Image processing method and device, computer device and storage medium

Abstract

The application relates to an image processing method and device, computer equipment and a storage medium. The method comprises the following steps: acquiring an initial image; performing speckle noise suppression processing and edge enhancement processing on the initial image respectively to obtain a denoising image and an enhanced image corresponding to the initial image; performing fusion processing on the denoising image and the enhanced image to obtain a first fusion image; performing fusion processing on the first fusion image and the initial image to obtain a second fusion image; taking the second fusion image as a new initial image, and returning to the steps of performing speckle noise suppression processing and edge enhancement processing on the initial image until a preset iteration end condition is reached to obtain a target image corresponding to the initial image. The method can suppress three-dimensional speckle noise and enhance three-dimensional edges while further preserving the original information of a three-dimensional image.

Description

Technical Field

[0001] This application relates to the field of ultrasound imaging technology, and in particular to an image processing method, apparatus, computer equipment, storage medium, and computer program product. Background Technology

[0002] Ultrasonic imaging is a method of obtaining visible images of objects using ultrasound waves. The ultrasound system receives the echo signals from the scanned structure and displays the intensity of the reflected signals as brightness. Due to the interference effect of the echoes and the mutual interference between the scattered ultrasound beams, when two echoes from related sources overlap, particles and textures of varying brightness are produced in the ultrasound image; this is the origin of speckle noise. The presence of speckle noise significantly affects image quality; therefore, suppressing speckle noise is a crucial step in ultrasound image processing.

[0003] Current methods for suppressing speckle noise in ultrasound images mostly employ filtering techniques, such as mean filtering, median filtering, and local filtering. These methods replace noisy pixels with statistical values ​​(such as mean or median) of pixels within a window. However, these methods do not differentiate between different regions of the image, and while filtering out noise, they also replace pixel information of non-noise pixels, resulting in the loss of some image details.

[0004] Therefore, this traditional three-dimensional ultrasound image enhancement method is prone to causing excessive image blurring while removing image speckle noise, and cannot simultaneously satisfy speckle noise suppression and image clarity. Summary of the Invention

[0005] Therefore, it is necessary to address the technical problem that the aforementioned ultrasound image enhancement methods, while removing image speckle noise, easily cause excessive image blurring, failing to simultaneously satisfy speckle noise suppression and image clarity. This necessitates providing an image processing method, apparatus, computer equipment, computer-readable storage medium, and computer program product.

[0006] Firstly, this application provides an image processing method. The method includes:

[0007] Get the initial image;

[0008] The initial image is subjected to noise suppression processing and edge enhancement processing respectively to obtain the denoised image and the enhanced image corresponding to the initial image;

[0009] The denoised image and the enhanced image are fused together to obtain a first fused image;

[0010] The first fused image and the initial image are fused to obtain a second fused image;

[0011] The second fused image is used as the new initial image, and the steps of performing speckle noise suppression and edge enhancement processing on the initial image are returned until the preset iteration end condition is reached to obtain the target image corresponding to the initial image.

[0012] In one embodiment, the step of fusing the denoised image and the enhanced image to obtain a first fused image includes:

[0013] The edge enhancement results of the enhanced image are filtered using a preset filtering method to obtain the filtered enhanced image.

[0014] The denoised image and the filtered enhanced image are fused together to obtain the first fused image.

[0015] In one embodiment, the step of fusing the denoised image and the filtered enhanced image to obtain the first fused image includes:

[0016] Obtain the first weight of the filtered enhanced image;

[0017] Based on the first weight of the enhanced image after filtering, a second weight of the denoised image is determined; wherein the second weight of the denoised image is negatively correlated with the first weight;

[0018] The denoised image and the filtered enhanced image are fused according to the first weight and the second weight to obtain the first fused image.

[0019] In one embodiment, the step of fusing the first fused image and the initial image to obtain a second fused image includes:

[0020] Obtain the third weight of the first fused image and the fourth weight of the initial image; the sum of the third weight and the fourth weight is 1;

[0021] The first fused image and the initial image are fused according to the third weight and the fourth weight to obtain the second fused image.

[0022] In one embodiment, before performing speckle noise suppression processing and edge enhancement processing on the initial image, the method further includes:

[0023] The initial image is filtered to obtain the filtered image corresponding to the initial image;

[0024] The process of performing speckle noise suppression and edge enhancement processing on the initial image to obtain a denoised image and an enhanced image corresponding to the initial image includes:

[0025] The filtered image corresponding to the initial image is subjected to speckle noise suppression processing and edge enhancement processing respectively to obtain the denoised image and enhanced image of the filtered image, which are used as the denoised image and enhanced image corresponding to the initial image.

[0026] In one embodiment, the iteration termination condition is determined based on image processing rate requirements and image quality requirements;

[0027] The iteration termination condition is either a preset number of iterations or the grayscale value change of the second fused image obtained from two adjacent iterations is less than a threshold.

[0028] In one embodiment, the denoised image corresponding to the initial image is determined in the following manner:

[0029] Obtain the diffusion tensor and image gradient of the initial image;

[0030] Based on the diffusion tensor and the image gradient, the anisotropic diffusion information of the initial image is obtained; the anisotropic diffusion information represents the rate of change of the image grayscale value of the initial image at a certain moment;

[0031] Based on the anisotropic diffusion information, the initial image is subjected to speckle noise suppression processing to obtain the denoised image corresponding to the initial image.

[0032] In one embodiment, obtaining the diffusion tensor of the initial image includes:

[0033] Obtain the initial eigenvalues ​​and eigenvectors of the structure tensor of the initial image; the initial eigenvalues ​​represent the diffusion coefficients that diffuse along the direction of the eigenvectors.

[0034] The initial eigenvalues ​​of the structure tensor are corrected to obtain the corrected eigenvalues.

[0035] Based on the corrected eigenvalues ​​and the eigenvectors, the diffusion tensor of the initial image is obtained.

[0036] In one embodiment, the feature vector is a three-dimensional feature vector, and the initial feature values ​​include three feature values ​​corresponding to the three-dimensional feature vector; the step of correcting the initial feature values ​​of the structure tensor to obtain corrected feature values ​​includes:

[0037] In a uniformly organized region in the initial image, the three feature values ​​included in the corrected feature values ​​are approximately equal; in an edge region in the initial image, the largest feature value among the three feature values ​​included in the corrected feature values ​​will stop spreading as the correction target, and the initial feature values ​​are corrected to obtain the corrected feature values.

[0038] In one embodiment, the enhanced image corresponding to the initial image is determined in the following manner:

[0039] Obtain the corrected feature values ​​of the structure tensor of the initial image;

[0040] Based on the corrected feature values, edge extraction processing is performed on the initial image to obtain the image edge region of the initial image;

[0041] Edge enhancement processing is performed on the edge regions of the image to obtain the enhanced image corresponding to the initial image.

[0042] In one embodiment, the step of performing edge extraction processing on the initial image based on the corrected feature values ​​to obtain the image edge region of the initial image includes:

[0043] For each pixel in the initial image, two target feature values ​​are determined from the three corrected feature values ​​corresponding to the pixel; the two target feature values ​​include the feature value with the largest value and the feature value with the smallest value among the three corrected feature values.

[0044] If the difference between the two target feature values ​​is greater than a threshold, then the pixel is determined to be at the image edge of the initial image;

[0045] Based on the extracted pixels located at the image edges, the image edge region of the initial image is obtained.

[0046] Secondly, this application also provides an image processing apparatus. The apparatus includes:

[0047] The acquisition unit is used to acquire the initial image;

[0048] The processing unit is used to perform noise suppression processing and edge enhancement processing on the initial image respectively to obtain a denoised image and an enhanced image corresponding to the initial image;

[0049] The first fusion unit is used to fuse the denoised image and the enhanced image to obtain a first fused image;

[0050] The second fusion unit is used to fuse the first fused image and the initial image to obtain the second fused image;

[0051] The determining unit is used to take the second fused image as a new initial image and return to the steps of performing speckle noise suppression processing and edge enhancement processing on the initial image respectively, until a preset iteration end condition is reached to obtain the target image corresponding to the initial image.

[0052] Thirdly, this application also provides a computer device. The computer device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to perform the following steps:

[0053] Get the initial image;

[0054] The initial image is subjected to noise suppression processing and edge enhancement processing respectively to obtain the denoised image and the enhanced image corresponding to the initial image;

[0055] The denoised image and the enhanced image are fused together to obtain a first fused image;

[0056] The first fused image and the initial image are fused to obtain a second fused image;

[0057] The second fused image is used as the new initial image, and the steps of performing speckle noise suppression and edge enhancement processing on the initial image are returned until the preset iteration end condition is reached to obtain the target image corresponding to the initial image.

[0058] Fourthly, this application also provides a computer-readable storage medium. The computer-readable storage medium stores a computer program thereon, which, when executed by a processor, performs the following steps:

[0059] Get the initial image;

[0060] The initial image is subjected to noise suppression processing and edge enhancement processing respectively to obtain the denoised image and the enhanced image corresponding to the initial image;

[0061] The denoised image and the enhanced image are fused together to obtain a first fused image;

[0062] The first fused image and the initial image are fused to obtain a second fused image;

[0063] The second fused image is used as the new initial image, and the steps of performing speckle noise suppression and edge enhancement processing on the initial image are returned until the preset iteration end condition is reached to obtain the target image corresponding to the initial image.

[0064] Fifthly, this application also provides a computer program product. The computer program product includes a computer program that, when executed by a processor, performs the following steps:

[0065] Get the initial image;

[0066] The initial image is subjected to noise suppression processing and edge enhancement processing respectively to obtain the denoised image and the enhanced image corresponding to the initial image;

[0067] The denoised image and the enhanced image are fused together to obtain a first fused image;

[0068] The first fused image and the initial image are fused to obtain a second fused image;

[0069] The second fused image is used as the new initial image, and the steps of performing speckle noise suppression and edge enhancement processing on the initial image are returned until the preset iteration end condition is reached to obtain the target image corresponding to the initial image.

[0070] The aforementioned image processing method, apparatus, computer equipment, storage medium, and computer program product, after acquiring an initial image, iteratively executes the steps of performing speckle noise suppression processing and edge enhancement processing on the initial image to obtain a denoised image and an enhanced image corresponding to the initial image; fusing the denoised image and the enhanced image to obtain a first fused image; and fusing the first fused image and the initial image to obtain a second fused image, until a preset iteration termination condition is reached to obtain the target image corresponding to the initial image. This method, by fusing the denoised image after speckle noise suppression processing and the enhanced image after edge enhancement processing, allows for the simultaneous satisfaction of image edge detail requirements while performing speckle noise suppression. By fusing the first fused image and the initial image, image detail information not processed by edge enhancement can be further compensated. Through multiple iterations, while performing three-dimensional speckle noise suppression and three-dimensional edge enhancement on the image, the original information of the three-dimensional image can be further preserved, simultaneously satisfying the requirements for image speckle noise suppression and image clarity. Attached Figure Description

[0071] Figure 1 This is a flowchart illustrating an image processing method in one embodiment;

[0072] Figure 2 This is a schematic diagram of the complete process of a three-dimensional ultrasound image enhancement method in one embodiment;

[0073] Figure 3 This is a schematic diagram of a three-dimensional ultrasound image enhancement system in one embodiment;

[0074] Figure 4 This is a comparison diagram of the original three-dimensional ultrasound image and the image after enhancement processing using the enhancement method of this application in one embodiment;

[0075] Figure 5 This is a structural block diagram of an image processing device in one embodiment;

[0076] Figure 6 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0077] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0078] It should be noted that the terms "first," "second," etc., used in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. It should also be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties.

[0079] In one embodiment, such as Figure 1 As shown, an image processing method is provided. This embodiment illustrates the application of this method to a terminal. It is understood that this method can also be applied to a server, and to a system including both a terminal and a server, and is implemented through interaction between the terminal and the server. The terminal can be, but is not limited to, various personal computers, laptops, smartphones, tablets, IoT devices, and portable wearable devices. IoT devices can include smart speakers, smart TVs, smart air conditioners, smart in-vehicle devices, etc. Portable wearable devices can include smartwatches, smart bracelets, head-mounted devices, etc. The server can be implemented using a standalone server or a server cluster consisting of multiple servers. In this embodiment, the method includes the following steps:

[0080] Step S110: Obtain the initial image.

[0081] The initial image can be a three-dimensional ultrasound image.

[0082] Understandably, since the acquired 3D ultrasound images contain speckle noise, post-processing is required after obtaining the initial 3D ultrasound images to improve image quality.

[0083] Step S120: Perform noise suppression processing and edge enhancement processing on the initial image to obtain the denoised image and enhanced image corresponding to the initial image.

[0084] In the specific implementation, speckle noise suppression processing is performed on the initial image, including calculating the structure tensor of the initial image, determining the initial eigenvalues ​​and eigenvectors of the structure tensor, and determining the diffusion tensor of the initial image based on the initial eigenvalues ​​and eigenvectors of the structure tensor. After obtaining the image gradient of the initial image, the anisotropic diffusion information of the initial image is obtained based on the image gradient and the diffusion tensor. Finally, speckle noise suppression processing is performed on the initial image based on the anisotropic diffusion information of the initial image to obtain the denoised image corresponding to the initial image. More specifically, determining the diffusion tensor of the initial image based on the initial eigenvalues ​​and eigenvectors of the structure tensor also includes correcting the initial eigenvalues ​​of the structure tensor to obtain corrected eigenvalues, and obtaining the diffusion tensor of the initial image based on the corrected eigenvalues ​​and eigenvectors.

[0085] Edge enhancement processing of the initial image can be performed on the entire initial image, or the edge regions of the initial image can be extracted and the edge enhancement processing can be performed only on the edge regions to reduce the scope of edge enhancement processing and improve image processing efficiency.

[0086] Step S130: The denoised image and the enhanced image are fused to obtain the first fused image.

[0087] In a specific implementation, the fusion of the denoised image and the enhanced image can be weighted fusion. That is, a first weight of the denoised image and a second weight of the enhanced image are determined. The pixel values ​​of pixels at the same position in the denoised image and the enhanced image are weighted and summed using the first weight and the second weight to obtain the first weighted pixel value of each pixel. Based on the first weighted pixel values ​​of each pixel, the first fused image is obtained.

[0088] In this step, by fusing the denoised image and the enhanced image of the initial image, it is possible to remove image speckle noise while preserving the edge details of the image.

[0089] Step S140: Perform fusion processing on the first fused image and the initial image to obtain the second fused image.

[0090] In a specific implementation, the fusion of the first fused image and the initial image can also be a weighted fusion. That is, the third weight of the first fused image and the fourth weight of the initial image are determined. The pixel values ​​of the pixels at the same position in the first fused image and the initial image are weighted and summed using the third weight and the fourth weight to obtain the second weighted pixel value of each pixel. Based on the second weighted pixel values ​​of each pixel, the second fused image is obtained.

[0091] Step S150: Use the second fused image as the new initial image, and return to the steps of performing noise suppression processing and edge enhancement processing on the initial image respectively, until the preset iteration end condition is reached, and obtain the target image corresponding to the initial image.

[0092] The iteration termination condition can be determined based on the image processing rate requirement and the image quality requirement, so as to ensure that the image quality is improved as much as possible while meeting the image rate requirement.

[0093] For example, the iteration termination condition can be a preset number of iterations. Specifically, when a high frame rate (the number of frames or images displayed per second) is required, a smaller number of iterations (e.g., 1 time) can be set; when a lower frame rate is required, a larger number of iterations (e.g., 3-5 times) can be set. The iteration termination condition can also be that the change in grayscale value of the second fused image obtained from two adjacent iterations is less than a threshold.

[0094] In specific implementation, after obtaining the second fused image, the second fused image can be used as the new initial image, and the process of steps S120 to S140 is repeated until the preset number of iterations is reached. The second fused image obtained in the last iteration is then determined as the target image corresponding to the initial image.

[0095] In the aforementioned image processing method, after acquiring the initial image, the method iteratively performs speckle noise suppression and edge enhancement processing on the initial image to obtain a denoised image and an enhanced image corresponding to the initial image. The denoised image and the enhanced image are then fused to obtain a first fused image. The first fused image and the initial image are then fused to obtain a second fused image. This process continues until a preset iteration termination condition is met, resulting in the target image corresponding to the initial image. This method, by fusing the denoised image after speckle noise suppression and the enhanced image after edge enhancement, allows for the simultaneous satisfaction of image edge detail requirements while performing speckle noise suppression. By fusing the first fused image and the initial image, image detail information not processed by edge enhancement can be further compensated. Through multiple iterations, while performing three-dimensional speckle noise suppression and three-dimensional edge enhancement, the original information of the three-dimensional image can be further preserved, simultaneously satisfying the requirements for image speckle noise suppression and image clarity.

[0096] In an exemplary embodiment, fusing a denoised image and an enhanced image to obtain a first fused image includes: filtering the edge enhancement results of the enhanced image using a preset filtering method to obtain a filtered enhanced image; and fusing the denoised image and the filtered enhanced image to obtain the first fused image.

[0097] In practice, since edge enhancement usually only needs to enhance the strong texture region, in order to avoid introducing too much noise into the edge enhancement process, it is necessary to further filter the edge enhancement results of the enhanced image. Specifically, the edge enhancement results can be filtered using the findContours function. There are various filtering methods, such as morphological operations, filtering based on features such as contour area and perimeter, and using machine learning to classify contours. The filtered enhanced image and the denoised image are then fused to obtain the first fused image.

[0098] Further, in an exemplary embodiment, fusing the denoised image and the filtered enhanced image to obtain a first fused image includes: obtaining a first weight of the filtered enhanced image; determining a second weight of the denoised image based on the first weight of the filtered enhanced image; wherein the second weight of the denoised image is negatively correlated with the first weight; and fusing the denoised image and the filtered enhanced image according to the first weight and the second weight to obtain the first fused image.

[0099] Specifically, let the enhanced image obtained after edge enhancement processing be E, and the filtered enhanced image can be represented as findContours(E). Assume that the first weight of the filtered enhanced image is w, the denoised image is R, and the second weight of the denoised image is negatively correlated with the first weight. Specifically, the second weight can be represented as 1 - d1w. Then, according to the first and second weights, the denoised image and the filtered enhanced image are weighted and summed to obtain the first fused image, which can be expressed by the following formula:

[0100] (1-d1w)R+w﹒ findContours(E)

[0101] Where d1 is a set coefficient, and d1 and w together determine the weight of the denoised image R after speckle suppression. The weight of speckle suppression needs to decrease as the weight of edge enhancement increases.

[0102] In this embodiment, by filtering the edge enhancement results, the image processing efficiency can be improved. By fusing the denoised image and the enhanced image of the initial image, the edge detail information of the image can be preserved while removing the image's speckle noise.

[0103] In an exemplary embodiment, fusing a first fused image and an initial image to obtain a second fused image includes: obtaining a third weight of the first fused image and a fourth weight of the initial image; and fusing the first fused image and the initial image according to the third weight and the fourth weight to obtain the second fused image.

[0104] In the specific implementation, since the initial image needs to be filtered to obtain a filtered image after the initial image is obtained, and post-processing is performed based on the filtered image, the fusion processing of the first fused image and the initial image is actually the fusion processing of the first fused image and the filtered image. Specifically, the fusion can be performed according to the third weight of the first fused image and the fourth weight of the filtered image.

[0105] Let the third weight of the first fused image be d2, the fourth weight of the initial image (filtered image) be 1-d2, and the filtered image be S. Then the fusion process of the second fused image H can be represented as:

[0106] H=(1-d2)S+d2[(1-d1w)R+w﹒ findContours(E)]

[0107] In this embodiment, by fusing the first fused image obtained by fusing the denoised image and the enhanced image with the initial image again, it is possible to further preserve the original information of the three-dimensional image while performing three-dimensional noise suppression and three-dimensional edge enhancement on the image.

[0108] In an exemplary embodiment, before performing speckle noise suppression processing and edge enhancement processing on the initial image in step S120, the method further includes: performing filtering processing on the initial image to obtain a filtered image corresponding to the initial image;

[0109] Step S120 further includes: performing noise suppression processing and edge enhancement processing on the filtered image corresponding to the initial image, respectively, to obtain a denoised image and an enhanced image of the filtered image, which are used as the denoised image and enhanced image corresponding to the initial image.

[0110] In this embodiment, after acquiring the initial image, the initial image can first be filtered, for example, by three-dimensional Gaussian filtering, to obtain the filtered image corresponding to the initial image. Then, the filtered image is subjected to speckle noise suppression and edge enhancement processing respectively to obtain the denoised image and enhanced image of the filtered image, which are used as the denoised image and enhanced image corresponding to the initial image. This method first filters the initial image to achieve preliminary suppression of image noise while preserving image detail features as much as possible, so as to improve the effectiveness and reliability of subsequent image processing and analysis.

[0111] In an exemplary embodiment, determining the denoised image corresponding to the initial image may include: obtaining the diffusion tensor and image gradient of the initial image; obtaining anisotropic diffusion information of the initial image based on the diffusion tensor and image gradient; the anisotropic diffusion information represents the rate of change of the image grayscale value of the initial image at a certain moment; and performing speckle suppression processing on the initial image based on the anisotropic diffusion information to obtain the denoised image corresponding to the initial image.

[0112] The diffusion tensor, used to control the diffusion of the image, can be represented by a 3D matrix as follows:

[0113]

[0114] Image gradients calculate the rate of change in an image. Edges of an image experience larger changes in grayscale values, resulting in larger gradient values; conversely, smoother areas of the image exhibit smaller changes in grayscale values, leading to smaller gradient values. If an image is viewed as a two-dimensional discrete function, the image gradient can be understood as the derivative of this function. Image gradients are typically approximated (approximate derivative values) by calculating the differences between pixel values.

[0115] The three-dimensional anisotropic diffusion equation can be expressed as:

[0116] in, The image gradient is represented by , D represents the diffusion tensor, and div represents the divergence function. Divergence is a vector operator in vector analysis that maps a vector field in vector space to a scalar field. Divergence describes whether a point in the vector field is a convergence point or a source point. The anisotropic diffusion equation can be understood as the rate of change of the image grayscale values ​​at time t.

[0117] In practice, after obtaining the diffusion tensor and image gradient of the initial image, the anisotropic diffusion information of the initial image can be obtained by substituting the diffusion tensor and image gradient into the three-dimensional anisotropic diffusion equation. This can be expressed by the following relation:

[0118]

[0119] Among them, in the above formula It can represent anisotropic diffusion information, div represents the divergence function, and I represents the initial image. These represent the diffusion information of the initial image in the x, y, and z directions, respectively.

[0120] Furthermore, by applying speckle noise suppression processing to the initial image based on anisotropic diffusion information, the product of anisotropic diffusion information and time can be considered as an increment, which is added to the initial image to obtain the denoised image after speckle noise suppression. This can be expressed by the formula:

[0121]

[0122] Among them, I * I represents the denoised image, and I represents the original image. This represents the increment corresponding to anisotropic diffusion information.

[0123] In this embodiment, anisotropic diffusion information is first obtained through the diffusion tensor and image gradient of the initial image. Then, based on the anisotropic diffusion information, the initial image is subjected to speckle noise suppression processing, which can enhance the image edges to a certain extent while preserving image details.

[0124] Further, in an exemplary embodiment, obtaining the diffusion tensor of the initial image includes: step S2101, obtaining the initial eigenvalues ​​and eigenvectors of the structure tensor of the initial image; the initial eigenvalues ​​represent the diffusion coefficients that diffuse along the direction of the eigenvectors; step S2102, correcting the initial eigenvalues ​​of the structure tensor to obtain corrected eigenvalues; step S2103, obtaining the diffusion tensor of the initial image based on the corrected eigenvalues ​​and eigenvectors.

[0125] The structure tensor is primarily used to distinguish between flat regions, edge regions, and corner regions in an image. The structure tensor of an image can be viewed as a structure matrix of the image.

[0126] The eigenvalues ​​of the structure tensor can also be understood as the degree of diffusion in the direction of the eigenvector.

[0127] In practice, the initial eigenvalues ​​μ1, μ2, μ3 and eigenvectors v1, v2, v3 of the structure tensor can be obtained by performing feature decomposition on the structure tensor of the initial image:

[0128]

[0129]

[0130] in, Represents the structure tensor. I represents the image gradient of the image. x I y I z These represent the image gradients of the initial image in the x, y, and z directions, respectively.

[0131] Furthermore, the initial eigenvalues ​​μ1, μ2, and μ3 of the structure tensor are corrected to obtain corrected eigenvalues ​​λ1, λ2, and λ3. Based on these corrected eigenvalues ​​λ1, λ2, and λ3 and the eigenvectors v1, v2, and v3, the diffusion tensor D is calculated, expressed by the formula:

[0132]

[0133] Furthermore, the feature vector is a three-dimensional feature vector, and the corrected feature values ​​include three feature values ​​corresponding to the three-dimensional feature vector; in the uniformly organized regions of the initial image, the three feature values ​​included in the corrected feature values ​​are approximately equal; in the edge regions of the initial image, the largest feature value among the three feature values ​​included in the corrected feature values ​​will stop spreading.

[0134] Whether the largest eigenvalue stops diffusion can be determined by whether the largest eigenvalue approaches zero. The smaller the largest eigenvalue, the closer it is to zero, indicating that it is closer to the goal of stopping diffusion.

[0135] Specifically, the relational expression for correcting the initial eigenvalues ​​can be expressed as:

[0136] λ1=c1

[0137] λ Z =c1

[0138]

[0139] Where C = (μ1 - μ3) 2 C represents the squared difference between the maximum and minimum values ​​of the three initial eigenvalues ​​μ1, μ2, and μ3. α is a constant that determines the diffusion amount and can be equal to λ1, λ2, and λ3. α is usually set to 1. β is a negative number that determines the edge enhancement amount; the absolute value of β determines the edge enhancement amount. The larger the absolute value, the more blurred the edge. Its value range is [-1, -0.1]. K and c1 are constants that are set. The larger the value of K, the more blurred the tissue. The value of c1 ranges from [0, 0.5].

[0140] Understandably, in uniformly organized regions, the value of C will be very small, and the diffusion coefficients along the three eigenvector directions will be approximately equal, meaning the eigenvalues ​​will be approximately equal. In edge regions, the diffusion coefficient will stop spreading and may even enhance the edges, because the corrected eigenvalue λ3 will be negative. Since the diffusion coefficient determines the amount of diffusion, a negative diffusion amount in edge regions can enhance the edges to some extent.

[0141] In the above embodiments, after obtaining the initial eigenvalues ​​and eigenvectors of the structure tensor, the initial eigenvalues ​​are corrected by taking advantage of the different characteristics of the uniformly organized region and the edge region. This makes the corrected eigenvalues ​​more consistent with the characteristics of the image, thereby ensuring the accuracy of the anisotropic diffusion information determined subsequently based on the corrected eigenvalues.

[0142] In an exemplary embodiment, the enhanced image corresponding to the initial image is determined by: obtaining the corrected feature values ​​of the structure tensor of the initial image; performing edge extraction processing on the initial image based on the corrected feature values ​​to obtain the image edge region of the initial image; and performing edge enhancement processing on the image edge region to obtain the enhanced image corresponding to the initial image.

[0143] In specific implementation, edge extraction of the initial image can be achieved by modifying the feature values ​​of the structure tensor of the initial image. Each pixel corresponds to three feature values ​​along three feature vector directions. The difference between the largest and smallest feature values ​​among the three modified feature values ​​can be calculated and compared with a preset threshold. Based on the comparison result, pixels in the image edge region are extracted to obtain the image edge region of the initial image. Further edge enhancement processing is performed on the extracted image edge region to obtain the enhanced image corresponding to the initial image.

[0144] Further, in an exemplary embodiment, edge extraction processing is performed on the initial image based on the corrected feature values ​​to obtain the image edge region of the initial image, including: for each pixel in the initial image, determining two target feature values ​​of the pixel from the three corrected feature values ​​corresponding to the pixel; the two target feature values ​​include the feature value with the largest value and the feature value with the smallest value among the three corrected feature values; if the difference between the two target feature values ​​is greater than a threshold, then the pixel is determined to be at the image edge of the initial image; and the image edge region of the initial image is obtained based on each pixel extracted at the image edge.

[0145] Specifically, pixels in a uniform region dominated by speckle noise exhibit isotropic properties. Therefore, for each pixel in an image, if the difference between the two target feature values ​​of that pixel is small, the pixel is considered to be speckle noise; if the difference between the two target feature values ​​of that pixel is large, the pixel can be considered to be located in an image edge region. Therefore, the edge region of the image can be obtained by setting a threshold or by automatically calculating a threshold. One implementation method can be expressed as follows:

[0146]

[0147] Where M(x,y,z) represents the edge information at point (x,y,z), μ1(x,y,z) and μ2(x,y,z) represent the two target feature values ​​corresponding to point (x,y,z), and l, m, and n represent the length, width, and height of the initial image, respectively. This represents the calculated threshold.

[0148] The above embodiments improve the accuracy of the extracted image edge regions by using the modified feature values ​​of the structure tensor for edge extraction.

[0149] In one embodiment, to facilitate understanding of the embodiments of this application by those skilled in the art, specific examples will be described below in conjunction with the accompanying drawings.

[0150] refer to Figure 2 This paper illustrates a complete flowchart of a three-dimensional ultrasound image enhancement method. After acquiring a three-dimensional ultrasound image, it performs three-dimensional filtering. The filtered ultrasound image is then used as the input image for iterative enhancement. Specifically, the enhancement operation includes: first, performing three-dimensional speckle noise suppression and three-dimensional edge enhancement on the input image; then, fusing the image after speckle noise suppression and the image after edge enhancement to obtain a first fused image; further, fusing the first fused image with the filtered image again to obtain a second fused image. The relationship between the number of iterations and a preset threshold N is determined. If the number of iterations is less than the threshold N, the second fused image is used as the new input image for the enhancement operation, and this process continues until the number of iterations exceeds the threshold N. The second fused image obtained in the last iteration is then output as the final image. This method enhances the three-dimensional ultrasound image, resulting in a three-dimensional ultrasound image with better image quality and stereoscopic effect.

[0151] refer to Figure 3 A schematic diagram of a three-dimensional ultrasound image enhancement system is shown, including: an image acquisition module, a three-dimensional filtering module, a speckle noise suppression module, an edge enhancement module, an image fusion module, and an iterative control module, wherein:

[0152] The image acquisition module is used to acquire three-dimensional ultrasound images.

[0153] The three-dimensional filtering module is used to filter the acquired three-dimensional ultrasound images.

[0154] The speckle noise suppression module is used to perform three-dimensional speckle noise suppression processing on the filtered image.

[0155] The edge enhancement module is used to perform three-dimensional edge enhancement processing on the filtered image.

[0156] The image fusion module is used to fuse the image after noise suppression with the image after edge enhancement to obtain a first fused image, and to fuse the first fused image with the filtered image to obtain a second fused image.

[0157] The iteration control module controls the number of iterations. When the number of iterations reaches a set value, the iteration stops and the final result graph is output.

[0158] The purpose of the three-dimensional ultrasound image enhancement algorithm provided in this application is to improve image quality. This improvement can be subjectively evaluated visually, for example... Figure 4As shown, this is a comparison of the original three-dimensional ultrasound image and the image after enhancement processing using the enhancement method of this application. The image on the left is the original three-dimensional ultrasound image, and the image on the right is the image after enhancement processing using the enhancement method of this application. The comparison shows that after using the enhancement method of this application, the three-dimensional ultrasound image is smoother, the speckle noise is eliminated, and the detailed information of the original image is still preserved.

[0159] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0160] Based on the same inventive concept, this application also provides an image processing apparatus for implementing the image processing method described above. The solution provided by this apparatus is similar to the implementation scheme described in the above method; therefore, the specific limitations in one or more image processing apparatus embodiments provided below can be found in the limitations of the image processing method described above, and will not be repeated here.

[0161] In one embodiment, such as Figure 5 As shown, an image processing apparatus is provided, comprising:

[0162] Acquisition unit 510 is used to acquire an initial image;

[0163] The processing unit 520 is used to perform noise suppression processing and edge enhancement processing on the initial image respectively to obtain a denoised image and an enhanced image corresponding to the initial image;

[0164] The first fusion unit 530 is used to fuse the denoised image and the enhanced image to obtain the first fused image;

[0165] The second fusion unit 540 is used to fuse the first fused image and the initial image to obtain the second fused image;

[0166] The determining unit 550 is used to take the second fused image as a new initial image and return to the steps of performing speckle noise suppression processing and edge enhancement processing on the initial image respectively, until the preset iteration end condition is reached to obtain the target image corresponding to the initial image.

[0167] In one embodiment, the first fusion unit 530 is further configured to perform edge enhancement processing on the enhanced image using a preset filtering method to obtain a filtered enhanced image; and to perform fusion processing on the denoised image and the filtered enhanced image to obtain a first fused image.

[0168] In one embodiment, the first fusion unit 530 is further configured to obtain a first weight of the filtered enhanced image; determine a second weight of the denoised image based on the first weight of the filtered enhanced image; wherein the second weight of the denoised image is negatively correlated with the first weight; and perform fusion processing on the denoised image and the filtered enhanced image according to the first weight and the second weight to obtain a first fused image.

[0169] In one embodiment, the second fusion unit 540 is further configured to obtain the third weight of the first fused image and the fourth weight of the initial image; and to perform fusion processing on the first fused image and the initial image according to the third weight and the fourth weight to obtain the second fused image.

[0170] In one embodiment, the above-mentioned apparatus further includes a filtering unit for filtering the initial image to obtain a filtered image corresponding to the initial image;

[0171] The aforementioned processing unit 520 is further configured to perform noise suppression processing and edge enhancement processing on the filtered image corresponding to the initial image, respectively, to obtain a denoised image and an enhanced image of the filtered image, which are used as the denoised image and the enhanced image corresponding to the initial image.

[0172] In one embodiment, the iteration termination condition is determined based on image processing rate requirements and image quality requirements; the iteration termination condition is a preset number of iterations or the grayscale value change of the second fused image obtained from two adjacent iterations is less than a threshold.

[0173] In one embodiment, the processing unit 520 includes a suppression subunit, configured to acquire the diffusion tensor and image gradient of the initial image; obtain anisotropic diffusion information of the initial image based on the diffusion tensor and image gradient; the anisotropic diffusion information represents the rate of change of the image grayscale value of the initial image at a certain moment; and perform speckle and noise suppression processing on the initial image based on the anisotropic diffusion information to obtain a denoised image corresponding to the initial image.

[0174] In one embodiment, the suppression subunit is further configured to obtain the initial eigenvalues ​​and eigenvectors of the structure tensor of the initial image; the initial eigenvalues ​​represent the diffusion coefficients that diffuse along the direction of the eigenvectors; the initial eigenvalues ​​of the structure tensor are corrected to obtain the corrected eigenvalues; and the diffusion tensor of the initial image is obtained based on the corrected eigenvalues ​​and eigenvectors.

[0175] In one embodiment, the suppression subunit is further configured to uniformly organize regions in the initial image, wherein the three feature values ​​included in the corrected feature values ​​are approximately equal; in the edge regions of the initial image, the largest feature value among the three feature values ​​included in the corrected feature values ​​will stop spreading as the correction target, and the initial feature values ​​will be corrected to obtain the corrected feature values.

[0176] In one embodiment, the processing unit 520 further includes an enhancement subunit, configured to obtain the corrected feature values ​​of the structure tensor of the initial image; perform edge extraction processing on the initial image based on the corrected feature values ​​to obtain the image edge region of the initial image; and perform edge enhancement processing on the image edge region to obtain the enhanced image corresponding to the initial image.

[0177] In one embodiment, the enhancement subunit is further configured to determine two target feature values ​​for each pixel in the initial image from three corrected feature values ​​corresponding to the pixel; the two target feature values ​​include the feature value with the largest value and the feature value with the smallest value among the three corrected feature values; if the difference between the two target feature values ​​is greater than a threshold, the pixel is determined to be at the image edge of the initial image; and based on the extracted pixels at the image edge, the image edge region of the initial image is obtained.

[0178] Each module in the aforementioned image processing device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the operations corresponding to each module.

[0179] In one embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 6As shown, the computer device includes a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, NFC (Near Field Communication), or other technologies. When the computer program is executed by the processor, it implements an image processing method. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad mounted on the computer device casing, or an external keyboard, touchpad, or mouse.

[0180] Those skilled in the art will understand that Figure 6 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0181] In one embodiment, a computer device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above method embodiments.

[0182] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon that, when executed by a processor, implements the steps in the above method embodiments.

[0183] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above method embodiments.

[0184] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0185] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0186] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. An image processing method, characterized in that, The method includes: Get the initial image; The initial image is subjected to noise suppression processing and edge enhancement processing respectively to obtain the denoised image and the enhanced image corresponding to the initial image; The denoised image and the enhanced image are fused together to obtain a first fused image; The first fused image and the initial image are fused to obtain a second fused image; The second fused image is used as the new initial image, and the steps of performing speckle noise suppression and edge enhancement processing on the initial image are returned until the preset iteration end condition is reached to obtain the target image corresponding to the initial image.

2. The method according to claim 1, characterized in that, The step of fusing the denoised image and the enhanced image to obtain a first fused image includes: The edge enhancement results of the enhanced image are filtered using a preset filtering method to obtain the filtered enhanced image. The denoised image and the filtered enhanced image are fused together to obtain the first fused image.

3. The method according to claim 2, characterized in that, The step of fusing the denoised image and the filtered enhanced image to obtain the first fused image includes: Obtain the first weight of the filtered enhanced image; Based on the first weight of the enhanced image after filtering, a second weight of the denoised image is determined; wherein the second weight of the denoised image is negatively correlated with the first weight; The denoised image and the filtered enhanced image are fused according to the first weight and the second weight to obtain the first fused image.

4. The method according to claim 1, characterized in that, The step of fusing the first fused image and the initial image to obtain the second fused image includes: Obtain the third weight of the first fused image and the fourth weight of the initial image; the sum of the third weight and the fourth weight is 1; The first fused image and the initial image are fused according to the third weight and the fourth weight to obtain the second fused image.

5. The method according to claim 1, characterized in that, Before performing speckle noise suppression and edge enhancement processing on the initial image, the process further includes: The initial image is filtered to obtain the filtered image corresponding to the initial image; The process of performing speckle noise suppression and edge enhancement processing on the initial image to obtain a denoised image and an enhanced image corresponding to the initial image includes: The filtered image corresponding to the initial image is subjected to noise suppression processing and edge enhancement processing respectively to obtain the denoised image and enhanced image of the filtered image, which are used as the denoised image and enhanced image corresponding to the initial image.

6. The method according to claim 1, characterized in that, The iteration termination condition is determined based on image processing rate requirements and image quality requirements; The iteration termination condition is either a preset number of iterations or the grayscale value change of the second fused image obtained from two adjacent iterations is less than a threshold.

7. The method according to claim 1, characterized in that, The denoised image corresponding to the initial image is determined in the following way: Obtain the diffusion tensor and image gradient of the initial image; Based on the diffusion tensor and the image gradient, the anisotropic diffusion information of the initial image is obtained; the anisotropic diffusion information represents the rate of change of the image grayscale value of the initial image at a certain moment; Based on the anisotropic diffusion information, the initial image is subjected to speckle noise suppression processing to obtain the denoised image corresponding to the initial image.

8. The method according to claim 7, characterized in that, The step of obtaining the diffusion tensor of the initial image includes: Obtain the initial eigenvalues ​​and eigenvectors of the structure tensor of the initial image; the initial eigenvalues ​​represent the diffusion coefficients that diffuse along the direction of the eigenvectors. The initial eigenvalues ​​of the structure tensor are corrected to obtain the corrected eigenvalues. Based on the corrected eigenvalues ​​and the eigenvectors, the diffusion tensor of the initial image is obtained.

9. The method according to claim 8, characterized in that, The feature vector is a three-dimensional feature vector, and the corrected feature values ​​include three feature values ​​corresponding to the three-dimensional feature vector. The process of correcting the initial eigenvalues ​​of the structure tensor to obtain corrected eigenvalues ​​includes: In a uniformly organized region in the initial image, the corrected feature values ​​include three feature values ​​that are approximately equal. In the edge region of the initial image, the largest of the three feature values ​​included in the corrected feature value will stop spreading and become the correction target. The initial feature value is then corrected to obtain the corrected feature value.

10. The method according to claim 1, characterized in that, The enhanced image corresponding to the initial image is determined in the following manner: Obtain the corrected feature values ​​of the structure tensor of the initial image; Based on the corrected feature values, edge extraction processing is performed on the initial image to obtain the image edge region of the initial image; Edge enhancement processing is performed on the edge regions of the image to obtain the enhanced image corresponding to the initial image.

11. The method according to claim 10, characterized in that, The corrected feature values ​​include three feature values. Based on these corrected feature values, edge extraction processing is performed on the initial image to obtain the image edge region of the initial image, including: For each pixel in the initial image, two target feature values ​​are determined from the three corrected feature values ​​corresponding to the pixel; the two target feature values ​​include the feature value with the largest value and the feature value with the smallest value among the three corrected feature values. If the difference between the two target feature values ​​is greater than a threshold, then the pixel is determined to be at the image edge of the initial image; Based on the extracted pixels located at the image edges, the image edge region of the initial image is obtained.

12. An image processing apparatus, characterized in that, The device includes: The acquisition unit is used to acquire the initial image; The processing unit is used to perform noise suppression processing and edge enhancement processing on the initial image respectively to obtain a denoised image and an enhanced image corresponding to the initial image; The first fusion unit is used to fuse the denoised image and the enhanced image to obtain a first fused image; The second fusion unit is used to fuse the first fused image and the initial image to obtain the second fused image; The determining unit is used to take the second fused image as a new initial image and return to the steps of performing speckle noise suppression processing and edge enhancement processing on the initial image respectively, until a preset iteration end condition is reached to obtain the target image corresponding to the initial image.

13. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the image processing method according to any one of claims 1 to 11.

14. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the image processing method according to any one of claims 1 to 11.

Images