A method, device and electronic device for removing CT image artifacts

Abstract

The embodiments of the present application provide a method, device and electronic device for removing artifacts from CT images. The electronic device can obtain an original sinogram; solve the sinogram that makes a preset objective function obtain a minimum value as a smoothed sinogram; determine the characteristic attributes of the original projection image based on the difference between the original sinogram and the smoothed sinogram; use the characteristic attributes to correct the image to be processed to obtain a corrected image of the target object. In order to remove artifacts in the original sinogram, a sinogram that makes the objective function obtain a minimum value can be solved as a smoothed sinogram. Next, the original sinogram can be corrected using the artifacts in the original sinogram, and the smoothed sinogram can also be corrected using the smoothed details to obtain a corrected image of the target object. Since the corrected image includes all the details of the original sinogram but does not include artifacts, the artifact removal effect can be improved and the quality of the CT image can be improved.

Description

Technical Field

[0001] The present application relates to the technical field of CT image processing, and in particular to a method, device and electronic device for removing CT image artifacts. Background Art

[0002] CT (Computed Tomography) technology uses X-rays to perform layer-by-layer scanning of target objects, which can quickly and clearly image and is widely used for security inspections, object detection, disease diagnosis, etc. Due to the non-uniformity of the detector, ring artifacts are common in CT images. On this basis, since a detector with a large cone angle is used during the scanning process, the detector pixels have inconsistent responses to the energy of X-rays, and the detector will also be offset and jittered during the scanning process, which will cause the ring artifact to deform and produce arc-shaped or elliptical arc artifacts.

[0003] The above artifacts will seriously affect the quality of CT images. Currently, the following two methods are usually used to eliminate artifacts:

[0004] The first method is to scan the air image and the target object image, and correct the bad pixels and non-uniformity in the target object image through the air image. The second method is to transform the CT image into a polar coordinate image, correct the stripes in the polar coordinate image, filter the polar coordinate image, and then transform it back to the rectangular coordinate system.

[0005] The above two methods can eliminate heavier ring artifacts, but cannot eliminate lighter and more complex ring artifacts and arc artifacts. Therefore, the removal effect of artifacts in the current related technology is still poor. Summary of the invention

[0006] The purpose of the embodiments of the present application is to provide a method, device and electronic device for removing artifacts from CT images, so as to improve the effect of removing artifacts from CT images and enhance the quality of CT images. The specific technical solution is as follows:

[0007] In a first aspect, an embodiment of the present application provides a method for removing CT image artifacts, the method comprising:

[0008] Acquire an original sinusoidal graph obtained by converting an original projection image, wherein the original projection image is an image obtained by performing a CT scan on the target object;

[0009] Based on a preset objective function and the original sinusoidal graph, solving a sinusoidal graph that makes the preset objective function obtain a minimum value as a smoothed sinusoidal graph, wherein the preset objective function is used to characterize the error between the smoothed sinusoidal graph and the original sinusoidal graph;

[0010] Determine a characteristic attribute included in the original projection image based on a difference between the original sinusoidal graph and the smoothed sinusoidal graph, wherein the characteristic attribute is an artifact included in the original sinusoidal graph or a smoothed detail;

[0011] The image to be processed is corrected using the characteristic attribute to obtain a corrected image of the target object, wherein when the characteristic attribute is an artifact included in the original sinusoidal graph, the image to be processed is the original sinusoidal graph, and when the characteristic attribute is the smoothed details, the image to be processed is the smoothed sinusoidal graph.

[0012] Optionally, the step of determining the characteristic attributes of the original projection image based on the difference between the original sinusoidal graph and the smoothed sinusoidal graph comprises:

[0013] Subtract the pixel value of each pixel point in the original sinusoidal image from the pixel value of the same position in the smoothed sinusoidal image to obtain a texture image;

[0014] Performing artifact filtering on the texture image to obtain a filtered texture image;

[0015] Based on the difference between the texture image and the feature image, a feature attribute is determined, wherein, when the feature image is the filtered texture image, the feature attribute is an artifact included in the original sinusoidal graph, and when the feature image is an artifact feature, the feature attribute is the smoothed details.

[0016] Optionally, the characteristic attribute is an artifact included in the original sinusoidal graph, and the abscissa and ordinate of the sinusoidal graph are respectively the detector channel width and the projection angle;

[0017] The step of determining the feature attribute based on the difference between the texture image and the feature image comprises:

[0018] Subtracting the pixel value of each pixel point in the texture image from the pixel point at the same position in the filtered texture image to obtain a difference image;

[0019] At each projection angle, artifact features are determined based on pixel values ​​corresponding to the width of each detector channel in the difference image.

[0020] Optionally, the step of correcting the image to be processed by using the characteristic attribute to obtain a corrected image of the target object includes:

[0021] The pixel value of each pixel point in the original sinusoidal graph is subtracted from the pixel value of the same position in the artifact feature to obtain a corrected image of the target object.

[0022] Optionally, the characteristic attribute is the smoothed detail, and the abscissa and ordinate of the sinusoidal graph are the detector channel width and the projection angle, respectively;

[0023] The step of determining the feature attribute based on the difference between the texture image and the feature image comprises:

[0024] Subtracting the pixel value of each pixel point in the texture image from the pixel point at the same position in the filtered texture image to obtain a difference image;

[0025] At each projection angle, determining artifact features based on pixel values ​​corresponding to the width of each detector channel in the difference image;

[0026] The pixel value of each pixel point in the texture image is subtracted from the pixel value of the pixel point at the same position in the artifact feature to obtain the smoothed detail feature.

[0027] Optionally, the step of correcting the image to be processed by using the characteristic attribute to obtain a corrected image of the target object includes:

[0028] The pixel value of each pixel point in the smoothed detail feature is added to the pixel point at the same position in the smoothed sinusoidal graph to obtain a corrected image of the target object.

[0029] Optionally, the step of determining the artifact feature based on the pixel value corresponding to each detector channel width in the difference image at each projection angle includes:

[0030] For each row of the difference image, calculating the average value of the pixel values ​​of each pixel point in the row;

[0031] The average value of the pixel values ​​of each pixel point in the row is taken as the pixel value of each pixel point in the row to obtain the artifact feature.

[0032] Optionally, the step of solving a sinusoidal graph that minimizes the preset objective function based on the preset objective function and the original sinusoidal graph includes:

[0033] According to the difference between the smoothed sinusoidal graph represented by the preset objective function and the original sinusoidal graph, the sinusoidal graph is adjusted in an iterative manner so that the preset objective function obtains a minimum value until the preset objective function converges to obtain a smoothed sinusoidal graph.

[0034] In a second aspect, an embodiment of the present application provides a device for removing CT image artifacts, the device comprising:

[0035] A sinusoidal image acquisition module, used to acquire an original sinusoidal image obtained by converting an original projection image, wherein the original projection image is an image obtained by performing a CT scan on the target object;

[0036] A sinusoidal graph smoothing module, used for solving a sinusoidal graph that makes the preset objective function obtain a minimum value based on a preset objective function and the original sinusoidal graph, as a smoothed sinusoidal graph, wherein the preset objective function is used to characterize the error between the smoothed sinusoidal graph and the original sinusoidal graph;

[0037] A characteristic attribute determination module, used to determine the characteristic attribute included in the original projection image based on the difference between the original sinusoidal graph and the smoothed sinusoidal graph, wherein the characteristic attribute is an artifact included in the original sinusoidal graph or a smoothed detail;

[0038] An image correction module is used to correct the image to be processed using the characteristic attribute to obtain a corrected image of the target object, wherein when the characteristic attribute is an artifact included in the original sinusoidal graph, the image to be processed is the original sinusoidal graph; when the characteristic attribute is the smoothed details, the image to be processed is the smoothed sinusoidal graph.

[0039] In a third aspect, an embodiment of the present application provides an electronic device, including:

[0040] Memory, used to store computer programs;

[0041] The processor is used to implement any method described in the first aspect when executing a program stored in the memory.

[0042] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, any method described in the first aspect is implemented.

[0043] Beneficial effects of the embodiments of the present application:

[0044] In the scheme provided by the embodiment of the present application, the electronic device can obtain the original sinogram obtained by converting the original projection image, wherein the original projection image is an image obtained by performing CT scanning on the target object; based on the preset objective function and the original sinogram, the sinogram that makes the preset objective function obtain the minimum value is solved as the smoothed sinogram, wherein the preset objective function is used to characterize the error between the smoothed sinogram and the original sinogram; based on the difference between the original sinogram and the smoothed sinogram, the characteristic attribute included in the original projection image is determined, wherein the characteristic attribute is the artifact included in the original sinogram or the smoothed details; the image to be processed is corrected using the characteristic attribute to obtain the corrected image of the target object, wherein when the characteristic attribute is the artifact included in the original sinogram, the image to be processed is the original sinogram, and when the characteristic attribute is the smoothed details, the image to be processed is the smoothed sinogram. The original sinogram includes artifacts and details, wherein the artifacts need to be removed, while the details need to be retained. In order to remove the artifacts in the original sinogram, the sinogram that makes the objective function obtain the minimum value can be solved as the smoothed sinogram. The objective function represents the error between the smoothed sinogram and the original sinogram. When the error between the smoothed sinogram and the original sinogram is minimized, the smoothed sinogram can be as close to the original sinogram as possible. Therefore, the smoothed sinogram can retain the image edge details in the original sinogram as much as possible without including artifacts. However, some details will inevitably be smoothed out, that is, the smoothed sinogram only includes some details in the original sinogram. Next, the corrected image can be obtained by one of the following two implementations. In the first implementation, since the original sinogram includes artifacts and details, and the smoothed sinogram does not include artifacts but only includes some details, the artifacts and smoothed details in the original sinogram can be obtained based on the difference between the original sinogram and the smoothed sinogram. Then, the smoothed details can be obtained by artifact filtering, and the artifacts in the original sinogram filtered out by the artifact filtering are determined. Since the original sinogram includes artifacts and all details, the artifacts in the original sinogram can be used to correct the original sinogram to obtain a corrected image of the target object. In a second embodiment, the artifacts and smoothed details in the original sinogram can be obtained based on the difference between the original sinogram and the smoothed sinogram. Then, the smoothed details can be obtained by artifact filtering. Since the details not included in the smoothed sinogram are smoothed details, the smoothed details can be used to correct the smoothed sinogram to obtain a corrected image of the target object.By adopting the above two implementations, the corrected image of the target object can include all the details in the original sinusoidal image without artifacts, thereby improving the effect of removing artifacts in the CT image and improving the quality of the CT image. Of course, implementing any product or method of the present application does not necessarily require achieving all the advantages described above at the same time. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application, and for ordinary technicians in this field, other embodiments can also be obtained based on these drawings.

[0046] Figure 1 A flowchart of a method for removing CT image artifacts provided in an embodiment of the present application;

[0047] Figure 2(a) is based on Figure 1 A schematic diagram of an elliptical artifact of the illustrated embodiment;

[0048] Figure 2(b) is based on Figure 1 A schematic diagram of a ring artifact of the illustrated embodiment;

[0049] Figure 3 (a) is based on Figure 1 A schematic diagram of a CT image and an artifact image of a skull slice of the illustrated embodiment;

[0050] Figure 3(b) is based on Figure 1 A schematic diagram of a CT image of the skull coronal plane and an artifact image of the illustrated embodiment;

[0051] Figure 4 for Figure 1 A specific flow chart of step S103 in the illustrated embodiment;

[0052] Figure 5 Based on Figure 1 A schematic diagram of a sinusoidal graph and a texture image of the illustrated embodiment;

[0053] Figure 6 for Figure 4 A specific flow chart of step S403 in the illustrated embodiment;

[0054] Figure 7 for Figure 4 Another specific flow chart of step S403 in the illustrated embodiment;

[0055] Figure 8 for Figure 6 A specific flow chart of step S602 in the illustrated embodiment;

[0056] Fig. 9 Based on Figure 1 A flow chart of removing artifacts included in the original projection image according to the embodiment shown;

[0057] Fig.10 A schematic diagram of the structure of a device for removing CT image artifacts provided in an embodiment of the present application;

[0058] Fig.11 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0059] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field based on the present application belong to the scope of protection of the present application.

[0060] In order to improve the effect of removing artifacts in CT images and enhance the quality of CT images, the embodiments of the present application provide a method, device, electronic device, computer-readable storage medium and computer program product for removing artifacts in CT images. The following first introduces a method for removing artifacts in CT images provided in the embodiments of the present application.

[0061] The method for removing CT image artifacts provided in the embodiment of the present application can be applied to any electronic device that needs to remove CT image artifacts, for example, it can be an image processor, an image processor, a processing device, etc., which is not specifically limited here. For the sake of clarity, it is subsequently referred to as an electronic device in this article.

[0062] like Figure 1 As shown, a method for removing CT image artifacts, the method comprising:

[0063] S101, obtaining an original sinusoidal graph obtained by converting an original projection graph;

[0064] The original projection image is an image obtained by performing CT scanning on the target object.

[0065] S102, based on a preset objective function and the original sinusoidal graph, solving a sinusoidal graph that makes the preset objective function obtain a minimum value as a smoothed sinusoidal graph;

[0066] The preset objective function is used to characterize the error between the smoothed sinusoidal graph and the original sinusoidal graph.

[0067] S103, determining characteristic attributes included in the original projection image based on the difference between the original sinusoidal graph and the smoothed sinusoidal graph;

[0068] The characteristic attribute is the artifacts or smoothed details included in the original sinusoidal graph.

[0069] S104: Correct the image to be processed using the characteristic attribute to obtain a corrected image of the target object.

[0070] Wherein, when the characteristic attribute is the artifact included in the original sinusoidal graph, the image to be processed is the original sinusoidal graph; when the characteristic attribute is the smoothed details, the image to be processed is the smoothed sinusoidal graph.

[0071] It can be seen that in the embodiment of the present application, the electronic device can obtain the original sinusoidal graph obtained by converting the original projection image, wherein the original projection image is an image obtained by performing CT scanning on the target object; based on the preset objective function and the original sinusoidal graph, the sinusoidal graph that makes the preset objective function obtain the minimum value is solved as the smoothed sinusoidal graph, wherein the preset objective function is used to characterize the error between the smoothed sinusoidal graph and the original sinusoidal graph; based on the difference between the original sinusoidal graph and the smoothed sinusoidal graph, the characteristic attribute included in the original projection image is determined, wherein the characteristic attribute is the artifact included in the original sinusoidal graph or the smoothed details; the image to be processed is corrected using the characteristic attribute to obtain the corrected image of the target object, wherein when the characteristic attribute is the artifact included in the original sinusoidal graph, the image to be processed is the original sinusoidal graph, and when the characteristic attribute is the smoothed details, the image to be processed is the smoothed sinusoidal graph. The original sinusoidal graph includes artifacts and details, wherein the artifacts need to be removed, while the details need to be retained. In order to remove the artifacts in the original sinusoidal graph, the sinusoidal graph that makes the objective function obtain the minimum value can be solved as the smoothed sinusoidal graph. The objective function represents the error between the smoothed sinogram and the original sinogram. When the error between the smoothed sinogram and the original sinogram is minimized, the smoothed sinogram can be as close to the original sinogram as possible. Therefore, the smoothed sinogram can retain the image edge details in the original sinogram as much as possible without including artifacts. However, some details will inevitably be smoothed out, that is, the smoothed sinogram only includes some details in the original sinogram. Next, the corrected image can be obtained by one of the following two implementations. In the first implementation, since the original sinogram includes artifacts and details, and the smoothed sinogram does not include artifacts but only includes some details, the artifacts and smoothed details in the original sinogram can be obtained based on the difference between the original sinogram and the smoothed sinogram. Then, the smoothed details can be obtained by artifact filtering, and the artifacts in the original sinogram filtered out by the artifact filtering are determined. Since the original sinogram includes artifacts and all details, the artifacts in the original sinogram can be used to correct the original sinogram to obtain a corrected image of the target object. In a second embodiment, the artifacts and smoothed details in the original sinogram can be obtained based on the difference between the original sinogram and the smoothed sinogram. Then, the smoothed details can be obtained by artifact filtering. Since the details not included in the smoothed sinogram are smoothed details, the smoothed details can be used to correct the smoothed sinogram to obtain a corrected image of the target object.By adopting the above two implementations, the corrected image of the target object can include all the details in the original sinusoidal image without including artifacts, thereby improving the effect of removing artifacts in the CT image and improving the quality of the CT image.

[0072] When a target object is CT scanned, the original projection image of the target object can be collected from the CT scanning system. The original projection image may include annular, arc-shaped or elliptical artifacts. For example, a schematic diagram of an elliptical artifact can be shown in FIG2 (a), and a schematic diagram of an annular artifact can be shown in FIG2 (b). The target object is an object to be CT scanned, for example, industrial materials, human body parts, geological samples, etc.

[0073] Since the annular, arc-shaped or elliptical artifacts in the original projection image will become parallel vertical stripes in the original sinusoidal image after the original projection image is converted into the original sinusoidal image, in order to facilitate the removal of the artifacts in the original projection image, in step S101, the electronic device can obtain the original sinusoidal image obtained by converting the original projection image. The horizontal coordinate and the vertical coordinate of the original sinusoidal image can be the detector channel width and the projection angle, respectively. The original sinusoidal image includes artifacts and details. The original sinusoidal image can be represented as a superposition of an ideal image including details but not including artifacts and artifacts.

[0074] The execution subject of converting the original projection image into the original sine graph can be an electronic device or other device. Accordingly, the electronic device can obtain the original sine graph in the following two ways: In one embodiment, the electronic device can convert the original projection image into the original sine graph, and then obtain the original sine graph obtained by its own processing. In another embodiment, other devices can convert the original projection image into the original sine graph, and the electronic device can obtain the original sine graph from other devices.

[0075] In order to remove the artifacts included in the original sinusoidal graph, the original sinusoidal graph can be smoothed, but in the process of smoothing, some details included in the original sinusoidal graph will inevitably be smoothed out. Therefore, in order to retain the image edge details included in the original sinusoidal graph as much as possible during the smoothing process, a preset objective function can be pre-constructed, and the preset objective function is used to characterize the error between the smoothed sinusoidal graph and the original sinusoidal graph.

[0076] If the preset objective function obtains the minimum value, it means that the error between the smoothed sinusoidal graph and the original sinusoidal graph is the minimum, that is, the smoothed sinusoidal graph retains as many details as possible. Therefore, the electronic device can solve the sinusoidal graph that makes the preset objective function obtain the minimum value based on the preset objective function and the original sinusoidal graph, and use it as the smoothed sinusoidal graph, that is, execute step S102.

[0077] In one embodiment, the preset objective function The formula can be shown as follows:

[0078] ;

[0079] in, is the original sinusoidal graph, is the smoothed sinusoidal graph, is the position of the pixel, is the smoothing coefficient, is the penalty function, , Is the power of the penalty function. In order to perform edge-preserving smoothing filtering, it is usually The value range is set to , is the first preset constant, with a smaller value, usually set to . is the smoothed sinusoidal graph in The gradient of the position along the x-axis and y-axis, the gradient direction .

[0080] The first term in the above formula It means subtracting the pixel value of each pixel point in the original sinusoidal graph from the pixel value of the same position in the smoothed sinusoidal graph. The physical meaning of this term is the error between the smoothed sinusoidal graph and the original sinusoidal graph. It represents the sum of the gradients of each pixel along the x-axis and y-axis in the smoothed sinusoidal graph. The physical meaning of this item is the optimization direction of the sinusoidal graph.

[0081] After obtaining the smoothed sinusoidal graph, the electronic device can obtain the corrected image through step S103 and step S104. There are two implementations of step S103, and there are two corresponding implementations of step S104. The following will take step S103 and step S104 as a whole to introduce the two implementations mentioned above respectively:

[0082] In a first embodiment, the electronic device can determine the artifacts included in the original projection image based on the difference between the original sinusoidal graph and the smoothed sinusoidal graph, and then use the artifacts included in the original projection image to correct the original sinusoidal graph, thereby obtaining a corrected image of the target object.

[0083] Since the smoothed sinogram is obtained by smoothing the original projection image, the smoothed sinogram does not include artifacts but only some details. The difference between the original sinogram and the smoothed sinogram is the artifacts included in the original sinogram and the smoothed details. Through filtering, only the smoothed details can be retained through artifact filtering, and the artifacts included in the original sinogram filtered out by the artifact filtering can be determined.

[0084] Since the original sinusoidal image includes artifacts and all details, the artifacts included in the original projection image can be used to correct the original sinusoidal image, thereby obtaining a corrected image of the target object.

[0085] In a second embodiment, the electronic device may determine the smoothed details based on the difference between the original sinusoidal graph and the smoothed sinusoidal graph, and then use the smoothed details to correct the smoothed sinusoidal graph, thereby obtaining a corrected image of the target object.

[0086] As described above, based on the difference between the original sinogram and the smoothed sinogram, the artifacts included in the original projection image can be determined. On this basis, the artifacts in the original sinogram and the smoothed sinogram can be determined by the original sinogram and the smoothed sinogram, and the smoothed details can be obtained by combining the artifacts included in the original projection image. Since the details not included in the smoothed sinogram are the smoothed details, the smoothed details can be used to correct the smoothed sinogram, thereby obtaining a corrected image of the target object.

[0087] In the above two embodiments, the corrected image can be a corrected sinusoidal image. If a corrected CT image needs to be obtained, the corrected sinusoidal image can be subjected to an inverse process of converting the original projection image to the original sinusoidal image, thereby obtaining a corrected CT image, which is the original projection image after removing artifacts.

[0088] For example, the schematic diagrams of the original projection image, the corrected CT image, and the artifact image can be shown in FIG3 (a) and FIG3 (b): In FIG3 (a), from left to right are the original projection image of the skull slice, the corrected CT image of the skull slice, and the ring artifact image. The original projection image of the skull slice includes a ring artifact, and the corrected CT image of the skull slice is obtained by removing the ring artifact from the original projection image of the skull slice.

[0089] In Figure 3 (b), from left to right are the original projection image of the skull coronal plane, the corrected CT image of the skull coronal plane, and the stripe artifact image. Among them, the original projection image of the skull coronal plane includes stripe artifacts, and the corrected CT image of the skull coronal plane is obtained by removing the stripe artifacts from the original projection image of the skull coronal plane.

[0090] It can be seen that in the embodiment of the present application, since the annular, arc-shaped or elliptical artifacts in the original projection image will become parallel vertical stripes in the original sinusoidal image after the original projection image is converted into the original sinusoidal image, it can be processed based on the projection domain, thereby eliminating the inconsistency of the same detector channel, and the operation is relatively simple. In this way, not only can the standard annular artifacts be effectively removed, but also the arc-shaped and elliptical artifacts can be effectively suppressed, thereby making the CT image more uniform and improving the image quality.

[0091] As an implementation method of the present application, Figure 4 As shown, the step of determining the characteristic attributes of the original projection image based on the difference between the original sinusoidal graph and the smoothed sinusoidal graph may include:

[0092] S401, subtracting the pixel value of each pixel point in the original sinusoidal image from the pixel value of the same position in the smoothed sinusoidal image to obtain a texture image;

[0093] Since the smoothing operation does not change the image dimension, the dimensions of the original sinogram and the smoothed sinogram are the same, and the pixel value of each pixel in the original sinogram can be subtracted from the pixel value of the pixel in the same position in the smoothed sinogram to obtain the texture image. Since the original sinogram includes artifacts and all details, and the smoothed sinogram only includes some details, the texture image includes artifacts and smoothed details.

[0094] For example, a schematic diagram of a sinusoidal graph and a texture image can be shown as Figure 5 Shown: In Figure 5 In the figure, from left to right are the original sinogram, the smoothed sinogram, and the texture image. The smoothed sinogram is obtained by smoothing the original sinogram, and the texture image is obtained by subtracting the pixel value of each pixel point in the original sinogram from the pixel value of the same position in the smoothed sinogram.

[0095] The following formula will be used to describe the above steps. Assuming that the original sinusoidal graph is , the smoothed sinusoidal graph is , the texture image is , then we can use the formula , determine the texture image .

[0096] S402, performing artifact filtering processing on the texture image to obtain a filtered texture image;

[0097] After obtaining the texture image, the texture image can be subjected to an artifact filtering process to obtain a filtered texture image. Since the artifact filtering process can remove the artifacts included in the texture image, the filtered texture image does not include the artifacts, but only includes the smoothed details. The artifact filtering process can adopt any filtering method, for example, Gaussian filtering, median filtering, mean filtering, etc. can be adopted.

[0098] S403: Determine a feature attribute based on the difference between the texture image and the feature image.

[0099] The electronic device may determine the feature attribute based on the difference between the texture image and the feature image. The feature image may be a filtered texture image or an artifact feature. When the feature image is a filtered texture image, the feature attribute is an artifact included in the original sinusoidal image; when the feature image is an artifact feature, the feature attribute is a smoothed detail.

[0100] It can be seen that in the embodiment of the present application, the electronic device can subtract the pixel value of each pixel point in the original sinusoidal graph from the pixel value of the same position in the smoothed sinusoidal graph to obtain a texture image; perform artifact filtering on the texture image to obtain a filtered texture image; and determine the feature attributes based on the difference between the texture image and the feature image. Since the original sinusoidal graph includes artifacts and all details, and the smoothed sinusoidal graph only includes some details, the texture image includes artifacts and smoothed details, and the texture image can be subjected to artifact filtering to obtain a filtered texture image. Next, based on the filtered texture image or artifact features, combined with the texture image, the artifacts or smoothed details included in the original sinusoidal graph can be determined. In this way, the artifacts or smoothed details included in the original sinusoidal graph can be quickly and accurately determined.

[0101] As an implementation method of the present application, the above-mentioned characteristic attribute may be an artifact included in the original sinusoidal graph, and the abscissa and ordinate of the sinusoidal graph may be the detector channel width and the projection angle, respectively. In this case, Figure 6 As shown, the step of determining the feature attribute based on the difference between the texture image and the feature image may include:

[0102] S601, subtracting the pixel value of each pixel point in the texture image from the pixel point at the same position in the filtered texture image to obtain a difference image;

[0103] Since the artifact filtering process does not change the image dimension, the texture image and the filtered texture image have the same dimension, so the pixel value of each pixel in the texture image can be subtracted from the pixel value of the same position in the filtered texture image to obtain a difference image. Since the texture image includes artifacts and smoothed details, and the filtered texture image only includes smoothed details, the difference image only includes artifacts, and the pixel value of each pixel in the difference image is the pixel value of the artifact.

[0104] S602 , determining artifact features based on pixel values ​​corresponding to the width of each detector channel in the difference image at each projection angle.

[0105] Since annular, arc-shaped or elliptical artifacts appear as horizontal vertical stripes in the difference image, and the horizontal and vertical coordinates of the difference image are the detector channel width and the projection angle, respectively, the electronic device can determine the artifact features based on the pixel values ​​corresponding to the width of each detector channel in the difference image at each projection angle. Among them, the artifact features are also called artifact lateral features.

[0106] It can be seen that in the embodiment of the present application, the electronic device can subtract the pixel value of each pixel point in the texture image from the pixel value of the same position in the filtered texture image to obtain a difference image; at each projection angle, the artifact features are determined based on the pixel values ​​corresponding to the widths of each detector channel in the difference image. Since the texture image includes artifacts and smoothed details, and the filtered texture image only includes smoothed details, the difference image only includes artifacts. Since artifacts appear as horizontal vertical stripes in the difference image, the artifact features can be determined at each projection angle based on the pixel values ​​corresponding to the widths of each detector channel in the difference image. In this way, artifact features can be extracted quickly and accurately.

[0107] As an implementation manner of the embodiment of the present application, the step of correcting the image to be processed by using the feature attribute to obtain the corrected image of the target object may include:

[0108] The pixel value of each pixel point in the original sinusoidal graph is subtracted from the pixel value of the same position in the artifact feature to obtain a corrected image of the target object.

[0109] Since the artifact feature is obtained by smoothing the original sinusoidal graph and performing artifact filtering, and these operations do not change the image dimension, the artifact feature has the same image dimension as the original sinusoidal graph, and the pixel value of each pixel point in the original sinusoidal graph and the pixel value of the same position in the artifact feature can be subtracted to obtain the corrected image of the target object. Since the original sinusoidal graph includes artifacts and details, by removing the artifacts in the original sinusoidal graph, the corrected image obtained only includes the details of the target object.

[0110] It can be seen that in the embodiment of the present application, the electronic device can subtract the pixel value of each pixel point in the original sinusoidal graph from the pixel value of the same position in the artifact feature to obtain a corrected image of the target object. Since the corrected image is obtained by removing the artifacts in the original sinusoidal graph, the corrected image only includes the details of the target object but not the artifacts, so the quality of the CT image can be improved.

[0111] As an implementation method of the embodiment of the present application, the above-mentioned characteristic attribute can be a smoothed detail, and the horizontal axis and the vertical axis of the sinusoidal graph are the detector channel width and the projection angle respectively. In this case, Figure 7 As shown, the step of determining the feature attribute based on the difference between the texture image and the feature image may include:

[0112] S701, subtracting the pixel value of each pixel point in the texture image from the pixel point at the same position in the filtered texture image to obtain a difference image;

[0113] Since step S701 is the same as the above step S601, it will not be described again here.

[0114] S702, determining artifact features based on pixel values ​​corresponding to the width of each detector channel in the difference image at each projection angle;

[0115] Since step S702 is the same as the above step S602, it will not be described again here.

[0116] S703: Subtract the pixel value of each pixel point in the texture image from the pixel point at the same position in the artifact feature to obtain the smoothed detail feature.

[0117] Since the artifact feature has the same image dimension as the original sinusoidal image, and the texture image has the same image dimension as the original sinusoidal image, the artifact feature has the same image dimension as the texture image, and the pixel value of each pixel in the texture image can be subtracted from the pixel value of the same position in the artifact feature. Since the texture image includes artifacts and smoothed details, and the artifact feature only includes artifacts, the smoothed detail feature can be obtained by subtracting the pixel value of each pixel in the texture image from the pixel value of the same position in the artifact feature.

[0118] The following formula will be used to describe the above steps, assuming that the texture image is , the artifact characteristics are , the smoothed detail features are , then we can use the formula , determine the detail features that are smoothed out .

[0119] It can be seen that in the embodiment of the present application, the electronic device can subtract the pixel value of each pixel point in the texture image from the pixel value of the pixel point at the same position in the filtered texture image to obtain a difference image; at each projection angle, the artifact feature is determined based on the pixel value corresponding to the width of each detector channel in the difference image; and each pixel point in the texture image is subtracted from the pixel value of the pixel point at the same position in the artifact feature to obtain the smoothed detail feature. Since the texture image includes artifacts and smoothed details, and the artifact feature only includes artifacts, the smoothed detail feature can be quickly and accurately obtained by subtracting the pixel value of each pixel point in the texture image from the pixel value of the pixel point at the same position in the artifact feature.

[0120] As an implementation manner of the embodiment of the present application, the step of correcting the image to be processed by using the feature attribute to obtain the corrected image of the target object may include:

[0121] The pixel value of each pixel point in the smoothed detail feature is added to the pixel point at the same position in the smoothed sinusoidal graph to obtain a corrected image of the target object.

[0122] Since the smoothed detail features and the smoothed sinogram have the same image dimension as the original sinogram, the smoothed detail features have the same image dimension as the smoothed sinogram, and the pixel values ​​of each pixel in the smoothed detail features and the pixel in the same position in the smoothed sinogram can be added. Since the missing details in the smoothed sinogram are the smoothed details, the pixel values ​​of each pixel in the smoothed detail features and the pixel in the same position in the smoothed sinogram are added, and the corrected image obtained includes all the details.

[0123] The following formula will be used to describe the above steps, assuming that the corrected image of the target object is , the smoothed detail features are , the smoothed sinusoidal graph is , then we can use the formula , determine the corrected image as .

[0124] It can be seen that in the embodiment of the present application, the electronic device can add the pixel value of each pixel point in the smoothed detail feature to the pixel value of the pixel point at the same position in the smoothed sinusoidal map to obtain a corrected image of the target object. Since the pixel value of each pixel point in the original sinusoidal map is subtracted from the pixel value of the pixel point at the same position in the artifact feature, uneven noise points will be present in the image, and these noise points will make the uniformity of the image worse. However, by adding the pixel value of each pixel point in the smoothed detail feature to the pixel value of the pixel point at the same position in the smoothed sinusoidal map, the above problem can be optimized, thereby obtaining an image with better uniformity.

[0125] As an implementation method of the present application, Figure 8 As shown, the step of determining the artifact feature based on the pixel value corresponding to each detector channel width in the difference image at each projection angle may include:

[0126] S801, for each row of the difference image, calculating the average value of pixel values ​​of each pixel point in the row;

[0127] In order to eliminate the influence of noise on the image, the electronic device can calculate the average value of the pixel values ​​of each pixel point in each row of the difference image. The following formula will be used to describe the above steps, assuming that the average value is , the texture image is , the texture image after filtering is , we can use the formula , solve for the average .

[0128] S802: Taking the average value of the pixel values ​​of each pixel point in the row as the pixel value of each pixel point in the row, and obtaining an artifact feature.

[0129] Next, the electronic device can use the average value of the pixel values ​​of each pixel point in each row as the pixel value of each pixel point in the row for each row of the difference image to obtain the artifact feature. The average value of the pixel values ​​of each pixel in the row is , then we can The pixel values ​​of each pixel in the row are set to .

[0130] It can be seen that in the embodiment of the present application, the electronic device can calculate the average value of the pixel values ​​of each pixel point in each row of the difference image; the average value of the pixel values ​​of each pixel point in the row is used as the pixel value of each pixel point in the row to obtain the artifact feature. By taking the average value, the influence of noise on the image can be eliminated, so that the artifact feature does not include noise, which can further improve the quality of the CT image.

[0131] As an implementation manner of an embodiment of the present application, the step of solving the sine graph that makes the preset objective function obtain the minimum value based on the preset objective function and the original sine graph may include:

[0132] According to the difference between the smoothed sinusoidal graph represented by the preset objective function and the original sinusoidal graph, the sinusoidal graph is adjusted in an iterative manner so that the preset objective function obtains a minimum value until the preset objective function converges to obtain a smoothed sinusoidal graph.

[0133] As mentioned above, the preset objective function The formula can be shown as follows:

[0134] ;

[0135] In order to reduce the difference between the smoothed sinusoidal graph and the original sinusoidal graph, the sinusoidal graph can be adjusted iteratively so that the preset objective function Get the minimum value to solve the preset objective function Get the minimum value of the sine graph and get the smoothed sine graph , the specific formula can be as follows:

[0136] ;

[0137] in, is the number of iterations, Indicates The sine graph of the iteration, Indicates Next, we can use the preset objective function Substituting the formula into the previous formula, we get:

[0138] ;

[0139] Next, in order to solve When taking the minimum value The value of can be used to solve the extreme value of this formula:

[0140] Assumptions is a strictly convex function, then Satisfies the following relationship:

[0141] ;

[0142] in, for The derivative of When the equality sign holds, ,at this time Get extreme values.

[0143] because is a strictly convex function, so we can get:

[0144] ;

[0145] in, for The derivative of can define the intermediate variable , then we can get , yes The inverse function of .

[0146] Based on this, the above The formula is rewritten as:

[0147] ;

[0148] By deforming the right side, we can get:

[0149] ;

[0150] Among them, the intermediate variable ,constant , for The derivative of .

[0151] In the previous formula is a constant and can be omitted in the solution process, so the final formula can be obtained:

[0152] .

[0153] Based on this formula, electronic devices can solve the smoothed sine graph through iteration.

[0154] It can be seen that in the embodiment of the present application, the electronic device can adjust the sinogram in an iterative manner according to the difference between the smoothed sinogram represented by the preset objective function and the original sinogram, so that the preset objective function obtains a minimum value, until the preset objective function converges, and a smoothed sinogram is obtained. Since the objective function can characterize the difference between the smoothed sinogram and the original sinogram, the difference between the sinogram that makes the objective function obtain the minimum value and the original sinogram is the smallest, that is, the details in the original sinogram are retained to the greatest extent. Therefore, by processing the original sinogram in the above manner, the smoothed sinogram can retain the details in the original sinogram to the greatest extent while not including artifacts.

[0155] As an implementation method of the embodiment of the present application, a flowchart for removing artifacts included in the original projection image can be as follows: Fig. 9 As shown, the following steps may be specifically included:

[0156] S901, obtaining an original sinusoidal image including artifacts;

[0157] The electronic device can obtain an original sinusoidal graph obtained by converting the original projection image, wherein the original sinusoidal graph includes artifacts and details.

[0158] S902, inputting a smoothing coefficient, a power of a penalty function, and the number of iterations, and solving a smoothed sinusoidal graph based on a preset objective function;

[0159] The electronic device can be based on the smoothing coefficient input by the user , the power of the penalty function And the number of iterations , based on the preset objective function , through iteration, solve the preset objective function Get the minimum value of the sine graph as the smoothed sine graph .

[0160] Among them, the preset objective function , solve the preset objective function by iteration Based on the formula .

[0161] S903, solving the texture image;

[0162] The electronic device can subtract the pixel value of each pixel point in the original sinusoidal image from the pixel value of the pixel point at the same position in the smoothed sinusoidal image to obtain a texture image.

[0163] S904, performing artifact filtering on the texture image to obtain a filtered texture image;

[0164] The electronic device can perform artifact filtering processing on the texture image to obtain a filtered texture image, and the filtered texture image only includes smoothed details but does not include artifacts.

[0165] S905, solving artifact features;

[0166] The electronic device can subtract the pixel value of each pixel point in the texture image from the pixel value of the pixel point at the same position in the filtered texture image to obtain a difference image. At each projection angle, the artifact feature is determined based on the pixel value corresponding to each detector channel width in the difference image.

[0167] S906, solving the smoothed detail features;

[0168] The electronic device can subtract the pixel value of each pixel point in the texture image from the pixel point at the same position in the artifact feature to obtain the smoothed detail feature.

[0169] S907, solving the corrected image.

[0170] The electronic device can add the pixel value of each pixel point in the smoothed detail feature to the pixel point at the same position in the smoothed sinusoidal graph to obtain a corrected image of the target object.

[0171] In the technical solution of this application, the operations involved in obtaining, storing, using, processing, transmitting, providing and disclosing user personal information are all carried out with the user's authorization.

[0172] Corresponding to the above-mentioned method for removing CT image artifacts, the embodiment of the present application further provides a device for removing CT image artifacts. The device for removing CT image artifacts provided by the embodiment of the present application is introduced below.

[0173] like Fig.10 As shown, a device for removing CT image artifacts, the device comprising:

[0174] A sinusoidal graph acquisition module 1001 is used to acquire an original sinusoidal graph obtained by converting an original projection graph, wherein the original projection graph is an image obtained by performing a CT scan on a target object;

[0175] A sinusoidal graph smoothing module 1002 is used to solve a sinusoidal graph that makes the preset objective function obtain a minimum value based on a preset objective function and the original sinusoidal graph, as a smoothed sinusoidal graph, wherein the preset objective function is used to characterize the error between the smoothed sinusoidal graph and the original sinusoidal graph;

[0176] A characteristic attribute determination module 1003 is used to determine the characteristic attribute included in the original projection image based on the difference between the original sinusoidal graph and the smoothed sinusoidal graph, wherein the characteristic attribute is an artifact included in the original sinusoidal graph or a smoothed detail;

[0177] The image correction module 1004 is used to correct the image to be processed using the characteristic attribute to obtain a corrected image of the target object, wherein when the characteristic attribute is the artifact included in the original sinusoidal graph, the image to be processed is the original sinusoidal graph, and when the characteristic attribute is the smoothed details, the image to be processed is the smoothed sinusoidal graph.

[0178] It can be seen that in the embodiment of the present application, the electronic device can obtain the original sinusoidal graph obtained by converting the original projection image, wherein the original projection image is an image obtained by performing CT scanning on the target object; based on the preset objective function and the original sinusoidal graph, the sinusoidal graph that makes the preset objective function obtain the minimum value is solved as the smoothed sinusoidal graph, wherein the preset objective function is used to characterize the error between the smoothed sinusoidal graph and the original sinusoidal graph; based on the difference between the original sinusoidal graph and the smoothed sinusoidal graph, the characteristic attribute included in the original projection image is determined, wherein the characteristic attribute is the artifact included in the original sinusoidal graph or the smoothed details; the image to be processed is corrected using the characteristic attribute to obtain the corrected image of the target object, wherein when the characteristic attribute is the artifact included in the original sinusoidal graph, the image to be processed is the original sinusoidal graph, and when the characteristic attribute is the smoothed details, the image to be processed is the smoothed sinusoidal graph. The original sinusoidal graph includes artifacts and details, wherein the artifacts need to be removed, while the details need to be retained. In order to remove the artifacts in the original sinusoidal graph, the sinusoidal graph that makes the objective function obtain the minimum value can be solved as the smoothed sinusoidal graph. The objective function represents the error between the smoothed sinogram and the original sinogram. When the error between the smoothed sinogram and the original sinogram is minimized, the smoothed sinogram can be as close to the original sinogram as possible. Therefore, the smoothed sinogram can retain the image edge details in the original sinogram as much as possible without including artifacts. However, some details will inevitably be smoothed out, that is, the smoothed sinogram only includes some details in the original sinogram. Next, the corrected image can be obtained by one of the following two implementations. In the first implementation, since the original sinogram includes artifacts and details, and the smoothed sinogram does not include artifacts but only includes some details, the artifacts and smoothed details in the original sinogram can be obtained based on the difference between the original sinogram and the smoothed sinogram. Then, the smoothed details can be obtained by artifact filtering, and the artifacts in the original sinogram filtered out by the artifact filtering are determined. Since the original sinogram includes artifacts and all details, the artifacts in the original sinogram can be used to correct the original sinogram to obtain a corrected image of the target object. In a second embodiment, the artifacts and smoothed details in the original sinogram can be obtained based on the difference between the original sinogram and the smoothed sinogram. Then, the smoothed details can be obtained by artifact filtering. Since the details not included in the smoothed sinogram are smoothed details, the smoothed details can be used to correct the smoothed sinogram to obtain a corrected image of the target object.By adopting the above two implementations, the corrected image of the target object can include all the details in the original sinusoidal image without including artifacts, thereby improving the effect of removing artifacts in the CT image and improving the quality of the CT image.

[0179] As an implementation of the embodiment of the present application, the above-mentioned feature attribute determination module 1003 may include:

[0180] A texture image determination submodule, used for subtracting the pixel value of each pixel point in the original sinusoidal image from the pixel value of the same position in the smoothed sinusoidal image to obtain a texture image;

[0181] A texture image filtering submodule, used for performing an artifact filtering process on the texture image to obtain a filtered texture image;

[0182] A feature attribute determination submodule is used to determine the feature attribute based on the difference between the texture image and the feature image, wherein, when the feature image is the filtered texture image, the feature attribute is the artifact included in the original sinusoidal graph; when the feature image is an artifact feature, the feature attribute is the smoothed details.

[0183] As an implementation of an embodiment of the present application, the above-mentioned characteristic attribute may be an artifact included in the original sinusoidal graph, and the abscissa and ordinate of the sinusoidal graph may be the detector channel width and the projection angle, respectively. In this case, the above-mentioned characteristic attribute determination submodule may include:

[0184] A difference image determining unit, configured to subtract the pixel value of each pixel point in the texture image from the pixel value of the same position in the filtered texture image to obtain a difference image;

[0185] The artifact feature determination unit is used to determine the artifact feature at each projection angle based on the pixel value corresponding to the width of each detector channel in the difference image.

[0186] As an implementation of the embodiment of the present application, the above-mentioned image correction module 1004 may include:

[0187] The first image correction submodule is used to subtract the pixel value of each pixel point in the original sinusoidal image from the pixel value of the same position in the artifact feature to obtain a corrected image of the target object.

[0188] As an implementation of an embodiment of the present application, the above-mentioned characteristic attribute may be a smoothed detail, and the abscissa and ordinate of the sinusoidal graph may be the detector channel width and the projection angle, respectively. In this case, the above-mentioned characteristic attribute determination submodule may include:

[0189] A difference image determining unit, configured to subtract the pixel value of each pixel point in the texture image from the pixel value of the same position in the filtered texture image to obtain a difference image;

[0190] An artifact feature determination unit, configured to determine an artifact feature based on a pixel value corresponding to a width of each detector channel in the difference image at each projection angle;

[0191] The detail feature determination unit is used to subtract the pixel value of each pixel point in the texture image from the pixel value of the same position in the artifact feature to obtain the smoothed detail feature.

[0192] As an implementation of the embodiment of the present application, the above-mentioned image correction module 1004 may include:

[0193] The second image correction submodule is used to add the pixel value of each pixel point in the smoothed detail feature and the pixel point at the same position in the smoothed sinusoidal graph to obtain a corrected image of the target object.

[0194] As an implementation manner of the embodiment of the present application, the above-mentioned artifact feature determination unit may include:

[0195] an average value calculating subunit, used for calculating, for each row of the difference image, an average value of the pixel values ​​of each pixel point in the row;

[0196] The artifact feature determination unit is used to take the average value of the pixel values ​​of each pixel point in the row as the pixel value of each pixel point in the row to obtain the artifact feature.

[0197] As an implementation of the embodiment of the present application, the sinusoidal graph smoothing module 1002 may include:

[0198] The sine graph smoothing submodule is used to adjust the sine graph in an iterative manner according to the difference between the smoothed sine graph represented by the preset objective function and the original sine graph, so that the preset objective function obtains a minimum value until the preset objective function converges to obtain a smoothed sine graph.

[0199] The present application also provides an electronic device, such as Fig.11 As shown, including:

[0200] Memory 1101, used for storing computer programs;

[0201] The processor 1102 is used to implement the method for removing CT image artifacts described in any of the above embodiments when executing the program stored in the memory 1101.

[0202] In addition, the electronic device may further include a communication bus and / or a communication interface, and the processor 1102, the communication interface, and the memory 1101 communicate with each other via the communication bus.

[0203] It can be seen that in the embodiment of the present application, the electronic device can obtain the original sinusoidal graph obtained by converting the original projection image, wherein the original projection image is an image obtained by performing CT scanning on the target object; based on the preset objective function and the original sinusoidal graph, the sinusoidal graph that makes the preset objective function obtain the minimum value is solved as the smoothed sinusoidal graph, wherein the preset objective function is used to characterize the error between the smoothed sinusoidal graph and the original sinusoidal graph; based on the difference between the original sinusoidal graph and the smoothed sinusoidal graph, the characteristic attribute included in the original projection image is determined, wherein the characteristic attribute is the artifact included in the original sinusoidal graph or the smoothed details; the image to be processed is corrected using the characteristic attribute to obtain the corrected image of the target object, wherein when the characteristic attribute is the artifact included in the original sinusoidal graph, the image to be processed is the original sinusoidal graph, and when the characteristic attribute is the smoothed details, the image to be processed is the smoothed sinusoidal graph. The original sinusoidal graph includes artifacts and details, wherein the artifacts need to be removed, while the details need to be retained. In order to remove the artifacts in the original sinusoidal graph, the sinusoidal graph that makes the objective function obtain the minimum value can be solved as the smoothed sinusoidal graph. The objective function represents the error between the smoothed sinogram and the original sinogram. When the error between the smoothed sinogram and the original sinogram is minimized, the smoothed sinogram can be as close to the original sinogram as possible. Therefore, the smoothed sinogram can retain the image edge details in the original sinogram as much as possible without including artifacts. However, some details will inevitably be smoothed out, that is, the smoothed sinogram only includes some details in the original sinogram. Next, the corrected image can be obtained by one of the following two implementations. In the first implementation, since the original sinogram includes artifacts and details, and the smoothed sinogram does not include artifacts but only includes some details, the artifacts and smoothed details in the original sinogram can be obtained based on the difference between the original sinogram and the smoothed sinogram. Then, the smoothed details can be obtained by artifact filtering, and the artifacts in the original sinogram filtered out by the artifact filtering are determined. Since the original sinogram includes artifacts and all details, the artifacts in the original sinogram can be used to correct the original sinogram to obtain a corrected image of the target object. In a second embodiment, the artifacts and smoothed details in the original sinogram can be obtained based on the difference between the original sinogram and the smoothed sinogram. Then, the smoothed details can be obtained by artifact filtering. Since the details not included in the smoothed sinogram are smoothed details, the smoothed details can be used to correct the smoothed sinogram to obtain a corrected image of the target object.By adopting the above two implementations, the corrected image of the target object can include all the details in the original sinusoidal image without including artifacts, thereby improving the effect of removing artifacts in the CT image and improving the quality of the CT image.

[0204] The communication bus mentioned in the above electronic device can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. The communication bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, only one thick line is used in the figure, but it does not mean that there is only one bus or one type of bus.

[0205] The communication interface is used for communication between the above electronic device and other devices.

[0206] The memory may include a random access memory (RAM) or a non-volatile memory (NVM), such as at least one disk memory. Optionally, the memory may also be at least one storage device located away from the aforementioned processor.

[0207] The above-mentioned processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.

[0208] In another embodiment provided in the present application, a computer-readable storage medium is provided, in which a computer program is stored. When the computer program is executed by a processor, the steps of any of the above-mentioned methods for removing artifacts from CT images are implemented.

[0209] In another embodiment provided by the present application, a computer program product including instructions is also provided, which, when executed on a computer, enables the computer to execute any method for removing CT image artifacts in the above embodiments.

[0210] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on the computer, the process or function according to the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from a website site, computer, server or data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode to another website site, computer, server or data center. The computer-readable storage medium can be any available medium that can be accessed by the computer or a data storage device such as a server or data center that includes one or more available media integrated. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a solid-state hard disk (SSD), etc.

[0211] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0212] Each embodiment in this specification is described in a related manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the device, electronic device, computer-readable storage medium, and computer program product embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiments.

[0213] The above are only preferred embodiments of the present application and are not intended to limit the protection scope of the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application are included in the protection scope of the present application.

Claims

1. A method for removing CT image artifacts, characterized in that: The method comprises: Acquire an original sinusoidal graph obtained by converting an original projection graph, wherein the original projection graph is an image obtained by performing a CT scan on the target object; Based on a preset objective function and the original sinusoidal graph, solving a sinusoidal graph that makes the preset objective function obtain a minimum value as a smoothed sinusoidal graph, wherein the preset objective function is used to characterize the error between the smoothed sinusoidal graph and the original sinusoidal graph; Determine a characteristic attribute included in the original projection image based on a difference between the original sinusoidal graph and the smoothed sinusoidal graph, wherein the characteristic attribute is an artifact included in the original sinusoidal graph or a smoothed detail; The image to be processed is corrected using the characteristic attribute to obtain a corrected image of the target object, wherein when the characteristic attribute is an artifact included in the original sinusoidal graph, the image to be processed is the original sinusoidal graph, and when the characteristic attribute is the smoothed details, the image to be processed is the smoothed sinusoidal graph.

2. The method according to claim 1, characterized in that: The step of determining the characteristic attributes of the original projection image based on the difference between the original sinusoidal graph and the smoothed sinusoidal graph comprises: Subtract the pixel value of each pixel point in the original sinusoidal image from the pixel value of the same position in the smoothed sinusoidal image to obtain a texture image; Performing artifact filtering on the texture image to obtain a filtered texture image; Based on the difference between the texture image and the feature image, a feature attribute is determined, wherein, when the feature image is the filtered texture image, the feature attribute is an artifact included in the original sinusoidal graph, and when the feature image is an artifact feature, the feature attribute is the smoothed details.

3. The method according to claim 2, characterized in that The characteristic attribute is the artifact included in the original sinusoidal graph, and the abscissa and ordinate of the sinusoidal graph are the detector channel width and the projection angle respectively; The step of determining the feature attribute based on the difference between the texture image and the feature image comprises: Subtracting the pixel value of each pixel point in the texture image from the pixel point at the same position in the filtered texture image to obtain a difference image; At each projection angle, artifact features are determined based on pixel values ​​corresponding to the width of each detector channel in the difference image.

4. The method according to claim 3, characterized in that The step of correcting the image to be processed by using the characteristic attribute to obtain a corrected image of the target object comprises: The pixel value of each pixel point in the original sinusoidal graph is subtracted from the pixel value of the same position in the artifact feature to obtain a corrected image of the target object.

5. The method according to claim 2, characterized in that: The characteristic attribute is the smoothed details, and the abscissa and ordinate of the sinusoidal graph are the detector channel width and the projection angle, respectively; The step of determining the feature attribute based on the difference between the texture image and the feature image comprises: Subtracting the pixel value of each pixel point in the texture image from the pixel point at the same position in the filtered texture image to obtain a difference image; At each projection angle, determining artifact features based on pixel values ​​corresponding to the width of each detector channel in the difference image; The pixel value of each pixel point in the texture image is subtracted from the pixel value of the pixel point at the same position in the artifact feature to obtain the smoothed detail feature.

6. The method according to claim 5, characterized in that The step of correcting the image to be processed by using the characteristic attribute to obtain a corrected image of the target object comprises: The pixel value of each pixel point in the smoothed detail feature is added to the pixel point at the same position in the smoothed sinusoidal graph to obtain a corrected image of the target object.

7. The method according to any one of claims 3 to 6, characterized in that: The step of determining the artifact feature based on the pixel value corresponding to each detector channel width in the difference image at each projection angle comprises: For each row of the difference image, calculating the average value of the pixel values ​​of each pixel point in the row; The average value of the pixel values ​​of each pixel point in the row is taken as the pixel value of each pixel point in the row to obtain the artifact feature.

8. The method according to any one of claims 1 to 6, characterized in that: The step of solving the sine graph that makes the preset objective function obtain the minimum value based on the preset objective function and the original sine graph comprises: According to the difference between the smoothed sinusoidal graph represented by the preset objective function and the original sinusoidal graph, the sinusoidal graph is adjusted in an iterative manner so that the preset objective function obtains a minimum value until the preset objective function converges to obtain a smoothed sinusoidal graph.

9. A device for removing CT image artifacts, characterized in that: The device comprises: A sinusoidal image acquisition module, used to acquire an original sinusoidal image obtained by converting an original projection image, wherein the original projection image is an image obtained by performing a CT scan on the target object; A sinusoidal graph smoothing module, used for solving a sinusoidal graph that makes the preset objective function obtain a minimum value based on a preset objective function and the original sinusoidal graph, as a smoothed sinusoidal graph, wherein the preset objective function is used to characterize the error between the smoothed sinusoidal graph and the original sinusoidal graph; A characteristic attribute determination module, used to determine the characteristic attribute included in the original projection image based on the difference between the original sinusoidal graph and the smoothed sinusoidal graph, wherein the characteristic attribute is an artifact included in the original sinusoidal graph or a smoothed detail; An image correction module is used to correct the image to be processed using the characteristic attribute to obtain a corrected image of the target object, wherein when the characteristic attribute is an artifact included in the original sinusoidal graph, the image to be processed is the original sinusoidal graph; when the characteristic attribute is the smoothed details, the image to be processed is the smoothed sinusoidal graph.

10. An electronic device, characterized in that: include: Memory, used to store computer programs; A processor, for implementing any of the methods described in claims 1-8 when executing a program stored in a memory.

Images