An ultrasound image enhancement method and system

Abstract

This specification discloses an ultrasound image enhancement method and system. The ultrasound image enhancement method includes: acquiring an ultrasound image; determining the structure tensor of the ultrasound image; determining the diffusion tensor, image mask, and radiation pattern of the ultrasound image based on the structure tensor; performing speckle and noise suppression processing on the ultrasound image based on the diffusion tensor to obtain a first image; performing edge enhancement processing on the ultrasound image based on the image mask and the radiation pattern to obtain a second image; and fusing the first image and the second image to obtain a target image.

Description

Technical Field

[0001] This specification relates to the field of image processing, and in particular to a method and system for ultrasound image enhancement. Background Technology

[0002] Medical images acquired by ultrasound imaging systems typically undergo a series of image processing steps before being used by medical professionals for diagnosis. Multiplicative speckle noise in ultrasound images can mask minute details and reduce the detectability of low-contrast lesions, while image enhancement can significantly improve image quality. Existing ultrasound image denoising methods mainly include nonlocal mean filtering algorithms, algorithms based on anisotropic diffusion, and algorithms based on multi-scale filtering. Existing ultrasound image edge enhancement algorithms mainly include algorithms based on improved unsharpened masks and enhancement algorithms based on directional filtering. However, these algorithms suffer from poor adaptability, coarse mask results, and high computational costs.

[0003] Therefore, it is necessary to provide an ultrasound image enhancement method that can significantly improve image quality, reduce computational load, and meet real-time requirements. Summary of the Invention

[0004] One embodiment of this specification provides an ultrasound image enhancement method. The ultrasound image enhancement method includes: acquiring an ultrasound image; determining the structure tensor of the ultrasound image; determining the diffusion tensor, image mask, and radiation pattern of the ultrasound image based on the structure tensor; performing speckle and noise suppression processing on the ultrasound image based on the diffusion tensor to obtain a first image; performing edge enhancement processing on the ultrasound image based on the image mask and the radiation pattern to obtain a second image; and fusing the first image and the second image to obtain a target image.

[0005] One embodiment of this specification provides an ultrasound image enhancement system. The ultrasound image enhancement system includes: an acquisition module for acquiring an ultrasound image; a determination module for determining the structure tensor of the ultrasound image, and determining the diffusion tensor, image mask, and radiation pattern of the ultrasound image based on the structure tensor; a speckle noise suppression module for performing speckle noise suppression processing on the ultrasound image based on the diffusion tensor to obtain a first image; an edge enhancement module for performing edge enhancement processing on the ultrasound image based on the image mask and the radiation pattern to obtain a second image; and an image fusion module for fusing the first image and the second image to obtain a target image.

[0006] One embodiment of this specification provides a computer-readable storage medium that stores computer instructions. When a computer reads the computer instructions from the storage medium, the computer executes the method described in any embodiment of this specification.

[0007] The ultrasound image enhancement method, system, and computer-readable storage medium provided in the embodiments of this specification have the following advantages compared with the prior art: (1) Parallel computation based on structure tensors, while simultaneously performing speckle noise suppression and edge enhancement processing on the image, can accelerate computation and reduce repetitive computation, effectively improving the efficiency of ultrasound image processing; (2) Obtaining the image mask through structure tensor calculation can yield accurate filtering regions of interest, avoiding unnecessary computation and improving computational and filtering efficiency; (3) Obtaining the orientation map through structure tensor calculation can reduce the computational load of the orientation map, and the corrected orientation map contains accurate pixel orientation information, which can significantly improve the edge enhancement effect; (4) Fusing the result image after edge enhancement processing and the result image after speckle noise suppression processing can effectively avoid introducing too much noise into the edge enhancement result. Attached Figure Description

[0008] This specification will be further described by way of exemplary embodiments, which will be described in detail with reference to the accompanying drawings. These embodiments are not limiting; in these embodiments, the same reference numerals denote the same structures, wherein:

[0009] Figure 1 These are schematic diagrams illustrating application scenarios of the ultrasound image enhancement system according to some embodiments of this specification;

[0010] Figure 2 This is a block diagram of an ultrasound image enhancement system according to some embodiments of this specification;

[0011] Figure 3 This is an exemplary flowchart of an ultrasound image enhancement method according to some embodiments of this specification;

[0012] Figure 4 This is an exemplary flowchart of a speckle noise suppression method according to some embodiments of this specification;

[0013] Figure 5 This is an exemplary flowchart of an edge enhancement method according to some embodiments of this specification; and

[0014] Figure 6A and Figure 6B These are comparison images of exemplary original ultrasound images and exemplary enhanced ultrasound images shown in some embodiments of this specification. Detailed Implementation

[0015] To more clearly illustrate the technical solutions of the embodiments in this specification, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are merely some examples or embodiments of this specification. For those skilled in the art, these drawings can be applied to other similar scenarios without creative effort. Unless obvious from the context or otherwise specified, the same reference numerals in the drawings represent the same structures or operations.

[0016] It should be understood that the terms “system,” “device,” “unit,” and / or “module” used herein are one way to distinguish different components, elements, parts, sections, or assemblies at different levels. However, if other terms can achieve the same purpose, they may be replaced by other expressions.

[0017] As indicated in this specification and claims, unless the context clearly indicates otherwise, the words "a," "an," "an," and / or "the" do not specifically refer to the singular and may also include the plural. Generally speaking, the terms "comprising" and "including" only indicate the inclusion of expressly identified steps and elements, which do not constitute an exclusive list, and the method or apparatus may also include other steps or elements.

[0018] Flowcharts are used in this specification to illustrate the operations performed by the system according to embodiments of this specification. It should be understood that the preceding or following operations are not necessarily performed in exact order. Instead, the steps can be processed in reverse order or simultaneously. Furthermore, other operations can be added to these processes, or one or more steps can be removed from them.

[0019] Figure 1 This is a schematic diagram illustrating an application scenario of an ultrasound image enhancement system according to some embodiments of this specification.

[0020] Figure 1 This is a diagram illustrating application scenarios of the ultrasound image enhancement system 100 according to some embodiments of this application. For example... Figure 1 As shown, the ultrasound image enhancement system 100 may include an ultrasound device 110, a network 120, a terminal 130, a processing device 140, and a storage device 150.

[0021] Ultrasound device 110 can be used to perform scans on an object for diagnostic imaging. Ultrasound device 110 can be used to view image information of the object's internal tissues to assist doctors in disease diagnosis. Ultrasound device 110 can transmit high-frequency sound waves (e.g., ultrasound waves) to the object via a probe to generate ultrasound images. In some embodiments, ultrasound device 110 may include an ultrasound pulse-echo imaging device, an ultrasound echo Doppler imaging device, an ultrasound electronic endoscope, an ultrasound Doppler blood flow analysis device, an ultrasound human tissue measurement device, etc. In some embodiments, the object may include biological and / or non-biological objects. For example, the object may include specific parts of the human body, such as the neck, chest, abdomen, etc., or combinations thereof. Another example is that the object may be a patient to be scanned by ultrasound device 110. In some embodiments, the object being scanned can undergo ultrasound examination in any position, such as supine, lateral, prone, semi-recumbent, or sitting. In some embodiments, the scanning mode of ultrasound device 110 may include A-mode ultrasound, B-mode ultrasound, M-mode ultrasound, and / or D-mode ultrasound, etc. In some embodiments, the ultrasound device 110 may be installed in healthcare facilities or locations, such as medical examination centers, wards, delivery rooms, examination rooms, operating rooms, emergency rooms, ambulances, etc. In some embodiments, the ultrasound device 110 may be installed in other locations, such as marathon tracks, extreme sports venues, racing tracks, disaster relief sites, etc. In some embodiments, the ultrasound device 110 may also receive imaging object information and / or imaging operation instruction information sent from the terminal 130 or processing device 140 via the network 120, and may send intermediate imaging result data or imaging images to the processing device 140, storage device 150, or terminal 130.

[0022] Network 120 may include any suitable network that facilitates the exchange of information and / or data between the ultrasound image enhancement system 100 and the ultrasound image enhancement system 100. In some embodiments, one or more other components of the ultrasound image enhancement system 100 (e.g., ultrasound device 110, terminal 130, processing device 140, storage device 150, etc.) may exchange information and / or data with each other via network 120. For example, processing device 140 may acquire image data from ultrasound device 110 via network 120. As another example, processing device 140 may acquire user instructions from terminal 130 via network 120. Network 120 may be and / or include public networks (e.g., the Internet), private networks (e.g., local area networks (LANs), wide area networks (WANs), etc.), wired networks (e.g., Ethernet), wireless networks (e.g., 802.11 networks, Wi-Fi networks, etc.), cellular networks (e.g., LTE networks), Frame Relay networks, virtual private networks (“VPNs”), satellite networks, telephone networks, routers, hubs, converters, server computers, and / or combinations thereof. For example, network 120 may include cable networks, wired networks, fiber optic networks, telecommunications networks, local area networks (LANs), wireless LANs (WLANs), metropolitan area networks (MANs), public switched telephone networks (PSTNs), and Bluetooth. TM Network, ZigBee TM A combination of one or more of the following: a network, a near-field communication network (NFC), etc. In some embodiments, network 120 may include one or more network access points. For example, network 120 may include wired and / or wireless network access points, such as base stations and / or network switching points, through which one or more components of system 100 may access network 120 for data and / or information exchange.

[0023] In some embodiments, a user can operate the ultrasound image enhancement system 100 through a terminal 130. The terminal 130 may include one or more combinations of mobile devices 131, tablet computers 132, laptop computers 133, etc. In some embodiments, the fused target image can be presented to the user through the terminal 130. In some embodiments, the mobile device 131 may include one or more combinations of smart home devices, wearable devices, mobile devices, virtual reality devices, augmented reality devices, etc. In some embodiments, smart home devices may include one or more combinations of smart lighting devices, smart appliance control devices, smart monitoring devices, smart TVs, smart cameras, walkie-talkies, etc. In some embodiments, wearable devices may include one or more combinations of bracelets, shoes and socks, glasses, helmets, watches, clothing, backpacks, smart accessories, etc. In some embodiments, mobile devices may include one or more combinations of mobile phones, personal digital assistants (PDAs), gaming devices, navigation devices, point-of-sale (POS) devices, laptop computers, tablet computers, desktop computers, etc. In some embodiments, virtual reality devices and / or augmented reality devices may include one or more combinations of virtual reality helmets, virtual reality glasses, virtual reality goggles, augmented reality helmets, augmented reality glasses, augmented reality goggles, etc. For example, virtual reality devices and / or augmented reality devices may include Google Glass. TM Oculus Rift TM HoloLens TM GearVR TM In some embodiments, terminal 130 may be part of processing device 140. In some embodiments, terminal 130 may be part of ultrasound device 110.

[0024] Processing device 140 can process data and / or information obtained from ultrasound device 110, terminal 130, and / or storage device 150. For example, processing device 140 can acquire ultrasound images from ultrasound device 110 and perform speckle noise suppression and / or edge enhancement processing on the ultrasound images. In some embodiments, processing device 140 can be a server or a server group. The server group can be centralized or distributed. In some embodiments, processing device 140 can be local or remote. For example, processing device 140 can access information and / or data stored in ultrasound device 110, terminal 130, and / or storage device 150 via network 120. For example, processing device 140 can directly connect to ultrasound device 110, terminal 130, and / or storage device 150 to access the information and / or data stored therein. In some embodiments, processing device 140 can be executed on a cloud platform. For example, the cloud platform can include one or more combinations of private cloud, public cloud, hybrid cloud, community cloud, distributed cloud, interconnected cloud, multi-cloud, etc. In some embodiments, the processing device 140 may be a computing device 200 having one or more components (such as...). Figure 2 The process is performed as described above. In some embodiments, the processing device 140 may be part of the ultrasound device 110 or the terminal 130.

[0025] Storage device 150 may store data, instructions, and / or other information. In some embodiments, storage device 150 may store data obtained from terminal 130 and / or processing device 140. In some embodiments, storage device 150 may store data and / or instructions executed or used by processing device 140 for performing the exemplary methods described in this application. In some embodiments, storage device 150 may include one or more combinations of mass storage, removable storage, volatile read-write storage, read-only storage (ROM), etc. Exemplary mass storage may include disks, optical disks, solid-state drives, etc. Exemplary removable storage may include flash drives, floppy disks, optical disks, memory cards, zipper disks, magnetic tapes, etc. Exemplary volatile read-write storage may include random access memory (RAM). Exemplary RAM may include dynamic random access memory (DRAM), dual data rate synchronous dynamic random access memory (DDR SDRAM), static random access memory (SRAM), thyristor random access memory (T-RAM), and zero capacitance random access memory (Z-RAM), etc. Exemplary read-only memory (ROM) may include mask read-only memory (MROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), optical disc read-only memory (CD-ROM), and digital multipurpose optical disc, etc. In some embodiments, storage device 150 may be implemented on a cloud platform. For example, the cloud platform may include one or more combinations of private cloud, public cloud, hybrid cloud, community cloud, distributed cloud, interconnected cloud, multi-cloud, etc.

[0026] In some embodiments, storage device 150 may be connected to network 120 to communicate with one or more other components in system 100 (e.g., processing device 140, terminal 130, etc.). One or more components in ultrasound image enhancement system 100 may access data or instructions stored in storage device 150 via network 120. In some embodiments, storage device 150 may be directly connected to or communicate with one or more other components in system 100 (e.g., processing device 140, terminal 130, etc.). In some embodiments, storage device 150 may be part of processing device 140.

[0027] Figure 2 This is a block diagram of an ultrasound image enhancement system according to some embodiments of this specification.

[0028] like Figure 2As shown, the ultrasound image enhancement system 200 may include an acquisition module 210, a determination module 220, a speckle noise suppression module 230, an edge enhancement module 240, and an image fusion module 250. In some embodiments, the acquisition module 210, the determination module 220, the speckle noise suppression module 230, the edge enhancement module 240, and the image fusion module 250 may be implemented by the processing device 140.

[0029] The acquisition module 210 can be used to acquire ultrasound images. In some embodiments, ultrasound images obtained from scanning an object by an ultrasound device can be acquired. Further details regarding the acquisition of ultrasound images can be found in the detailed description of step 310, and will not be repeated here.

[0030] The determining module 220 can be used to determine the structural tensor of an ultrasound image, and / or determine the diffusion tensor, image mask, and / or radiation pattern of the ultrasound image based on the structural tensor. In some embodiments, the determining module 220 can determine the eigenvectors and / or eigenvalues ​​of the structural tensor, and / or determine the diffusion tensor based on the eigenvectors and / or eigenvalues. In some embodiments, the determining module 220 can determine the eigenvalues ​​of the structural tensor, and / or determine the image mask based on the eigenvalues ​​of the structural tensor. In some embodiments, the determining module 220 can determine the eigenvalues ​​of the structural tensor, and / or determine the radiation pattern based on the eigenvalues ​​of the structural tensor. Further details regarding the determination of the structural tensor, diffusion tensor, image mask, and radiation pattern can be found in the detailed descriptions of steps 320, 330, 420, 520, and 530, and will not be repeated here.

[0031] The speckle noise suppression module 230 can be used to perform speckle noise suppression processing on the ultrasound image based on the diffusion tensor to obtain a first image. In some embodiments, the speckle noise suppression module 230 can apply the diffusion tensor to the ultrasound image to obtain the first image. For more details on obtaining the first image, please refer to the detailed description of step 340, which will not be repeated here.

[0032] The edge enhancement module 240 can be used to perform edge enhancement processing on the ultrasound image based on an image mask and / or a radiation pattern to obtain a second image. In some embodiments, the edge enhancement module 240 can perform Gabor filtering on the ultrasound image based on the image mask and the radiation pattern to obtain the second image. For more details on obtaining the second image, please refer to the detailed description of steps 350 and 540, which will not be repeated here.

[0033] The image fusion module 250 can be used to fuse the first image and the second image to obtain a target image. In some embodiments, the image fusion module 250 can filter the second image and fuse the first image and the filtered second image to obtain the target image. For more details on fusing the first image and the second image, please refer to the detailed description of step 360, which will not be repeated here.

[0034] It should be noted that the above description of the ultrasound image enhancement system 200 and its modules is for convenience only and should not be construed as limiting this specification to the embodiments described. It is understood that those skilled in the art, after understanding the principle of the system, may arbitrarily combine the various modules or construct subsystems connected to other modules without departing from this principle. For example, in some embodiments, for example, Figure 2 The acquisition module 210, determination module 220, speckle noise suppression module 230, edge enhancement module 240, and image fusion module 250 disclosed herein can be different modules within a single system, or a single module can implement the functions of two or more of the aforementioned modules. For example, the modules can share a single storage module, or each module can have its own separate storage module. Such variations are all within the scope of protection of this specification.

[0035] Figure 3 This is an exemplary flowchart of an ultrasound image enhancement method according to some embodiments of this specification.

[0036] Process 300 can be executed by a processing device (e.g., processing device 140). For example, process 300 can be implemented as a set of instructions (e.g., an application program) stored in a memory external to, and accessible by, the ultrasound image enhancement system 100, such as storage device 150. The processing device can execute the instruction set and, when executing the instructions, can be configured to execute process 300. The operational schematic diagram of process 300 presented below is illustrative. In some embodiments, the process can be accomplished using one or more additional operations not described and / or one or more operations not discussed. Additionally, Figure 3 The order of operations shown in and described below in process 300 is not intended to be restrictive.

[0037] Step 310: Acquire an ultrasound image. In some embodiments, step 310 may be performed by a processing device or an acquisition module 210.

[0038] In some embodiments, ultrasound images that require enhancement can be acquired. These ultrasound images can be used to view image information of internal tissues of a subject to assist doctors in disease diagnosis. Because noise in unprocessed ultrasound images (e.g., speckle noise) can mask and reduce certain details in the image, severely affecting the quality of the ultrasound image, it is necessary to perform speckle noise suppression and / or edge enhancement on the ultrasound images to improve their quality.

[0039] In some embodiments, ultrasound images obtained from a subject scanned by an ultrasound device can be acquired. In some embodiments, the subject may include biological and / or non-biological objects. For example, the subject may include specific parts of a human body, such as the neck, chest, abdomen, etc., or combinations thereof. Another example is the patient to be scanned by the ultrasound device 110. In some embodiments, the subject being scanned may be in any position for ultrasound examination, such as supine, lateral, prone, semi-recumbent, or sitting.

[0040] In some embodiments, the ultrasound image may be a real-time image, i.e., an image obtained in real time by scanning with an ultrasound device. In some embodiments, the ultrasound image may also be an image stored in the storage device 150.

[0041] Step 320: Determine the structural tensor of the ultrasound image. In some embodiments, step 320 may be performed by a processing device or a determination module 220.

[0042] A structure tensor can be a representation of the spatial information of an ultrasound image, used to distinguish between flat regions, edge regions, and corner regions. In some embodiments, the structure tensor of an ultrasound image can be presented in matrix form, represented by local contours and gradient directions, and the relative contrast between them. The relative contrast represents the degree of change between the local contours and gradient directions.

[0043] In some embodiments, the structural tensor and its eigendecomposition can be expressed as the following formula (1):

[0044]

[0045] Among them, J ρ (I) represents the structure tensor matrix of the input image I, K ρ This represents the Gaussian convolution kernel function with variance ρ, * denotes convolution, and I x I y Let w1 and w2 represent the gradient matrix along the x-axis or y-axis, w1 and w2 be the eigenvectors of the structure tensor, μ1 and μ2 be the two eigenvalues ​​of the structure tensor, and T denote the transpose.

[0046] In some embodiments, a coordinate system can be established in the ultrasound image with the midpoint (or other point) of the image, the x-axis direction being the horizontal direction in the ultrasound image coordinate system, the y-axis direction being the vertical direction in the ultrasound image coordinate system, and the x-axis and y-axis being perpendicular to each other.

[0047] In some embodiments, the Gaussian convolution kernel function K ρ The parameters can be set as empirical values ​​or adjusted according to the actual situation.

[0048] In some embodiments, the eigenvectors of the structure tensor may include a first eigenvector w1 and a second eigenvector w2, corresponding to the gradient direction and the contour direction, respectively. In some embodiments, the eigenvalues ​​of the structure tensor may include a first eigenvalue μ1 and a second eigenvalue μ2, representing the difference between the gradient direction and the contour direction, respectively. In some embodiments, the structure tensor of the ultrasound image can be obtained by combining the first eigenvector w1, the second eigenvector w2, the first eigenvalue μ1, and the second eigenvalue μ2 (e.g., as shown in formula (1)). In some embodiments, the eigenvalues ​​and eigenvectors of the structure tensor can be obtained by eigenvalue decomposition of a matrix. In some embodiments, the eigenvalues ​​and eigenvectors of the structure tensor can be solved by methods such as inverse power iteration, Sturm sequence method and inverse iteration, and synchronous iteration.

[0049] In some embodiments, the eigenvectors and eigenvalues ​​of the structural tensor are both matrices, and each pixel in the ultrasound image can correspond to a value in the matrix, that is, the size of the eigenvector and eigenvalue matrix is ​​consistent with the size of the ultrasound image.

[0050] In some embodiments, the formula (1) for the structure tensor can be derived by a nonlinear anisotropic diffusion model, which can be expressed as the following formula (2):

[0051]

[0052] Where D is the diffusion tensor, representing the amount of diffusion in the gradient and contour directions, I is the input image (e.g., the ultrasound image obtained in step 310), ΔI represents the gradient of the input image I, and div is the divergence (i.e., a vector operator that maps a vector field in vector space to a scalar field). For the derivative, I(x,y,t) represents the ultrasound image of the corresponding pixel (x,y) at time t. The diffusion tensor D can be determined using the structure tensor.

[0053] Step 330: Based on the structure tensor, determine the diffusion tensor, image mask, and radiation pattern of the ultrasound image. In some embodiments, step 330 may be performed by a processing device or a determination module 220.

[0054] The diffusion tensor of an ultrasound image can represent the amount of diffusion along the gradient and contour directions. In some embodiments, the diffusion tensor can be determined based on the eigenvalues ​​of the structure tensor.

[0055] In some embodiments, the eigenvectors and eigenvalues ​​of the structure tensor can be determined first, wherein the eigenvectors include a first eigenvector w1 and a second eigenvector w2, and the eigenvalues ​​may include a first eigenvalue μ1 and a second eigenvalue μ2. In some embodiments, based on the first eigenvalue μ1 and the second eigenvalue μ2, a ​​third eigenvalue λ1 and a fourth eigenvalue λ2 can be determined. In some embodiments, based on the first eigenvector w1, the second eigenvector w1, and the determined third eigenvalue λ1 and fourth eigenvalue λ2, the diffusion tensor of the ultrasound image can be determined.

[0056] In some embodiments, the calculation of the diffusion tensor mainly depends on the determination of eigenvalues. The third eigenvalue λ1 and the fourth eigenvalue λ2 can be determined based on the first eigenvalue μ1 and the second eigenvalue μ2 of the structure tensor, modified according to different situations. Since anisotropic diffusion can suppress image noise while preserving image edge information, the eigenvalues ​​of the structure tensor can be modified as needed to enhance image edges during diffusion. There are various methods for solving the third eigenvalue λ1 and the fourth eigenvalue λ2 based on the first eigenvalue μ1 and the second eigenvalue μ2, which are not limited here. For more information on determining the diffusion tensor based on the structure tensor, please refer to the relevant description in step 420, which will not be described in detail here.

[0057] An image mask can be a binary image that represents the edge information of one or more regions of interest (e.g., organs, tissues, etc.) in an ultrasound image. In the embodiments of this specification, the image mask can be a binary image with the same size as the ultrasound image, and can be represented by a matrix, which can include two values: 0 and 1. 0 represents a uniform tissue region that requires no processing, while 1 represents a non-uniform tissue region (i.e., an edge region) that needs enhancement. In some embodiments, accurate filtering regions of interest can be obtained through image mask calculation, filtering out some unwanted edge information, avoiding unnecessary computation, and improving both computational accuracy and filtering efficiency. For ultrasound images, due to the high frame rate requirements (e.g., cardiac ultrasound images can reach 100 frames per second or higher), the efficiency improvement of image masks is significant.

[0058] In some embodiments, the image mask of the ultrasound image can be determined by determining the eigenvalues ​​of the structure tensor and based on these eigenvalues. Calculating the image mask of the ultrasound image using the eigenvalues ​​of the structure tensor in some embodiments, compared to calculations using methods such as standard deviation, can significantly reduce the computational load of image processing. Further details on determining the image mask based on the structure tensor can be found in the description of step 520, and will not be elaborated upon here.

[0059] An ultrasound image's orientation map can be used to represent the orientation information of each pixel in the image. In some embodiments, the orientation map can be determined by identifying the eigenvectors of a structure tensor and then determining the orientation map based on those eigenvectors.

[0060] In some embodiments, the first feature vector w1 and the second feature vector w2 can represent the directions of maximum and minimum change in feature values, respectively. The direction information of the pixel can be obtained by using the arctangent, thereby determining the orientation pattern of the ultrasound image.

[0061] In some embodiments, determining the orientation pattern based on feature vectors requires less computation than methods based on gradient calculation, thus resulting in higher computational efficiency without affecting the image edge enhancement results. For ultrasound images, due to the high frame rate requirements (e.g., cardiac ultrasound images can reach 100 frames per second or higher), the efficiency improvement of calculating the orientation pattern based on feature vectors is significant. More details on determining the orientation pattern based on the structure tensor can be found in the description of step 530, and will not be elaborated upon here.

[0062] It should be understood that the methods described in the embodiments of this specification can maximize the computational efficiency of image enhancement processing. Without affecting the enhancement effect, the higher the computational efficiency, the higher the applicable image frame rate, and the wider the range of practical applications.

[0063] Step 340: Based on the diffusion tensor, perform speckle noise suppression processing on the ultrasound image to obtain a first image. In some embodiments, step 340 may be performed by a processing device or a speckle noise suppression module 230.

[0064] In some embodiments, the diffusion tensor can be applied to the ultrasound image as a direction-selective diffusion factor, that is, the ultrasound image is diffused in a corresponding direction based on the diffusion tensor to obtain a first image. The first image is the image after speckle noise suppression processing. Performing speckle noise suppression processing on the ultrasound image can reduce noise (e.g., speckle noise) in the ultrasound image, thereby improving the quality of the ultrasound image.

[0065] In some embodiments, image diffusion based on a diffusion tensor can yield a noise-suppressed image. In some embodiments, the diffusion tensor can be represented as a matrix with the same size as the ultrasound image, and the values ​​in the diffusion tensor matrix can include both positive and negative numbers. In some embodiments, the value corresponding to each pixel in the diffusion tensor can be applied as a diffusion factor to the corresponding pixel in the original ultrasound image to obtain the noise-suppressed result.

[0066] Step 350: Based on the image mask and the radiation pattern, edge enhancement processing is performed on the ultrasound image to obtain a second image. In some embodiments, step 350 may be performed by a processing device or an edge enhancement module 240.

[0067] In some embodiments, Gabor filtering can be applied to the ultrasound image based on an image mask and a radiation pattern to obtain a second image. The second image is an image after edge enhancement processing. Noise in ultrasound images can affect image quality (e.g., blurring edges and details), therefore, in addition to speckle noise suppression, edge enhancement processing is also required for ultrasound images.

[0068] In some embodiments, Gabor filtering exhibits frequency and direction selectivity, thus providing robustness for edge or texture enhancement. Texture refers to visual features in an image other than color, such as the grayscale distribution of pixels and their surrounding areas. In some embodiments, the general form of the Gabor filter can be expressed as formula (3):

[0069]

[0070] Where h represents the Gabor filter, φ represents the direction of the Gabor filter, f represents the texture frequency information, and δ x and δ y It is the spatial constant of the Gaussian envelope along the x and y axes, (x,y) represents a pixel, and exp represents an exponential function.

[0071] In some embodiments, Gabor filtering of an image requires specifying three parameters: a frequency map, a direction map, and a Gaussian envelope. The frequency map represents the texture frequency information f at each pixel. In some embodiments, the frequency map can be determined based on the sinusoidal plane wave of the texture, or it can be determined experimentally. In some embodiments, δ x and δ y The choice involves equilibrium, if δ x and δ y The larger the value chosen, the stronger the filter's robustness to noise, but it may generate false edges. x and δ yThe smaller the value of δ, the worse the filtering effect will be. Therefore, it is necessary to consider all factors when choosing an appropriate δ value. x and δ y In some embodiments, δ x and δ y It can be determined based on empirical values, or it can be determined experimentally. In some embodiments, the texture frequency information f can be set based on empirical values, or it can be adjusted according to specific circumstances.

[0072] In some embodiments, the image mask and orientation map determined in step 330 can be used to guide Gabor filtering, and the filtering process can be expressed as formula (4):

[0073]

[0074] Where E is the filtered result image, E(i,j) represents the filtered result at pixel (i,j), M is the image mask, M(i,j) represents the edge information at pixel (i,j), O is the direction map, O(i,j) represents the direction information at pixel (i,j), F is the frequency map, F(i,j) represents the texture frequency information at pixel (i,j), I is the original ultrasound image, s is the Gabor filter kernel size, h is the Gabor filter generation function, and a and b are variables of the summation function. In some embodiments, the Gabor filter kernel size s is frequency-dependent and is not adjusted as a parameter.

[0075] In some embodiments, the edge-enhanced result image E, i.e., the second image, can be obtained by performing Gabor filtering on the original ultrasound image based on the image mask and the orientation map.

[0076] Step 360: Fuse the first image and the second image to obtain the target image. In some embodiments, step 360 may be performed by a processing device or an image fusion module 250.

[0077] In some embodiments, fusing the result image after speckle noise suppression (i.e., the first image) and the result image after edge enhancement (i.e., the second image) yields the final result image (i.e., the target image). In some embodiments, the target image can be obtained by filtering the second image and fusing the first image and the filtered second image. The target image is the image after speckle noise suppression and edge enhancement processing. The target image generated after fusion can significantly improve image quality and effectively avoid introducing excessive noise. In some embodiments, speckle noise suppression and edge enhancement processing can be performed in parallel, which can accelerate calculations and reduce repetitive operations, effectively improving the efficiency of ultrasound image processing.

[0078] The final image is obtained by fusing the noise suppression result image and the edge enhancement result image. The image fusion process can be expressed as formula (5):

[0079] H=(1-cw)R+w·findContours(E) (5)

[0080] Where H represents the fusion result image, R represents the speckle noise suppression result image, findContours represents the function for filtering the edge enhancement result image, E represents the edge enhancement result image, c is the coefficient, and w represents the edge enhancement weight.

[0081] In some embodiments, edge enhancement typically only needs to amplify strong texture regions. To avoid introducing excessive noise during the edge enhancement process, further screening of the edge enhancement results is necessary. In some embodiments, if noise is introduced into the edge-enhanced image, for example, by enhancing a lot of unnecessary information, directly fusing this image with a noise-suppressed image may result in bright spots in the image, significantly affecting the quality of the ultrasound image. In some embodiments, edge-enhanced images can be screened using morphological methods, such as based on the shape, size, color, etc., of objects (e.g., regions of interest, connected components, etc.) in the edge-enhanced image. In some embodiments, morphological operations, such as erosion, dilation, opening, and closing operations, can be performed on the objects in the edge-enhanced image during the screening process. In some embodiments, edge-enhanced images can be screened based on features such as contour area and / or contour perimeter. In some embodiments, the edge-enhanced image may contain one or more edge information, each edge information being independent. The contour area and / or contour perimeter information of each edge in the edge-enhanced image can be statistically analyzed and screened. As an example only, if the contour area and / or perimeter of an edge is very small (e.g., less than a preset threshold), its corresponding edge can be removed. In some embodiments, machine learning can also be used to classify contours, thereby filtering edge-enhanced images.

[0082] In some embodiments, the final result image can be obtained by weighted summation of the noise suppression result image and the filtered edge enhancement result image, as shown in formula (5). The coefficient c and the edge enhancement weight w jointly determine the weight of the noise suppression result image. In some embodiments, the noise suppression weight (1-cw) decreases as the edge enhancement weight w increases, and increases as the edge enhancement weight w decreases. In some embodiments, the coefficient c can be set according to empirical values, or it can be adjusted according to specific circumstances.

[0083] In some embodiments, the fused target image can be transmitted to a terminal for display, allowing doctors to view the internal tissues of the body and assisting in disease diagnosis. In the embodiments described in this specification, the diffusion tensor, image mask, and orientation map of the ultrasound image can be calculated simultaneously using a structure tensor. This allows for parallel processing of noise suppression and edge enhancement on the ultrasound image, accelerating computation and reducing repetitive calculations, effectively improving the efficiency of ultrasound image processing, and thus meeting the real-time and frame rate requirements of ultrasound image processing.

[0084] It should be noted that the above description of process 300 is for illustrative purposes only and does not limit the scope of this specification. Those skilled in the art can make various modifications and changes to process 300 under the guidance of this specification. However, these modifications and changes remain within the scope of this specification. For example, steps 340 and 350 can be performed simultaneously.

[0085] Figure 4 This is an exemplary flowchart of a speckle noise suppression method according to some embodiments of this specification.

[0086] Process 400 can be executed by a processing device (e.g., processing device 140). For example, process 400 can be implemented as a set of instructions (e.g., an application program) stored in a memory external to, but accessible by, the ultrasound image enhancement system 100, such as storage device 150. The processing device can execute the instruction set and, when executing the instructions, can be configured to execute process 400. The operational schematic diagram of process 400 presented below is illustrative. In some embodiments, the process can be accomplished using one or more additional operations not described and / or one or more operations not discussed. Additionally, Figure 4 The order of operations shown in and described below in process 400 is not intended to be restrictive.

[0087] Step 410: Obtain the structural tensor of the ultrasound image. In some embodiments, step 410 may be performed by a processing device or an acquisition module 210. For details of step 410, please refer to the relevant description of step 310, which will not be repeated here.

[0088] Step 420: Determine the diffusion tensor of the ultrasound image based on the structure tensor. In some embodiments, step 420 may be performed by a processing device or a determination module 220.

[0089] In some embodiments, the eigenvectors and eigenvalues ​​of the structure tensor can be determined first. The eigenvectors may include a first eigenvector w1 and a second eigenvector w2, and the eigenvalues ​​may include a first eigenvalue μ1 and a second eigenvalue μ2. In some embodiments, based on the first eigenvalue μ1 and the second eigenvalue μ2, a ​​third eigenvalue λ1 and a fourth eigenvalue λ2 can be determined. In some embodiments, based on the first eigenvector w1, the second eigenvector w1, and the determined third eigenvalue λ1 and fourth eigenvalue λ2, the diffusion tensor of the ultrasound image can be determined.

[0090] In some embodiments, the diffusion tensor can be determined by the following formula (6):

[0091]

[0092] Where D(I) represents the diffusion tensor of the input image I, w1 is the first eigenvector, w2 is the second eigenvector, λ1 is the third eigenvalue, λ2 is the fourth eigenvalue, and T represents the transpose. The first eigenvector w1 and the second eigenvector w2 can be determined by the structure tensor, and the third eigenvalue λ1 and the fourth eigenvalue λ2 can be determined by the first eigenvalue μ1 and the second eigenvalue μ2.

[0093] In some embodiments, the calculation of the diffusion tensor mainly depends on the determination of eigenvalues. The third eigenvalue λ1 and the fourth eigenvalue λ2 can be determined based on the first eigenvalue μ1 and the second eigenvalue μ2 of the structure tensor, modified according to different situations. Since anisotropic diffusion can suppress image noise while preserving image edge information, the eigenvalues ​​of the structure tensor can be modified as needed to enhance image edges during the diffusion process.

[0094] In some embodiments, in homogeneous tissue, the first feature value μ1 and the second feature value μ2 can be modified so that the values ​​of the third feature value λ1 and the fourth feature value λ2 are approximately equal. In some embodiments, in non-homogeneous tissue, the values ​​of the third feature value λ1 and the fourth feature value λ2 can be reduced. Specifically, in a homogeneous tissue region of an ultrasound image, if the values ​​of the third feature value λ1 and the fourth feature value λ2 are made approximately equal, the speed and form of diffusion in different directions in that region can be equal or approximately equal. In non-homogeneous tissue regions of an ultrasound image (e.g., edge regions), where edge information needs to be enhanced, the third feature value λ1 and the fourth feature value λ2 in this region can be made smaller to avoid smoothing.

[0095] In some embodiments, there may be a clear transition between homogeneous and non-homogeneous tissue in an ultrasound image. Enhancing contrast can facilitate the doctor's observation of the specific situation. In some embodiments, the difference between a first feature value μ1 and a second feature value μ2 can be used to determine whether the feature value corresponds to a homogeneous or non-homogeneous region. In the feature value matrix, each pixel in the ultrasound image corresponds to a first feature value μ1 and a second feature value μ2. The homogeneity or non-homogeneity of the feature value can be determined by comparing the difference between the first feature value μ1 and the second feature value μ2 corresponding to each pixel and determining whether the difference exceeds a threshold.

[0096] In some embodiments, the third eigenvalue λ1 and the fourth eigenvalue λ2 can be determined by the following formulas (7)-(9):

[0097] C = (μ1 - μ2) 2 (7)

[0098] λ1=α (8)

[0099]

[0100] Where α is a set constant with a small value, K is a threshold constant defined according to the actual situation, μ1 is the first eigenvalue of the structure tensor, μ2 is the second eigenvalue of the structure tensor, exp represents the exponential function, and C is an intermediate coefficient used to represent the relationship between the first eigenvalue μ1 and the second eigenvalue μ2. In some embodiments, α can be set to 0.01. As can be seen from the above formulas (7)-(9), when the first eigenvalue μ1 and the second eigenvalue μ2 are equal, the values ​​of the third eigenvalue λ1 and the fourth eigenvalue λ2 are both set constant values ​​α. When the first eigenvalue μ1 and the second eigenvalue μ2 are not equal, the third eigenvalue λ1 is a constant value α, and the fourth eigenvalue λ2 is calculated according to formula (9).

[0101] There are various methods to solve for the third eigenvalue λ1 and the fourth eigenvalue λ2 based on the first eigenvalue μ1 and the second eigenvalue μ2, and no limitation is made here. In some embodiments, the determined third eigenvalue λ1 and the fourth eigenvalue λ2 can be substituted into formula (6) to calculate the diffusion tensor of the ultrasound image, which can be used for subsequent speckle noise suppression processing of the ultrasound image.

[0102] Step 430: Apply the diffusion tensor to the ultrasound image to obtain a first image. In some embodiments, step 430 may be performed by a processing device or a speckle noise suppression module 230.

[0103] In some embodiments, the diffusion tensor can be applied as a direction-selective diffusion factor to the ultrasound image to obtain a first image. The first image is an image after speckle noise suppression processing. Speckle noise suppression processing of the ultrasound image can reduce noise (e.g., speckle noise) in the ultrasound image, thereby improving the quality of the ultrasound image. Further details of step 430 can be found in the relevant description of step 340, and will not be repeated here.

[0104] In some embodiments, the result image after speckle suppression processing (i.e., the first image) and the result image after edge enhancement processing (i.e., the second image) can be fused to obtain the final result image (i.e., the target image). Further details regarding the fusion of the first and second images can be found in the relevant description of step 360, and will not be repeated here.

[0105] It should be noted that the above description of process 400 is for illustrative purposes only and does not limit the scope of this specification. Those skilled in the art can make various modifications and changes to process 400 under the guidance of this specification. However, these modifications and changes remain within the scope of this specification.

[0106] Figure 5 This is an exemplary flowchart of an edge enhancement method according to some embodiments of this specification.

[0107] Process 500 can be executed by a processing device (e.g., processing device 140). For example, process 500 can be implemented as a set of instructions (e.g., an application program) stored in a memory external to and accessible by the ultrasound image enhancement system, such as storage device 150. The processing device can execute the instruction set and, when executing the instructions, can be configured to execute process 500. The operational schematic diagram of process 500 presented below is illustrative. In some embodiments, the process can be accomplished using one or more additional operations not described and / or one or more operations not discussed. Additionally, Figure 5 The order of operations shown in and described below in process 500 is not intended to be restrictive.

[0108] Step 510: Obtain the structural tensor of the ultrasound image. In some embodiments, step 510 may be performed by a processing device or an acquisition module 210. For details of step 510, please refer to the relevant description of step 320, which will not be repeated here.

[0109] Step 520: Determine the image mask of the ultrasound image based on the structure tensor. In some embodiments, step 520 may be performed by a processing device or a determination module 220.

[0110] In some embodiments, the image mask of the ultrasound image can be determined by determining the eigenvalues ​​of the structure tensor and based on those eigenvalues.

[0111] For each pixel in an ultrasound image, if the difference between the two feature values ​​(first feature value μ1 and second feature value μ2) of the pixel is large (e.g., greater than a certain threshold), it can be considered to be located in the image edge region to some extent. However, in a uniform region dominated by speckle noise, pixels have isotropic properties. Therefore, if the difference between its two feature values ​​is small, the pixel is considered to be speckle noise. In some embodiments, the image mask information can be obtained by setting a threshold or by automatically calculating the threshold. In some embodiments, if the difference between the first feature value μ1 and the second feature value μ2 (e.g., the difference between the absolute value of the first feature value μ1 and the absolute value of the second feature value μ2) is greater than the threshold, the pixel can be determined to belong to a non-uniform region, represented as 1; if the difference between the first feature value μ1 and the second feature value μ2 is less than or equal to the threshold, the pixel can be determined to belong to a uniform region, represented as 0. In some embodiments, the threshold can be determined by the average of the differences between the two feature values, that is, by taking the average of the differences between the two feature values ​​to determine the threshold. In some embodiments, the threshold can be set empirically or adjusted according to actual conditions.

[0112] In some embodiments, the image mask can be calculated based on the first eigenvalue μ1 and the second eigenvalue μ2 of the structure tensor according to the following formula (10):

[0113]

[0114] Where M(x,y) represents the edge information at pixel (x,y), μ1(x,y) represents the first feature value μ1 corresponding to pixel (x,y), μ2(x,y) represents the second feature value μ2 corresponding to pixel (x,y), n and m represent the length and width of the image, and i and j are variables of the summation function. The first feature value μ1 and the second feature value μ2 are the feature value data in the structure tensor. The threshold mentioned above can be determined by taking the average of the differences between the two feature values ​​corresponding to each pixel in the ultrasound image.

[0115] In some embodiments, the image mask of an ultrasound image is calculated using the eigenvalues ​​of the structure tensor, which, unlike calculations using methods such as standard deviation, can significantly reduce the computational load of image processing.

[0116] Step 530: Determine the orientation pattern of the ultrasound image based on the structure tensor. In some embodiments, step 530 may be performed by a processing device or a determination module 220.

[0117] In some embodiments, the orientation pattern can be determined by determining the eigenvectors of the structure tensor and based on those eigenvectors.

[0118] In some embodiments, the first feature vector w1 and the second feature vector w2 can represent the directions of maximum and minimum change in feature values, respectively. The arctangent can be used to obtain the orientation information of the pixel points, thereby determining the orientation pattern of the ultrasound image. Specifically, the orientation pattern can be obtained as shown in the following formula (11):

[0119]

[0120] Where O(x,y) represents the directional information at point (x,y), the direction of the pixel can be represented by an angle, and the angle range can be [0,π]. cor() represents the direction correction function, G(n,σ) represents the Gaussian filter kernel with size n and standard deviation σ, * represents convolution, w1 is the first eigenvector of the structure tensor, and w2 is the second eigenvector of the structure tensor.

[0121] In some embodiments, the obtained direction information needs to be corrected to obtain an accurate radiation pattern, and a direction correction function can be introduced. In some embodiments, the direction correction function... The range of angles can be determined using the arctangent formula, but the arctangent formula... The calculated angles are not all within the range of [0, π], and the calculated angles may even be negative. Therefore, direction correction is required to ensure that the final direction information is within the range of [0, π]. In some embodiments, the parameters in the direction correction function can be set based on empirical values ​​or adjusted according to specific circumstances.

[0122] In some embodiments, to reduce the interference of noisy pixels in the real ultrasound image, the obtained radiation pattern can be further filtered. In some embodiments, the radiation pattern can be divided into blocks, and a certain statistical feature of the directional data within a block can be used as the block direction. In some embodiments, the parameters in the Gaussian filter kernel G(n,σ) can be set according to empirical values ​​or adjusted according to specific circumstances.

[0123] In some embodiments, the direction correction function can divide the angles within the range of [0, π], and select one value as the value within each angle interval. In some embodiments, the number of Gaussian filter kernels is the same as the number of angle intervals. For each angle interval, a corresponding database is generated for the information therein, and the Gaussian filter kernels can perform filtering processing based on the database of the corresponding angle interval.

[0124] In some embodiments, a larger angle interval results in fewer angle intervals and fewer corresponding Gaussian filter kernels, leading to relatively higher computational efficiency. Conversely, a smaller angle interval results in more angle intervals and more corresponding Gaussian filter kernels, but with relatively higher computational accuracy. Therefore, the selection of the angle interval requires a balance between computational efficiency and accuracy. In some embodiments, the angle interval can be 0 degrees, 10 degrees, 20 degrees, 30 degrees, etc., and can be added and / or adjusted according to the direction correction function.

[0125] In some embodiments, the generated directional map can be verified and its accuracy compared by dividing it into blocks. Specifically, within each block, the direction of each point (i.e., pixel) is statistically analyzed to determine the main direction of each block, which is then compared with the direction in the test image. The more parallel the main direction of each block is to the direction of the test image, the higher its accuracy. The test image is an image containing directional information and can be used to test the accuracy of the generated directional map. In some embodiments, the test image can be randomly generated, only needing to contain information from different angles. As an example, the main direction of each block of the directional map can be represented by lines of one color (e.g., red), and the directional information in the test image can be represented by lines of another color (e.g., black). Then, the red lines in the directional map and the black lines in the test image are fitted. If the red lines and the black lines fit well, the directional map result is relatively accurate; otherwise, the directional map result is not accurate enough. In some embodiments, the test image can include information from as many angles as possible to verify the accuracy of the directional map corresponding to these angles.

[0126] In some embodiments, determining the orientation pattern based on feature vectors requires less computation than methods based on gradient calculation, thus resulting in higher computational efficiency without affecting the image edge enhancement results. For ultrasound images, where high frame rates are required (e.g., cardiac ultrasound images can reach 100 frames per second or higher), the efficiency improvement of calculating the orientation pattern based on feature vectors is significant.

[0127] Step 540: Based on the image mask and the orientation map, Gabor filtering is performed on the ultrasound image to obtain a second image. In some embodiments, step 540 may be performed by a processing device or an edge enhancement module 240.

[0128] In some embodiments, the ultrasound image can be Gabor filtered based on the image mask and the orientation map to obtain a second image. The second image is an image after edge enhancement processing.

[0129] In some embodiments, Gabor filtering is frequency and direction selective, thus exhibiting good robustness in texture enhancement. In some embodiments, the general form of the Gabor filter can be expressed as equation (3). Further description of Gabor filtering can be found in the relevant description in step 350, and will not be repeated here.

[0130] In some embodiments, the edge-enhanced result image, i.e., the second image, is obtained by performing Gabor filtering on the original ultrasound image based on the image mask and the orientation map. In some embodiments, the result image after speckle suppression processing (i.e., the first image) and the result image after edge enhancement processing (i.e., the second image) are fused to obtain the final result image (i.e., the target image). For more details on the fusion of the first image and the second image, please refer to the relevant description of step 360, which will not be repeated here.

[0131] It should be noted that the above description of process 500 is for illustrative purposes only and does not limit the scope of this specification. Those skilled in the art can make various modifications and changes to process 500 under the guidance of this specification. However, these modifications and changes are still within the scope of this specification. For example, steps 520 and 530 can be performed simultaneously.

[0132] Figure 6A and Figure 6B These are comparison images of exemplary original ultrasound images and exemplary enhanced ultrasound images shown in some embodiments of this specification.

[0133] The ultrasound image enhancement method provided in the embodiments of this specification can significantly improve image quality and enhance the detectability of low-contrast lesions. This improvement can be subjectively assessed visually.

[0134] like Figure 6A The image shown is the original ultrasound image without enhancement processing. Figure 6B The target image after enhancement processing according to the method provided in the embodiments of the present invention. Comparison Figure 6A and Figure 6B In the image above, within the solid bounding boxes 610 and 640, it's clear that edge enhancement processing improved the contrast of the edge regions. (Comparison) Figure 6A and Figure 6B In the image frames 620 and 650, as well as 630 and 660, it can be clearly seen that the speckle noise suppression process improves the image quality of uniform areas in the image, making them clearer and easier for doctors to diagnose.

[0135] The beneficial effects that the embodiments of this specification may bring include, but are not limited to: (1) By performing parallel computation of structural tensors, the image can be processed simultaneously with speckle noise suppression and edge enhancement, which can accelerate the computation and reduce repetitive computation, effectively improving the efficiency of ultrasound image processing; (2) By calculating the image mask through structural tensors, the region of interest for filtering can be obtained accurately, which can avoid unnecessary computation and improve computational efficiency and filtering efficiency; (3) By calculating the orientation map through structural tensors, the computational load of the orientation map can be reduced, and the corrected orientation map contains accurate pixel orientation information, which can significantly improve the edge enhancement effect; (4) By fusing the filtered edge enhancement processing result map and the speckle noise suppression processing result map through weighted summation, excessive noise can be effectively avoided.

[0136] It should be noted that different embodiments may produce different beneficial effects. In different embodiments, the beneficial effects may be any one or a combination of the above, or any other possible beneficial effects.

[0137] The basic concepts have been described above. Obviously, for those skilled in the art, the detailed disclosure above is merely illustrative and does not constitute a limitation of this specification. Although not explicitly stated herein, those skilled in the art may make various modifications, improvements, and corrections to this specification. Such modifications, improvements, and corrections are suggested in this specification and therefore remain within the spirit and scope of the exemplary embodiments described herein.

[0138] Furthermore, this specification uses specific terms to describe embodiments thereof. For example, "an embodiment," "one embodiment," and / or "some embodiments" refer to a particular feature, structure, or characteristic associated with at least one embodiment of this specification. Therefore, it should be emphasized and noted that references to "an embodiment," "one embodiment," or "an alternative embodiment" in different locations throughout this specification do not necessarily refer to the same embodiment. Moreover, certain features, structures, or characteristics in one or more embodiments of this specification can be appropriately combined.

[0139] Furthermore, unless expressly stated in the claims, the order of processing elements and sequences, the use of numbers and letters, or other names described in this specification are not intended to limit the order of the processes and methods described herein. Although various examples have been discussed in the foregoing disclosure of some embodiments of the invention that are currently considered useful, it should be understood that such details are for illustrative purposes only, and the appended claims are not limited to the disclosed embodiments; rather, the claims are intended to cover all modifications and equivalent combinations that conform to the spirit and scope of the embodiments described herein. For example, while the system components described above can be implemented using hardware devices, they can also be implemented solely using software solutions, such as installing the described system on existing servers or mobile devices.

[0140] Similarly, it should be noted that, in order to simplify the description disclosed herein and thus aid in the understanding of one or more embodiments of the invention, the foregoing description of embodiments in this specification may sometimes combine multiple features into a single embodiment, drawing, or description thereof. However, this method of disclosure does not imply that the subject matter of this specification requires more features than those mentioned in the claims. In fact, the embodiments contain fewer features than all the features of a single embodiment disclosed above.

[0141] In some embodiments, numbers describing the quantity of components and attributes are used. It should be understood that such numbers used in the description of embodiments are modified in some examples with the terms "approximately," "approximately," or "generally." Unless otherwise stated, "approximately," "approximately," or "generally" indicates that the numbers are allowed to vary by ±20%. Accordingly, in some embodiments, the numerical parameters used in the specification and claims are approximate values, which may be changed depending on the characteristics required by individual embodiments. In some embodiments, numerical parameters should take into account specified significant digits and employ a general method of digit reservation. Although the numerical ranges and parameters used to confirm their breadth of range in some embodiments of this specification are approximate values, in specific embodiments, such values ​​are set as precisely as feasible.

[0142] For each patent, patent application, patent application publication, and other material, such as articles, books, specifications, publications, and documents, referenced in this specification, the entire contents of which are incorporated herein by reference. This excludes historical application documents that are inconsistent with or conflict with the content of this specification, as well as documents that limit the broadest scope of the claims in this specification (currently or subsequently appended to this specification). It should be noted that in the event of any inconsistency or conflict between the descriptions, definitions, and / or terminology used in the supplementary materials to this specification and the content of this specification, the descriptions, definitions, and / or terminology used in this specification shall prevail.

[0143] Finally, it should be understood that the embodiments described in this specification are merely illustrative of the principles of the embodiments described herein. Other variations may also fall within the scope of this specification. Therefore, alternative configurations of the embodiments described herein are intended to be illustrative rather than limiting, and should be considered consistent with the teachings of this specification. Accordingly, the embodiments described herein are not limited to those explicitly introduced and described herein.

Claims

1. An ultrasound image enhancement method, characterized by, include: Acquire ultrasound images; Determine the structure tensor of the ultrasound image, wherein the structure tensor is used to describe the spatial information of the ultrasound image; Based on the structure tensor, the diffusion tensor, image mask, and orientation map of the ultrasound image are determined, wherein the image mask represents the edge information of one or more regions of interest in the ultrasound image, and the orientation map represents the orientation information of each pixel in the ultrasound image. Based on the diffusion tensor, the ultrasound image is subjected to speckle noise suppression processing to obtain a first image; Based on the image mask and the orientation pattern, edge enhancement processing is performed on the ultrasound image to obtain a second image; By fusing the first image and the second image to obtain the target image, wherein determining the orientation pattern of the ultrasound image based on the structure tensor includes: Determine the eigenvectors of the structure tensor, wherein the eigenvectors include a first eigenvector and a second eigenvector; Based on the first feature vector and the second feature vector, the orientation information of the pixel is determined using the arctangent rule; and The direction information is corrected based on the direction correction function to determine the direction pattern.

2. The method as described in claim 1, characterized in that, Determining the diffusion tensor of the ultrasound image based on the structure tensor includes: Determine the eigenvectors and eigenvalues ​​of the structural tensor, wherein the eigenvectors include a first eigenvector and a second eigenvector, and the eigenvalues ​​include a first eigenvalue and a second eigenvalue; Based on the first and second feature values, a third and fourth feature value are determined, wherein, in a homogeneous tissue region of the ultrasound image, the first and second feature values ​​are modified so that the values ​​of the third and fourth feature values ​​are equal, or in a non-homogeneous tissue region of the ultrasound image, the values ​​of the third and fourth feature values ​​are reduced. The diffusion tensor is determined based on the first eigenvector, the second eigenvector, the third eigenvalue, and the fourth eigenvalue.

3. The method as described in claim 1, characterized in that, The process of determining the image mask of the ultrasound image based on the structure tensor includes: Determine the eigenvalues ​​of the structure tensor; The image mask is determined based on the eigenvalues ​​of the structure tensor.

4. The method as described in claim 3, characterized in that, Determining the image mask based on the feature values ​​of the structure tensor includes: For each pixel of the ultrasound image Determine the difference between the first feature value and the second feature value corresponding to the pixel; In response to a situation where the difference between the first and second feature values ​​corresponding to the pixel is greater than a threshold, the pixel is determined to belong to a non-uniform organization region, and the pixel is assigned a value of 1; and In response to the fact that the difference between the first feature value and the second feature value corresponding to the pixel is less than or equal to the threshold, it is determined that the pixel belongs to a uniformly organized region, and the pixel is assigned a value of 0.

5. The method as described in claim 1, characterized in that, The step of performing speckle noise suppression processing on the ultrasound image based on the diffusion tensor to obtain the first image includes: The diffusion tensor is applied to the ultrasound image to obtain the first image.

6. The method as described in claim 1, characterized in that, The step of performing edge enhancement processing on the ultrasound image based on the image mask and the orientation map to obtain a second image includes: Based on the image mask and the orientation pattern, the ultrasound image is subjected to Gabor filtering to obtain the second image.

7. The method as described in claim 1, characterized in that, The noise suppression process and the edge enhancement process are performed in parallel.

8. The method as described in claim 1, characterized in that, The process of fusing the first image and the second image to obtain the target image includes: The second image is then filtered; The target image is obtained by fusing the first image and the filtered second image.

9. An ultrasound image enhancement system, characterized in that, include: The acquisition module is used to acquire ultrasound images; A determination module is used to determine the structure tensor of the ultrasound image, and to determine the diffusion tensor, image mask, and orientation map of the ultrasound image based on the structure tensor. The structure tensor is used to describe the spatial information of the ultrasound image, the image mask represents the edge information of one or more regions of interest in the ultrasound image, and the orientation map represents the orientation information of each pixel in the ultrasound image. The speckle noise suppression module is used to perform speckle noise suppression processing on the ultrasound image based on the diffusion tensor to obtain a first image; An edge enhancement module is used to perform edge enhancement processing on the ultrasound image based on the image mask and the orientation pattern to obtain a second image; An image fusion module is used to fuse the first image and the second image to obtain a target image, wherein the determining module is further used to: Determine the eigenvectors of the structure tensor, wherein the eigenvectors include a first eigenvector and a second eigenvector; Based on the first feature vector and the second feature vector, the orientation information of the pixel is determined using the arctangent rule; and The direction information is corrected based on the direction correction function to determine the direction pattern.

10. A computer-readable storage medium, characterized in that, The storage medium stores computer instructions, which, when executed by a processor, implement the ultrasound image enhancement method as described in any one of claims 1 to 8.

Images