasonic sampling images with different resolutions and determining the original ultrasonic image and the ultrasonic image with the enhanced pre-edge comprises the following steps:
three layers of downsampling are carried out on an original ultrasonic image to obtain three layers of ultrasonic sampling images with different resolutions, wherein the resolution of a first layer of ultrasonic sampling image is greater than the resolution of a second layer of ultrasonic sampling image and greater than the resolution of a third layer of ultrasonic sampling image;
taking the first layer ultrasonic sampling image as an original ultrasonic image, and taking the second layer ultrasonic sampling image and the third layer ultrasonic sampling image as pre-edge enhanced ultrasonic images;
Performing edge enhancement processing on the ultrasonic image with the pre-edge enhancement according to a preset image edge enhancement strategy to obtain an ultrasonic enhancement image;
Fusing the pre-edge enhanced ultrasonic image and the ultrasonic enhanced image to obtain an ultrasonic fused image;
Up-sampling the ultrasonic fusion image to obtain an up-sampling image, and carrying out fusion reconstruction on the up-sampling image and the original ultrasonic image to obtain a final ultrasonic enhanced image;
Performing edge enhancement processing on the second-layer ultrasonic sampling image to obtain a second-layer ultrasonic enhancement image;
Correspondingly, the process of fusing the pre-edge enhanced ultrasonic image and the ultrasonic enhanced image to obtain an ultrasonic fused image, up-sampling the ultrasonic fused image to obtain an up-sampled image, and fusing and reconstructing the up-sampled image and the original ultrasonic image to obtain a final ultrasonic enhanced image comprises the following steps:
fusing the third-layer ultrasonic sampling image with the third-layer ultrasonic enhancement image, and upsampling the fused ultrasonic image to obtain a third-layer ultrasonic upsampling image;
Performing fusion reconstruction on the second-layer ultrasonic sampling image, the second-layer ultrasonic enhancement image and the third-layer ultrasonic upsampling image, and upsampling the fused reconstructed ultrasonic image to obtain a second-layer ultrasonic upsampling image;
and carrying out fusion reconstruction on the first layer ultrasonic sampling image and the second layer ultrasonic up-sampling image to obtain a final ultrasonic enhancement image.
2. The method for optimizing an ultrasound image according to claim 1, wherein the process of performing edge enhancement processing on the pre-edge enhanced ultrasound image according to a preset image edge enhancement policy to obtain an ultrasound enhanced image comprises:
Performing edge enhancement processing on the second-layer ultrasonic sampling image according to a preset image edge enhancement strategy to obtain a second-layer ultrasonic enhancement image;
And carrying out edge enhancement processing on the third-layer ultrasonic sampling image according to a preset image edge enhancement strategy to obtain a third-layer ultrasonic enhancement image.
3. The method for optimizing an ultrasound image according to claim 2, wherein the process of performing edge enhancement processing on the third layer ultrasound sampling image according to a preset image edge enhancement policy to obtain a third layer ultrasound enhancement image comprises:
and carrying out multi-scale edge enhancement processing on the third layer ultrasonic sampling image to obtain a third layer ultrasonic enhancement image.
4. The method for optimizing an ultrasound image according to claim 3, wherein the step of performing multi-scale edge enhancement processing on the third layer ultrasound sample image to obtain a third layer ultrasound enhanced image comprises:
Carrying out isotropic filtering denoising treatment on the ultrasonic image subjected to multi-scale edge enhancement treatment to obtain a denoised ultrasonic image The method comprises the steps of carrying out multi-scale edge enhancement in an iterative mode, wherein n is the number of iterations, and the maximum value of n is equal to a preset iteration threshold value;
Ultrasound imaging Respectively inputting the filter to a pre-constructed odd filter and an even filter to obtain odd filter response and even filter response on multiple scales, and calculating the difference between the odd filter response and the even filter response to obtain edge response on multiple scales;
Fusing the edge responses of the multiple scales to obtain an edge image, and performing image pair ultrasonic imaging based on the edge image Performing multi-scale edge enhancement processing and image denoising processing to obtain an ultrasonic image
When the iteration number of the multi-scale edge enhancement is smaller than a preset iteration threshold, the ultrasonic image is obtainedAnd when the iteration number of the multi-scale edge enhancement is equal to a preset iteration threshold value, stopping the iteration.
5. The method of claim 4, wherein the difference between the odd and even filter responses is calculated to obtain multi-scale edge responses, the multi-scale edge responses are fused to obtain an edge image, and the ultrasound image is imaged based on the edge imagePerforming multi-scale edge enhancement processing and image denoising processing to obtain an ultrasonic imageComprises the following steps:
According to the edge response relation Calculating the difference between the odd and even filter responses to obtain an edge response on the scale sWherein, For the even filter response on scale s; T s is a noise threshold value related to the scale s, x is a pixel point of the image;
based on the response fusion relationship The edge responses of a plurality of scales are fused to obtain an edge image Eg 3, wherein,The value range of (1) is 0,1, m is the number of filter scales;
Enhancing relationships from multiple scales of images For ultrasound imagesPerforming multi-scale edge enhancement processing and image denoising processing to obtain an ultrasonic imageWhere Δt is the time step and δ 3 is the scale factor.
6. The method for optimizing an ultrasound image according to claim 2, wherein the process of performing edge enhancement processing on the second-layer ultrasound sampling image according to a preset image edge enhancement policy to obtain a second-layer ultrasound enhancement image comprises:
And performing multidirectional edge enhancement processing on the second-layer ultrasonic sampling image to obtain a second-layer ultrasonic enhancement image.
7. The method for optimizing an ultrasound image according to claim 6, wherein the step of performing multi-directional edge enhancement processing on the second-layer ultrasound sample image to obtain a second-layer ultrasound enhanced image comprises:
multidirectional enhancement of relationships according to preset images Performing multidirectional edge enhancement processing and image denoising processing on the second-layer ultrasonic sampling image within a preset edge enhancement time T1 to obtain a second-layer ultrasonic enhancement image;
Wherein, The method comprises the steps of obtaining a gradient operator, wherein I is a gradient value, t is a time operator, k 1 is a parameter for controlling smoothing, u 0 is an initial iteration image, u is an ultrasonic image obtained after diffusion treatment for the last time, beta is a super parameter, beta is epsilon [0,1], eg is an edge operator based on phase, the value range is [0,1], x is a pixel point of the image, g (x) divides the ultrasonic image into two areas, omega 1 = { x epsilon omega, g (x)/(1 } and omega 2＝{x∈Ω,0≤g(x)<1},Ω＝Ω1+Ω2, epsilon represents a normal constant, s is a certain scale of a filter, m is the number of the filter scale, and F s is a phase-based filter characteristic on an s scale; for the even filter response on scale s; Is the odd filter response at scale s, and T s is a noise threshold associated with scale s.
8. The method for optimizing an ultrasound image according to claim 2, wherein the process of fusing the third-layer ultrasound sampling image with the third-layer ultrasound-enhanced image and upsampling the fused ultrasound image to obtain a third-layer ultrasound upsampled image includes:
According to the preset image coarse fusion relation Performing coarse fusion on the third-layer ultrasonic sampling image I 3 and the third-layer ultrasonic enhancement image Eh 3 to obtain a third-layer ultrasonic coarse fusion image Fu 3, wherein Th 3 is a given threshold value, mu is a parameter for controlling the image edge enhancement degree, alpha is a parameter for controlling the image enhancement degree, eg 3 is an edge image of the third-layer ultrasonic sampling image, and the value range is [0,1];
up-sampling the third-layer ultrasonic rough fusion image Fu 3 to obtain a third-layer ultrasonic up-sampling image U 3;
according to the preset image fine fusion relation Performing diffusion treatment on the third-layer ultrasonic up-sampling image U 3 within a preset fine fusion time T2 to obtain a third-layer ultrasonic fusion reconstruction image R 3, wherein U is an ultrasonic image obtained after the previous diffusion treatment, and div is a divergence operator; Is a gradient operator, the gradient value is t is a time operator, U 3 is an initial iteration image, k 2 is a factor for controlling diffusion, and c is a diffusion equation.
9. The method for optimizing an ultrasound image according to claim 8, wherein the process of performing fusion reconstruction on the second-layer ultrasound sampling image, the second-layer ultrasound enhancement image, and the third-layer ultrasound upsampling image, and upsampling the fusion reconstructed ultrasound image to obtain the second-layer ultrasound upsampling image includes:
Carrying out fusion reconstruction on the second-layer ultrasonic sampling image I 2, the second-layer ultrasonic enhancement image Fu 2 and the third-layer ultrasonic fusion reconstruction image R 3 according to a preset image weighted fusion relation F2＝w1Fu2+w2R3-sign(w1Fu2+w2R3-Th2)w3I2, to obtain a second-layer ultrasonic weighted fusion image F 2, wherein Th 2 is a given threshold value, and w 1、w2 and w 3 are predefined coefficients;
And up-sampling the second-layer ultrasonic weighted fusion image F 2 to obtain a second-layer ultrasonic up-sampling image U 2.
10. The method of optimizing ultrasound images according to claim 9, wherein the process of performing fusion reconstruction on the first layer ultrasound sampling image and the second layer ultrasound up-sampling image to obtain a final ultrasound enhanced image includes:
And according to a preset image fusion relation E=gammaU 2+(1-γ)I1, carrying out fusion reconstruction on the first-layer ultrasonic sampling image I 1 and the second-layer ultrasonic up-sampling image U 2 to obtain a final ultrasonic enhanced image E, wherein gammais a control factor.
11. An ultrasound image optimization processing system, comprising:
the downsampling module is used for downsampling the original ultrasonic image to obtain ultrasonic sampling images with different resolutions;
An image determining module for determining an original ultrasonic image and a pre-edge enhanced ultrasonic image from the ultrasonic sampling image;
wherein, the image determination module is used for:
three layers of downsampling are carried out on an original ultrasonic image to obtain three layers of ultrasonic sampling images with different resolutions, wherein the resolution of a first layer of ultrasonic sampling image is greater than the resolution of a second layer of ultrasonic sampling image and greater than the resolution of a third layer of ultrasonic sampling image;
taking the first layer ultrasonic sampling image as an original ultrasonic image, and taking the second layer ultrasonic sampling image and the third layer ultrasonic sampling image as pre-edge enhanced ultrasonic images;
The image enhancement module is used for carrying out edge enhancement processing on the ultrasonic image with the pre-edge enhancement according to a preset image edge enhancement strategy to obtain an ultrasonic enhancement image, wherein the second-layer ultrasonic sampling image is subjected to edge enhancement processing to obtain a second-layer ultrasonic enhancement image;
The image fusion module is used for fusing the ultrasonic image with the pre-edge enhancement and the ultrasonic enhancement image to obtain an ultrasonic fusion image;
the image reconstruction module is used for upsampling the ultrasonic fusion image to obtain an upsampled image, and carrying out fusion reconstruction on the upsampled image and the original ultrasonic image to obtain a final ultrasonic enhanced image;
The image fusion module and the image reconstruction module are used for:
Fusing the third-layer ultrasonic sampling image with the third-layer ultrasonic enhancement image, and upsampling the fused ultrasonic image to obtain a third-layer ultrasonic upsampling image;
Performing fusion reconstruction on the second-layer ultrasonic sampling image, the second-layer ultrasonic enhancement image and the third-layer ultrasonic upsampling image, and upsampling the fused reconstructed ultrasonic image to obtain a second-layer ultrasonic upsampling image;
and carrying out fusion reconstruction on the first layer ultrasonic sampling image and the second layer ultrasonic up-sampling image to obtain a final ultrasonic enhancement image.
12. An ultrasound image optimization processing device, characterized by comprising:
A memory for storing a computer program;
a processor for implementing the steps of the ultrasound image optimization processing method according to any one of claims 1-10 when executing the computer program.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010724086.8A CN111861929B (en) | 2020-07-24 | 2020-07-24 | Ultrasonic image optimization processing method, system and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010724086.8A CN111861929B (en) | 2020-07-24 | 2020-07-24 | Ultrasonic image optimization processing method, system and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111861929A CN111861929A (en) | 2020-10-30 |
| CN111861929B true CN111861929B (en) | 2025-01-03 |
Family
ID=72949511
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010724086.8A Active CN111861929B (en) | 2020-07-24 | 2020-07-24 | Ultrasonic image optimization processing method, system and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111861929B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114022444B (en) * | 2021-11-04 | 2025-07-01 | 青岛海信医疗设备股份有限公司 | Ultrasonic image processing method and device |
| CN114820378A (en) * | 2022-05-12 | 2022-07-29 | 亳州联岐医疗科技有限公司 | Artificial intelligence-based ultrasonic image optimization method |
| CN116977266A (en) * | 2023-04-13 | 2023-10-31 | 浙江易时科技股份有限公司 | Canopy defect detection method and system based on small sample image feature extraction |
| CN116523810B (en) * | 2023-07-04 | 2023-11-17 | 深圳开立生物医疗科技股份有限公司 | Ultrasonic image processing method, device, equipment and medium |
| CN117115465A (en) * | 2023-08-30 | 2023-11-24 | 杭州一隅千象科技有限公司 | Multi-channel high-resolution edge fusion method |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1860500A (en) * | 2003-09-22 | 2006-11-08 | 皇家飞利浦电子股份有限公司 | Enhancing medical images with temporal filter |
| CN108389214A (en) * | 2018-03-06 | 2018-08-10 | 青岛海信医疗设备股份有限公司 | The processing method and processing device of ultrasonoscopy, electronic equipment, storage medium |
Family Cites Families (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP5832737B2 (en) * | 2010-11-01 | 2015-12-16 | 株式会社東芝 | Ultrasonic diagnostic apparatus and ultrasonic image processing apparatus |
| ITGE20120048A1 (en) * | 2012-05-04 | 2013-11-05 | Esaote Spa | METHOD OF RECONSTRUCTION OF BIOMEDICAL IMAGES |
| KR101533968B1 (en) * | 2014-02-14 | 2015-07-06 | 서강대학교산학협력단 | Picture quality improvement method and system of ultrasonic medical image based on multiscale |
| US9225889B1 (en) * | 2014-08-18 | 2015-12-29 | Entropix, Inc. | Photographic image acquisition device and method |
| CN108205804B (en) * | 2016-12-16 | 2022-05-31 | 斑马智行网络(香港)有限公司 | Image processing method and device and electronic equipment |
| CN109584192B (en) * | 2018-10-19 | 2021-05-25 | 中国人民解放军海军工程大学 | Target feature enhancement method, device and electronic device based on multispectral fusion |
| CN111179275B (en) * | 2019-12-31 | 2023-04-25 | 电子科技大学 | Medical ultrasonic image segmentation method |
-
2020
- 2020-07-24 CN CN202010724086.8A patent/CN111861929B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1860500A (en) * | 2003-09-22 | 2006-11-08 | 皇家飞利浦电子股份有限公司 | Enhancing medical images with temporal filter |
| CN108389214A (en) * | 2018-03-06 | 2018-08-10 | 青岛海信医疗设备股份有限公司 | The processing method and processing device of ultrasonoscopy, electronic equipment, storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111861929A (en) | 2020-10-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111861929B (en) | Ultrasonic image optimization processing method, system and device | |
| CN109410219B (en) | Image segmentation method and device based on pyramid fusion learning and computer readable storage medium | |
| Abbasi et al. | Optical coherence tomography retinal image reconstruction via nonlocal weighted sparse representation | |
| US20130202177A1 (en) | Non-linear resolution reduction for medical imagery | |
| JP2019504659A (en) | Automated cardiac volume segmentation | |
| CN106339998A (en) | Multi-focus image fusion method based on contrast pyramid transformation | |
| Liu et al. | Medical image resolution enhancement for healthcare using nonlocal self-similarity and low-rank prior | |
| US20150371372A1 (en) | System and method for medical image quality enhancement using multiscale total variation flow | |
| Liu et al. | Automatic blur-kernel-size estimation for motion deblurring | |
| US20250191124A1 (en) | Machine learning enabled restoration of low resolution images | |
| Salvado et al. | Partial volume reduction by interpolation with reverse diffusion | |
| Liang | Medical image fusion with deep neural networks | |
| CN115312198A (en) | Deep learning brain tumor prognosis analysis modeling method and system combining attention mechanism and multi-scale feature mining | |
| CN113487536A (en) | Image segmentation method, computer device and storage medium | |
| CN114529473A (en) | Image reconstruction method, image reconstruction device, electronic apparatus, storage medium, and computer program | |
| CN117635426A (en) | A super-resolution reconstruction method for magnetic resonance images based on multi-scale attention | |
| CN111563843B (en) | Image super-resolution reconstruction method, system and related device | |
| CN111091504B (en) | Image offset field correction method, computer device, and storage medium | |
| Lu et al. | Non-convex joint bilateral guided depth upsampling | |
| CN112561802B (en) | Interpolation method of continuous sequence images, interpolation model training method and system thereof | |
| Dzyubachyk et al. | Automated algorithm for reconstruction of the complete spine from multistation 7T MR data | |
| Kezele et al. | Spherical wavelet transform for ODF sharpening | |
| CN117953027A (en) | DWI-FLAIR mismatch evaluation method, device, medium and product | |
| CN110895790A (en) | Scene image super-resolution method based on posterior degradation information estimation | |
| CN113538299B (en) | Ultrasonic image denoising method, device, equipment and computer readable storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| TG01 | Patent term adjustment | ||
| TG01 | Patent term adjustment |