CT ring artifact correction method and system based on sample displacement
The sample shifting method combined with geometric correction and median filtering addresses the issue of ring artifacts in CT imaging, improving image quality by converting artifacts to noise and reducing computational complexity.
Patent Information
- Application Number
- CN202510563399.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-04-30
- Publication Date
- 2025-07-15
AI Technical Summary
In the existing CT technology, the annular artifact problem caused by inconsistent pixel response of the detector affects the imaging quality and is difficult to effectively remove.
The projection data processing is performed by sample shifting and combining discrete cosine function law, and the filtered backprojection algorithm and two-dimensional median filtering technology are used to disperse and remove ring artifacts.
Significantly reduce artifact visibility, improve image quality, reduce hardware costs and computing resource requirements, simple operation and wide application range.
Smart Images
Figure CN120318360A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to a method and system for shifting a sample relative to a detector.
[0002] The present invention also relates to the field of image processing technology, and particularly to a method and system for correcting ring artifacts in CT reconstructed images. Background Art
[0003] Computed tomography (CT) technology has been widely used in the fields of clinical diagnosis, industrial inspection, and materials science research due to its non-invasive three-dimensional imaging ability. However, the problem of ring artifacts caused by inconsistent detector pixel responses has long restricted the improvement of imaging quality. These artifacts appear as concentric circular interferences around the rotation axis in the reconstructed image, and their causes include: production processes and performance degradation of imaging components, mechanical assembly errors, ray scattering and strong absorption, data acquisition errors, reconstruction model errors, etc.
[0004] Therefore, there is an urgent need to develop a collaborative solution that combines hardware improvement and image processing to effectively improve CT imaging quality while ensuring the artifact suppression effect and reducing system complexity and computational resource requirements. Summary of the Invention
[0005] The present invention provides a method and system for correcting CT ring artifacts by sample shifting. This method disperses the artifacts in the projection image through sample shifting, thereby weakening the original data mutations in the projection image, and then processes the CT reconstructed image through two-dimensional median filtering to suppress the ring artifacts in the CT reconstructed image.
[0006] The first aspect of the present invention provides a method for correcting CT ring artifacts. The core of this method is to disperse the ring artifacts into noise through a sample shifting scheme and effectively suppress the ring artifacts by combining image post-processing techniques. The steps of the method are as follows:
[0007] Step 1: During the projection process, the sample is horizontally shifted at each same projection angle, and the horizontal shift follows the law of the discrete cosine function.
[0008] Step 2: Obtain the projection data after the sample is shifted through the detector, and geometrically correct the projection data using the known shift parameters to restore it to a sinogram in the non-shifted state.
[0009] Step 3: Use the filtered back-projection (FBP) algorithm for CT image reconstruction to obtain the CT reconstructed image.
[0010] Step 4: Perform two-dimensional median filtering on the CT reconstructed image to obtain the final corrected image.
[0011] The second aspect of the present invention provides a CT ring artifact correction system. The system is as follows:
[0012] The displaceable sample stage module is used to precisely displace the sample and obtain projection data after the displacement.
[0013] The angle module is used to determine the displacement of the sample stage after a fixed projection angle.
[0014] The distance module is used to determine the displacement law when the sample stage is displaced; the displacement law follows the law of discrete cosine function.
[0015] The restoration module is used to restore the shifted projection data to a sinusoidal graph in a non-shifted state.
[0016] The reconstruction module is used to reconstruct the sinusoidal image into a CT image; the reconstruction method used is the filtered back projection algorithm.
[0017] The filtering module is used to perform two-dimensional median filtering on the CT reconstructed image.
[0018] Beneficial effects of the present invention: Compared with the existing removal methods, the advantages of the present invention are as follows: (1) The high-intensity ring artifact is converted into multiple low-intensity artifacts or noise through the discrete cosine shift mode, which significantly reduces the visibility of the artifact. The converted artifact or noise can be effectively removed by processing the CT reconstructed image through two-dimensional median filtering, thereby removing the ring artifact in the image and improving the image quality. (2) The hardware cost is low, and the sample stage can be shifted by configuring a stepper motor. (3) This method has a wide range of applications, does not require complex image processing algorithms, is simple to operate, can remove artifacts while removing noise in the image, and can retain image details as much as possible while improving image quality. BRIEF DESCRIPTION OF THE DRAWINGS
[0019] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.
[0020] Figure 1 : Schematic diagram of sample shift mode of the method provided in an embodiment of the present invention Wherein, the reference numerals are: x-x-axis of the spatial rectangular coordinate system with the sample rotation center as the origin y-y axis of the spatial rectangular coordinate system with the sample rotation center as the origin z-axis of the spatial rectangular coordinate system with the sample rotation center as the origin V-Vector V D-Vector D H - vector H S - vector S
[0021] Figure 2 : Comparison diagram of ring artifact correction results with different shift angles provided by the embodiments of the present invention.
[0022] Figure 3 : Comparison diagram of ring artifact correction results with different shift distances provided by the embodiments of the present invention.
[0023] Figure 4 : Structural schematic diagram of the correction system for ring artifacts in the CT reconstructed image described in the embodiments of the present application Among them, reference numerals: 51 - Shiftable sample stage module 52 - Angle module 53 - Distance module 54 - Reduction module 55 - Reconstruction module 56 - Filter module Specific implementation manners
[0024] In order to enable those skilled in the art to better understand the solution of the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present application.
[0025] The following uses method embodiments to illustrate a CT ring artifact correction method provided by the present application. First, this embodiment uses Figure 1 to illustrate the sample shift mode. In the scanning system, this embodiment establishes a projection model based on vector geometry, as Figure 1 shown. Each projection model is defined by four three - dimensional vectors in a spatial rectangular coordinate system. The spatial position of the X - ray source is represented by S, and the current spatial position of the detector is represented by D. Vectors H and V represent the direction of the detector plane, including tilt and rotation. Therefore, the detector offset is simulated by translating D within the detector plane. Considering the case where the detector undergoes a horizontal displacement s, the projection geometry after displacement can be expressed as follows:
[0026]
[0027]
[0028]
[0029]
[0030] Among them, i represents the current projection angle.
[0031] In the first group of embodiments of the present invention, the sample is shifted through this shift pattern, and the specific steps are as follows:
[0032] Step 1: During the projection process, when the projection angle of the sample is every 1 degree, 2 degrees... 9 degrees, a horizontal shift is performed once, and the shift distance s is 16 pixels; the horizontal shift mode follows the law of discrete cosine function, and the maximum shift distance is s, specifically including displacement amounts such as 2s, s, 0, -s, -2s, etc.
[0033] Step 2: The projection data of the sample after shifting is obtained through the detector, and the projection data is geometrically corrected using the known shift parameters to restore it to a sinogram in the non-shifted state.
[0034] Step 3: The filtered back projection (FBP) algorithm is used for CT image reconstruction to obtain a CT reconstructed image.
[0035] Step 4: The CT reconstructed image is subjected to two-dimensional median filtering to obtain the final corrected image; the size of the median filter kernel is 5*5.
[0036] The final result is as Figure 2 shown, where Figure 2 the first picture of is an image that only contains ring artifacts without any processing.
[0037] In the second group of embodiments of the present invention, the sample is shifted through this shift pattern, and the specific steps are as follows:
[0038] Step 1: During the projection process, when the projection angle of the sample is every 4 degrees, a horizontal shift is performed once, and the shift distances s are 2, 4, 8, 16, and 32 pixels respectively; the horizontal shift mode follows the law of discrete cosine function, and the maximum shift distance is s, specifically including displacement amounts such as 2s, s, 0, -s, -2s, etc.
[0039] Step 2: The projection data of the sample after shifting is obtained through the detector, and the projection data is geometrically corrected using the known shift parameters to restore it to a sinogram in the non-shifted state.
[0040] Step 3: The filtered back projection (FBP) algorithm is used for CT image reconstruction to obtain a CT reconstructed image.
[0041] Step 4: The CT reconstructed image is subjected to two-dimensional median filtering to obtain the final corrected image; the size of the median filter kernel is 5*5.
[0042] The final result is asFigure 3 As shown, among them, the image that only contains circular artifacts without any processing is Figure 2 the first picture of
[0043] Based on the various embodiments of the method for correcting circular artifacts in CT images disclosed above, the present application also discloses a device for correcting circular artifacts in CT images. Referring to the figure, the correction device includes:
[0044] A shiftable sample stage module for shifting the sample to obtain shifted projection data.
[0045] An angle module for controlling the sample stage to be shifted after passing through a fixed projection angle.
[0046] A distance module for determining the shift distance and shift rule when the sample stage is shifted; the shift rule follows the discrete cosine function rule.
[0047] A restoration module for restoring the shifted projection data to a sinogram in an unshifted state.
[0048] A reconstruction module for reconstructing the sinogram into a CT image; the reconstruction method used is the filtered back-projection algorithm.
[0049] A filtering module for performing two-dimensional median filtering on the CT reconstructed image.
Claims
1. A CT annular artifact correction method and system based on sample shifting, characterized in that Comprising the following steps: Step 1: During the projection process, control the sample stage to perform horizontal shifting at fixed projection angle intervals, and the shifting follows the law of discrete cosine function; Step 2: Obtain the projection data after the sample is shifted, and perform geometric correction on the projection data according to the shifting parameters to restore it to a sinogram in the non-shifted state; Step 3: Use the filtered back-projection algorithm to perform CT image reconstruction on the sinogram to obtain a CT reconstructed image; Step 4: Perform two-dimensional median filtering on the CT reconstructed image to generate an image after correcting the annular artifacts.
2. The method according to claim 1, wherein In the said Step 1: The maximum shifting distance corresponding to the law of discrete cosine function is s, and the shifting amounts include at least three of 2s, s, 0, -s, -2s.
3. The method according to claim 1 or 2, characterized in that, In the said Step 1: The horizontal shifting distance s is 16 pixels.
4. The method according to claim 1 or 2, characterized in that, In the said Step 1: The horizontal shifting distance s is selected from one or more combinations of 2, 4, 8, 16, 32 pixels.
5. The method according to claim 1, wherein In the said Step 4: The filtering kernel size of the two-dimensional median filtering is 5×5.
6. The method according to claim 1, characterized in that In the said Step 1: The fixed projection angle interval is an integer angle value within the range of 1 degree to 9 degrees.
7. A CT annular artifact correction system for implementing the method according to any one of claims 1-6, characterized in that, Comprising: A shiftable sample stage module (51) for horizontally shifting the sample according to the law of discrete cosine function; An angle module (52) for controlling the sample stage to trigger shifting at fixed projection angle intervals; A distance module (53) for setting the shifting distance parameter s and the law of discrete cosine function; A restoration module (54) for performing geometric correction on the shifted projection data according to the shifting parameters to generate a sinogram in the non-shifted state; A reconstruction module (55) for performing CT image reconstruction on the sinogram using the filtered back-projection algorithm; A filtering module (56) for performing two-dimensional median filtering on the reconstructed CT image.