Calibration method of ray blocking region, image calibration system and computing device

Abstract

The present disclosure provides a method for calibrating a ray blocking area, an image calibration system and a computing device, and belongs to the technical field of image processing. The method comprises: acquiring a ray projection image of a target blocking area formed by a ray blocking array; determining the geometric parameters of a candidate blocking area in the ray projection image based on the ray projection image; performing noise reduction processing on the candidate blocking area based on the geometric parameters of the candidate blocking area to obtain the target blocking area; and outputting calibration information of the target blocking area, wherein the calibration information comprises the center coordinates of the target blocking area. The present embodiment can accurately calibrate the ray blocking area corresponding to the blocking body.

Description

Technical Field

[0001] This disclosure relates to the field of image processing technology, and in particular to a method for calibrating ray-blocked areas, an image calibration system, and a computing device. Background Technology

[0002] In cone-beam computed tomography (CBCT) imaging, scattered radiation is one of the main factors leading to reduced image contrast and artifacts. To improve image quality, beam stop arrays (BSAs) are typically used for scatter correction. The basic principle is to place a BSA consisting of a regularly arranged array of dots between the X-ray source and the detector. The scattering distribution of the entire projected image is estimated based on the scattering signals in the blocked areas formed by the dot array. The projected image is a two-dimensional perspective image formed by the X-rays emitted from the X-ray source passing through the BSA (and the patient) and being received by the detector. Therefore, before performing scatter correction, the locations of these blocked areas must be accurately identified in the projected image; that is, the beam-blocking areas in the projected image must be calibrated.

[0003] In related technologies, threshold segmentation is used to calibrate ray-occluded areas. However, due to the diversity of BSA tooling installation, the edges of occluded areas may not be clear boundaries, but rather grayscale gradient structures, or there may be significant grayscale differences between different occluded areas in the projected image. Fixed threshold segmentation may misclassify grayscale areas in transition zones as occluded areas, or fail to detect occluded areas with significant grayscale differences from other areas. Summary of the Invention

[0004] This disclosure provides a method, image calibration system, and computing device for calibrating ray-blocking areas; it can accurately calibrate the ray-blocking areas corresponding to the blocking body.

[0005] The technical solution disclosed herein is implemented as follows: In a first aspect, this disclosure provides a method for calibrating a radiation-blocking region, the method comprising: Acquire ray projection images including the target occlusion area formed by the ray blocking array; Based on the ray projection image, determine the geometric parameters of the candidate occlusion region in the ray projection image; Based on the geometric parameters of the candidate occlusion region, noise reduction processing is performed on the candidate occlusion region to obtain the target occlusion region. Output the calibration information of the target occlusion area, including the center coordinates of the target occlusion area.

[0006] In a second aspect, this disclosure provides a computing device including a processor, a memory, and a program or instructions stored in the memory and executable on the processor, wherein the program or instructions, when executed by the processor, implement the steps of the calibration method for the ray shielding region as described in the first aspect.

[0007] Thirdly, this disclosure provides an image calibration system, including a radiation source, a detector, a radiation blocking array, and a computing device as described in the second aspect. The radiation source is configured to emit radiation. The detector, positioned opposite the radiation source, is configured to receive radiation emitted by the radiation source and generate a projection image of the radiation. A ray blocking array, positioned on the ray path between the ray source and the detector, is configured to block part of the ray to form a target blocking area in the ray projection image generated by the detector.

[0008] Fourthly, this disclosure provides a computer-readable storage medium on which a program or instructions are stored, which, when executed by a processor, implement the steps of the calibration method for the ray-blocking region as described in the first aspect.

[0009] Fifthly, this disclosure provides a computer program product, wherein the computer program product includes a computer program or instructions, which, when run on a processor, cause the processor to execute the computer program or instructions to implement the steps of the ray shielding region calibration method as described in the first aspect.

[0010] In a sixth aspect, this disclosure provides a chip including a processor and a communication interface coupled to the processor, the processor being used to run programs or instructions to implement the calibration method for the ray shielding region as described in the first aspect.

[0011] This disclosure provides a method, image calibration system, and computing device for calibrating ray-blocking regions. Since the geometric features of the shading region formed by the blocking body of the beam-projection image are determined by the physical properties of the beam-projection image (regular arrangement, fixed shape, fixed size), it possesses stable geometric rules. In contrast, noise regions (such as electronic noise and beam limiter artifacts) lack these rules. Therefore, the ray-blocking region calibration method of this disclosure filters and distinguishes regions based on geometric parameters, eliminating noise regions whose geometric parameters do not conform to the geometric rules, such as electronic noise that is too small or artifacts with irregular shapes. The final output center coordinates of the target shading region accurately correspond to the projection center of the beam-blocking body of the beam-projection image. Attached Figure Description

[0012] Figure 1 This is a schematic diagram of the structure of an image calibration system provided in this disclosure.

[0013] Figure 2 This is a flowchart illustrating a method for calibrating a radiation shielding area provided in this disclosure.

[0014] Figure 3 This is a flowchart illustrating another method for calibrating the radiation shielding area provided in this disclosure.

[0015] Figure 4 This is a schematic diagram of the noise reduction process provided in this disclosure.

[0016] Figure 5 This is a schematic diagram illustrating the principle of determining the target occlusion region based on area ratio constraints provided in this disclosure.

[0017] Figure 6 This is a schematic diagram of the process for removing interference regions provided in this disclosure.

[0018] Figure 7 A schematic diagram of the original ray projection image provided in this disclosure.

[0019] Figure 8 This is a schematic diagram of a binarized ray projection image provided in this disclosure.

[0020] Figure 9 This is a statistical diagram of the white pixels in the candidate connected regions provided in this disclosure.

[0021] Figure 10 This is a schematic diagram of the ray projection image after removing the interference region provided in this disclosure.

[0022] Figure 11 This is a schematic diagram showing the feature size after being enlarged according to the present disclosure.

[0023] Figure 12 This is a schematic diagram of the process for adjusting the binarization parameters provided in this disclosure.

[0024] Figure 13 This is a structural block diagram of a computing device provided in this disclosure. Detailed Implementation

[0025] The technical solutions in the embodiments of this disclosure will be clearly described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this disclosure. All other embodiments obtained by those skilled in the art based on the embodiments of this disclosure are within the scope of protection of this disclosure.

[0026] In CBCT imaging, to counteract the destructive effect of scattered radiation on image contrast, a background scattering (BSA) is introduced into the imaging path. The BSA is typically a regular lattice made of high atomic number materials such as tungsten or lead, fixed between the radiation source and the detector. The physical shadow it casts in the projected image is the occlusion region. Accurately locating these occlusion regions (e.g., center coordinates) is fundamental for full-field scattering correction.

[0027] refer to Figure 1 This disclosure provides an image calibration system 100, which can be applied to CBCT equipment. The image calibration system 100 mainly includes: an X-ray source 101, a detector 102, a BSA 103, and a computing device 104.

[0028] The X-ray source 101 is used to emit X-rays to pass through the object being scanned or the calibration fixture. In actual medical scenarios, the X-ray source 101 can be an X-ray tube. The X-ray source 101 is mounted on one side of the gantry and can rotate around the object being scanned (such as a patient's head).

[0029] The detector 102 is positioned opposite the X-ray source 101 and mounted on the other side of the rack. It is configured to receive X-rays after they pass through the scanned object and convert the received X-ray signals into digital signals to generate a two-dimensional X-ray projection image.

[0030] Detector 102 can be a flat panel detector (FPD). The ray projection image generated by detector 102 is a grayscale image. The grayscale value of each pixel reflects the intensity of the ray received at that location. The lower the grayscale value, the weaker the ray intensity, indicating that it is more severely blocked or attenuated. The higher the grayscale value, the stronger the ray intensity.

[0031] BSA 103 is positioned on the radiation path between radiation source 101 and detector 102. In this embodiment, BSA 103 can be composed of multiple high-density, high-atomic-number blocking bodies 1031 (which can be blocking points, such as lead beads, lead cylinders, tungsten beads, tungsten cylinders, etc.) arranged according to a preset rule, such as a uniformly distributed rectangular array or hexagonal array.

[0032] When the rays emitted by the X-ray source 101 pass through the BSA 103, each blocker 1031 will block part of the rays, forming a low grayscale shadow area on the ray projection image generated by the detector 102. These shadow areas are the multiple target blocking areas to be marked in this disclosure.

[0033] The shape of the target occlusion area is related to the shape of the obstruction body 1031. For example, the common obstruction body 1031 is cylindrical or spherical, and the corresponding target occlusion area is a circular or nearly circular dark spot in the ray projection image.

[0034] The computing device 104 is communicatively connected to the detector 102, specifically via a wired or wireless connection. The computing device 104 receives the X-ray projection image and executes the calibration method for the X-ray occlusion region provided in this disclosure. The computing device 104 can be hardware with data processing capabilities, such as a Central Processing Unit (CPU), Digital Signal Processor (DSP), Field Programmable Gate Array (FPGA), or Programmable Logic Array (PLA), or a chip integrating multiple of these processors. In actual use, due to the gravity descent of the robotic arm, minor vibrations of the rotating shaft, or deformation caused by temperature changes, the projection position of the BSA 103 on the detector 102 is not absolutely fixed. Therefore, the computing device 104 needs to perform real-time or near-real-time automatic calibration on the acquired X-ray projection image to obtain the accurate center coordinates of the target occlusion region in the current state.

[0035] Please see Figure 2 , Figure 2 This is a flowchart illustrating a method for calibrating a radiation shielding region according to an embodiment of this disclosure. This method can be executed by the aforementioned computing device 104. Figure 2 As shown, the method includes the following steps S201 to S204.

[0036] In step S201, a ray projection image including the target occlusion area formed by the ray blocking array is acquired.

[0037] In this embodiment of the disclosure, a ray projection image refers to a two-dimensional perspective image formed on the photosensitive plane of the detector after a ray beam passes through an object. For example, when imaging a human skeleton, the single-frame planar grayscale image received by the detector is the ray projection image.

[0038] The target occlusion region refers to the specific shadow area formed by the blocking body of the BSA in the ray projection image. For example, if the blocking body is a cylindrical or spherical dot matrix, the corresponding target occlusion region in the ray projection image is a series of regularly arranged near-circular dark spots. The target occlusion region is the shadow formed by the actual blocking body.

[0039] In step S202, the geometric parameters of the candidate occlusion regions in the ray projection image are determined based on the ray projection image.

[0040] Understandably, a candidate occlusion region refers to an independent connected region in a ray projection image that is preliminarily located as a potential ray occlusion region through common image processing methods such as conventional image segmentation, edge detection, and connected component extraction. The number of candidate occlusion regions is usually greater than or equal to the number of target occlusion regions. The geometric parameters of a candidate occlusion region refer to quantitative parameters that characterize the geometric shape, spatial location, and shape features of the region. These parameters may include, but are not limited to, at least one of the following: region contour size, coverage area, centroid location, center coordinates, boundary coordinates, shape regularity, circumscribed geometric feature parameters, and equivalent feature size. For example, if the candidate occlusion region is a circle, the geometric parameters may include the center coordinates, circumscribed radius, connected component area, and perimeter.

[0041] In practical implementation, the computing device first preprocesses the ray projection image. Preprocessing may include image binarization, converting a grayscale image into a black-and-white image. For determining candidate occlusion regions, some feasible methods employ contour detection. Contour detection refers to the process of extracting continuous boundaries with the same color or intensity in the ray projection image, identifying the closed region enclosed by these continuous boundaries as the candidate occlusion region. For near-circular target occlusion regions, other feasible methods may employ Hough circle detection. Detected Hough circles are identified as candidate occlusion regions. Hough circle detection is a feature extraction technique that uses mathematical transformations to find circular trajectories in image space. By accumulating votes in the parameter space, it can more robustly locate circular shadows.

[0042] In step S203, the candidate occlusion region is denoised based on its geometric parameters to obtain the target occlusion region.

[0043] Candidate occlusion regions typically include both the target occlusion region actually formed by the ray blocking array and invalid occlusion regions caused by factors such as noise, interference, image artifacts, or edge truncation. The noise reduction processing in this embodiment refers to the process of identifying and eliminating invalid occlusion regions through logical rules, thereby extracting the target occlusion region. Noise may include edge artifacts of the beam limiter, electronic interference noise, or structural ghosting of the workpiece itself.

[0044] Specifically, the geometric compliance conditions that the target occlusion area should meet can be determined in advance based on the prior design information of the ray blocking array; then, based on the geometric parameters corresponding to each candidate occlusion area, the candidate occlusion area is checked one by one to see if it meets the above geometric compliance conditions; the candidate occlusion area that meets the geometric compliance conditions is determined as the target occlusion area, and the candidate occlusion area that does not meet the geometric compliance conditions is determined as a noise invalid occlusion area and is removed.

[0045] For example, if the blocker of a BSA is circular, the blocker will form a near-circular or circular dark spot in the ray projection image. However, random noise or mechanical artifacts caused by the beam limiter in the ray projection image are usually messy and have no fixed shape. Therefore, the determined geometric compliance condition is that the circularity is greater than or equal to the circularity threshold. Candidate occlusion areas with a circularity less than the circularity threshold in the target occlusion area are removed.

[0046] In step S204, the calibration information of the target occlusion area is output.

[0047] The calibration information describes the geometric information of the target occlusion area. This calibration information may include the center coordinates of the target occlusion area. The center coordinates refer to the geometric center of the target occlusion area and are used to accurately locate the core position of the target occlusion area on the detector's imaging plane. For example, if the target occlusion area is circular, the center coordinates are the coordinates of the center of the circle.

[0048] In some embodiments, the calibration information may further include a template image containing the target occlusion region. A template image refers to a digital mask that logically classifies or masks image pixels in a spatial dimension. Template images are often simply referred to as masks. For example, a template image is a binary bitmap with the same dimensions as the original image. In this binary bitmap, pixel positions corresponding to the target occlusion region are marked with high-brightness values, while pixel positions in other regions are marked with zero values.

[0049] The center coordinates provide accurate positioning for scattering signal extraction, and the template image intuitively shows the range of the target occlusion area, which facilitates manual verification and monitoring by operators, thereby improving the operability and reliability of the calibration method.

[0050] The geometric features of the occlusion region formed by the blocker of the BSA in the ray projection image are determined by the physical properties of the BSA (regular arrangement, fixed shape, fixed size), and have stable geometric rules. However, noise regions (such as electronic noise and beam limiter artifacts) do not have such rules. Therefore, this embodiment of the present disclosure filters and distinguishes by geometric parameters, and eliminates noise regions whose geometric parameters do not conform to the geometric rules, such as electronic noise that is too small in size and artifacts with irregular shapes. Finally, the center coordinates of the target occlusion region can accurately correspond to the projection center of the blocker of the BSA.

[0051] In some embodiments, geometric parameters include: dimensions and center coordinates; such as Figure 3 As shown, the method for calibrating the radiation shielding area provided in this disclosure includes the following steps S301 to S304.

[0052] In step S301, a ray projection image including the target occlusion area formed by the ray blocking array is acquired.

[0053] In step S302, the geometric parameters of the candidate occlusion regions in the ray projection image are determined based on the ray projection image.

[0054] After acquiring the ray projection image, the computing device first determines the candidate occlusion regions. For example, in the binarized ray projection image, the candidate occlusion regions are discrete white connected regions. To describe the candidate occlusion regions, the computing device needs to determine their geometric parameters. In this embodiment, the geometric parameters include size and center coordinates. Size is a spatial measure of the candidate occlusion region, and center coordinates are the geometric center of the candidate occlusion region in the two-dimensional image coordinate system. For example, if the candidate occlusion region is circular, the size can be its radius or diameter, and the center coordinates are the center of the circle.

[0055] In step S303, based on the geometric parameters of the candidate occlusion region, the candidate occlusion region that meets the constraint conditions is determined as the target occlusion region.

[0056] The size of the candidate occlusion region can be an equivalent size. The area of ​​the candidate occlusion region can be determined by the effective pixel statistics method. For example, if the candidate occlusion region is equivalent to a circle and the size corresponds to the radius, the radius can be determined based on the area of ​​the candidate occlusion region obtained by statistics.

[0057] Constraints refer to the logical criteria used to determine whether a candidate occlusion region is formed by an obstruction. In this embodiment, the constraints include at least one of area ratio constraints, size threshold constraints, and boundary position constraints.

[0058] Specifically, such as Figure 4 As shown, the noise reduction process can be implemented in at least one of the following ways: steps S401 and S402, or steps S403 and S404, or step S405.

[0059] In step S401, in response to the constraint condition including the area ratio constraint, the ratio of the area of ​​the candidate occlusion region to the area of ​​the corresponding standard occlusion region is determined based on the size of the candidate occlusion region and the size of the corresponding standard occlusion region.

[0060] In this embodiment, the standard occlusion region refers to the standard shape of the minimum envelope candidate occlusion region. The standard shape corresponds to the shape of the shadow actually formed by the obstruction in the ray projection image. For example, if the shadow actually formed by the obstruction in the ray projection image is circular, then the standard shape is determined to be circular, and the minimum circumcircle of the candidate occlusion region is determined as the standard occlusion region corresponding to that candidate occlusion region; if the shadow actually formed by the obstruction in the ray projection image is rectangular, then the standard shape is determined to be rectangular, and the minimum circumcircle of the candidate occlusion region is determined as the standard occlusion region corresponding to that candidate occlusion region.

[0061] In step S402, candidate occlusion areas whose area ratio is greater than or equal to a preset ratio are determined as target occlusion areas.

[0062] If the ratio of the area of ​​the candidate occluded region to the area of ​​the corresponding standard occluded region is greater than or equal to a preset ratio, it can be indirectly indicated that the shape of the candidate occluded region is very similar to that of the corresponding standard occluded region, and it can be identified as the target occluded region. If the ratio of the area of ​​the candidate occluded region to the area of ​​the corresponding standard occluded region is less than a preset ratio, it can be indicated that the shape of the candidate occluded region is significantly different from that of the corresponding standard occluded region, and it fails to completely fill the corresponding standard occluded region, and is identified as noise.

[0063] Area ratio constraints can effectively eliminate irregularly shaped noise, such as long strip-shaped beam limiter artifacts and irregular clusters of electronic noise. The area ratio of these noises is usually much smaller than the preset ratio, and this constraint can quickly eliminate them, improving the accuracy of target occlusion area recognition.

[0064] For example, such as Figure 5 As shown, taking a circular standard occlusion area as an example, the ratio of the areas is also called the circularity, with a preset ratio of 0.3. The areas enclosed by the solid lines in the figure represent the determined candidate occlusion areas 501 and 502. The dashed circle surrounding candidate occlusion area 501 represents the corresponding standard occlusion area 503. The ratio of the area of ​​candidate occlusion area 501 to the area of ​​the corresponding standard occlusion area 503 is 0.2, which is less than the preset ratio of 0.3, and is therefore identified as noise. The dashed circle surrounding candidate occlusion area 502 represents the corresponding standard occlusion area 504. The ratio of the area of ​​candidate occlusion area 502 to the area of ​​the corresponding standard occlusion area 504 is 0.9, which is greater than the preset ratio of 0.3, and is therefore identified as the target occlusion area.

[0065] In step S403, in response to constraints including size threshold constraints, a dynamic size threshold is determined based on the feature size of the standard occlusion region corresponding to all candidate occlusion regions.

[0066] In step S404, the candidate occlusion regions corresponding to the standard occlusion regions with feature sizes greater than the dynamic size threshold are determined as the target occlusion regions.

[0067] Since the actual BSA blocks are of uniform size, the size of the target occlusion area is relatively uniform; while the size of noise (such as tiny electronic noise) is much smaller than the actual target occlusion area, it can be removed by setting a dynamic size threshold.

[0068] Size threshold constraints refer to filtering criteria based on a range of physical dimensions. The computing device determines a dynamic size threshold based on the feature dimensions of the standard occlusion region corresponding to all candidate occlusion regions. Feature dimensions are indicators representing the size of an object. For example, if the standard occlusion region is a circle, the corresponding feature dimension could be the radius. The dynamic size threshold refers to a dynamic decision boundary calculated in real-time based on all feature dimensions of the current image.

[0069] Specifically, the computing device first obtains the feature dimensions of the standard occlusion regions corresponding to all candidate occlusion regions; it then calculates the statistical values ​​of all feature dimensions corresponding to all standard occlusion regions. These statistical values ​​can be the mean, median, or mode, as the mean reflects the overall level; this disclosure uses the mean as an example. Based on the statistical values, a dynamic size threshold T is determined. The formula for calculating the dynamic size threshold is as follows: Where α is an adjustment coefficient, greater than 0 and less than or equal to 1, and its specific value range can be adjusted according to the imaging noise level. The value is the average of all feature sizes. It is determined whether the feature size of the standard occlusion region corresponding to each candidate occlusion region is greater than the dynamic size threshold. If it is, the size of the candidate occlusion region matches the size of the standard occlusion region and is retained as the target occlusion region; otherwise, the candidate occlusion region is considered to be minor noise (such as a small connected region formed by electronic noise) and is discarded.

[0070] Size threshold constraints can effectively eliminate minute noise interference, whose characteristic size is usually much smaller than that of the standard occlusion area. Dynamic size thresholds can accurately filter these noises. At the same time, the dynamic size threshold setting method allows the constraint to adapt to different working conditions, avoiding the limitations of fixed thresholds and further improving the adaptability of the calibration method.

[0071] In step S405, in response to the constraints including boundary position constraints, the candidate occlusion region corresponding to the standard occlusion region with a complete shape is determined as the target occlusion region based on the center coordinates, feature size, average feature size and boundary coordinates of the standard occlusion region corresponding to the candidate occlusion region and the ray projection image.

[0072] Boundary position constraints refer to the criteria for handling incomplete regions at the edges of ray-projected images. The computing device identifies whether a standard occlusion region is complete based on the center coordinates, feature size, and average feature size of the standard occlusion region corresponding to the candidate occlusion region, as well as the boundary coordinates of the ray-projected image. The boundary coordinates of the ray-projected image refer to the coordinates of the outermost pixel of the detector's effective imaging area. For example, the left boundary of the image is X=0 or the right boundary is X=W, where W is the width of the image. The computing device identifies candidate occlusion regions corresponding to standard occlusion regions with complete shapes as target occlusion regions.

[0073] The specific logic is as follows: if the center coordinates plus the feature size exceed the image boundary, or if the feature size of the standard occluded region is significantly smaller than the average feature size, then it is determined to be an incomplete shape. Boundary position constraints can accurately identify and eliminate incomplete occluded regions and boundary artifacts near the image boundary. The scattering signals from incomplete regions are not representative, therefore their presence affects the accuracy of scattering signal extraction. This constraint ensures that all target occluded regions are complete in shape and located within the image, providing a reliable sample area for subsequent scattering correction.

[0074] Specifically, the standard occlusion area is illustrated using a circle. The center coordinates of the standard occlusion area corresponding to the candidate occlusion area are obtained. , radius r_c0, average radius r_avg of all standard occlusion regions; obtain the boundary coordinates of the ray projection image, including left boundary x_left (usually 0 pixels), right boundary x_right (image width, e.g., 1024 pixels), top boundary y_top (usually 0 pixels), and bottom boundary y_bottom (image height, e.g., 1024 pixels); determine whether the standard occlusion region is a complete shape. The criteria for a complete shape include: left boundary constraint, ensuring that the left side of the occlusion region does not exceed the image boundary and maintains a certain distance from the boundary. The discriminant is: Right boundary constraint ensures that the right side of the occluded area does not extend beyond the image boundary; the discriminant is... Upper boundary constraint ensures that the upper side of the occluded area does not exceed the image boundary; the discriminant is... The lower boundary constraint ensures that the lower side of the occluded area does not exceed the image boundary; the discriminant is... The specific discriminant may be modified based on the origin setting in the image coordinate system, but this disclosure does not impose any limitations.

[0075] For example, the center coordinates of a standard occlusion region are x_c0 = 5, y_c0 = 300, radius r_c0 = 4 pixels, and r_avg = 10 pixels. Since... The left boundary constraint is not met. Therefore, the corresponding candidate occlusion region is determined to be close to the left boundary of the image, and its shape may be incomplete, so it is discarded. If the radius r_c0 is 5 pixels, since... The candidate occlusion region is determined to satisfy the left boundary constraint. The center coordinates of a standard occlusion region are x_c0 = 1020, y_c0 = 300, radius r_c0 = 4 pixels, and r_avg = 10 pixels. Since... The right boundary constraint is not met. Therefore, the corresponding candidate occlusion region is determined to be close to the left boundary of the image. Its shape may be incomplete, so it is discarded. The upper boundary constraint is similar to the lower boundary constraint, and will not be described in detail here.

[0076] In step S304, the calibration information of the target occlusion area is output.

[0077] This disclosure addresses complex background interference by introducing multi-dimensional constraints. Taking a circular standard occlusion area as an example, firstly, the area ratio constraint utilizes the physical nature of the obstruction being circular. Noise in non-target occlusion areas (such as mechanical scratches and artifacts at the edges of the limiter) typically presents as elongated or irregular shapes. By calculating the circularity ratio and comparing it with a preset ratio, the circular target occlusion area and irregular noise are logically distinguished. Secondly, the dynamic size threshold constraint considers the magnification fluctuations of the imaging system. Since the distance between the X-ray source focus and the detector may vary slightly, the projected size of the obstruction will fluctuate accordingly. By calculating the average feature size to determine the dynamic threshold, small noise points with non-compliant sizes can be accurately eliminated, avoiding the insufficient robustness of fixed thresholds under different operating conditions. Finally, the boundary position constraint solves the accuracy reduction caused by edge distortion. Obstructions located at the edge of the image often only partially develop. If the center coordinates are forcibly calculated, a large positioning error will inevitably occur due to the physical offset of the pixel centroid. By logically comparing the boundary coordinates with the feature size, these incomplete areas that cannot guarantee accuracy are automatically eliminated. In summary, this solution establishes a rigorous physical discrimination characterization from three dimensions: area, size, and spatial boundary. It can not only effectively suppress various artifacts in complex backgrounds, but also ensure stability under various installation errors through dynamic optimization.

[0078] In actual operation, CBCT equipment can switch between different imaging modes according to clinical needs. Imaging modes include full-field imaging and half-field imaging. Full-field imaging means that the X-ray cone beam covers the entire effective area of ​​the detector. Half-field imaging means that the X-ray cone beam covers only a portion of the effective area of ​​the detector.

[0079] When the X-ray projection image is in half-field imaging mode, the collimator obstructs one side of the detector, resulting in large areas of mechanical occlusion shadows at the edges of the X-ray projection image. The shape of these mechanical occlusion shadows is related to the position and shape of the machinery, and the shape of the corresponding interference regions can be predetermined, i.e., a pre-defined shape, such as a triangular region. These interference regions may adhere to candidate occlusion regions, affecting the accuracy of geometric parameter calculation and noise reduction processing. Therefore, before determining the geometric parameters of candidate occlusion regions, it is necessary to identify and remove these pre-defined interference regions.

[0080] Based on this, in some embodiments, such as Figure 6 As shown, in the calibration method for the ray blocking region, the process of eliminating interference regions before determining the geometric parameters of the candidate blocking region includes the following steps S601 and S602.

[0081] In step S601, interference regions of a preset shape are identified in the ray projection image.

[0082] The computing device first determines whether the imaging mode of the X-ray projection image is full-field imaging or half-field imaging. The determination can be based on the scanning parameters of the CBCT device, such as whether the scanning mode is set to half-field or full-field.

[0083] Interference regions refer to the set of unwanted pixels in a ray-projected image generated by non-obstructing objects that can obscure or confuse the target's occlusion area. The preset shape is determined in advance based on the mechanical design features of the system tooling. In practical applications, the preset shape is typically triangular.

[0084] In this embodiment of the disclosure, the computing device first identifies interference regions of a preset shape in the ray projection image, and then removes the interference regions of the preset shape from the ray projection image. Removal refers to setting the pixel values ​​of a specific region to background values ​​through image processing methods, such as setting the grayscale values ​​of all pixels in the interference region to 0.

[0085] The process of identifying interference regions of a preset shape is as follows. First, based on at least one of spatial location constraints, size constraints, and boundary consistency constraints, candidate connected regions are identified in the ray projection image. A candidate connected region refers to a set of pixels with the same pixel value and spatially adjacent to each other in a binarized image.

[0086] When the imaging mode is half-field imaging, the acquired ray projection image is denoted as the half-field image. Spatial position constraint refers to determining the search area where the interference area may exist from the half-field image based on the prior position information of the interference area in the half-field image. Prior position information refers to the coordinate distribution pattern known in advance based on the mechanical installation position of the limiter. The search area refers to the local pixel range defined by the computing device specifically for retrieving the interference area. For example, defining the search area within 10% of the width on the left side of the half-field image, and limiting the vertical coordinate to 30% of the top or bottom of the half-field image, thereby narrowing the search range and improving the accuracy of interference area localization.

[0087] The computing device then performs size constraints, which involves using the size of the connected regions and a set size threshold to determine candidate interference regions that may be interfering. Specifically, the computing device identifies connected regions within the search area whose width is greater than or equal to a minimum width threshold as candidate interference regions. The minimum width threshold is a preset length indicator used to distinguish between minute noise points and large-scale occlusion regions. Candidate interference regions refer to areas initially filtered by location and size that are suspected of being occluded by the clamp limiter. This eliminates excessively narrow or low minute connected regions, avoiding false detections.

[0088] Next, the computing device performs boundary consistency constraint determination. The computing device identifies candidate regions whose boundary geometric difference values ​​are less than a consistency threshold as candidate connected regions. The geometric difference value is a numerical difference describing the degree of alignment between two edge contours in a specific direction. The consistency threshold refers to the upper limit of allowable mechanical installation errors.

[0089] For example, the preset shape is a triangle-like region, and the right boundaries of two triangle-like regions (such as the upper left and lower left triangle-like regions) are close to each other. A right boundary difference threshold of 20 pixels is set. If the difference in the x-coordinates of the right boundaries of two candidate interference regions is less than or equal to 20 pixels, the boundary consistency constraint is satisfied, and the region is determined to be a candidate connected region; if the difference is greater than 20 pixels, the constraint is not satisfied, and the region is excluded.

[0090] Furthermore, the computing device needs to address the adhesion problem between candidate occlusion regions and interference regions. Based on the edge gradient features of candidate connected regions, the computing device determines the abrupt connection points of the candidate connected regions; these abrupt connection points are the adhesion locations. An abrupt connection point refers to the coordinate point where the shadow of the obstruction merges with the edge of the triangular-like shape.

[0091] Determining abrupt connection points can be achieved by scanning the pixel distribution within candidate connected regions row by row or column by column. Pixel distribution refers to the arrangement of the number or density of white pixels (i.e., signal points) on a specific scan line segment.

[0092] The computing device then detects the trend of pixel distribution along the scanning direction. When a local bulge exists that violates the monotonicity of the trend, the location of the local bulge is identified as abrupt connection. The trend refers to the direction of evolution of the pixel width value as the row or column number changes. Monotonicity refers to the mathematical property that a numerical sequence always increases or decreases. For example, the white pixel distribution in a triangular region exhibits a decreasing trend between rows from top to bottom or bottom to top. For the upper left triangular region, the number of white pixels in each row gradually decreases from top to bottom (because the area of ​​the triangular region gradually shrinks), while in the part adhering to the candidate occlusion area, the number of white pixels in each row suddenly increases, forming a local bulge. The location of this local bulge is the abrupt connection.

[0093] Finally, the computing device segments the candidate connected regions at the abrupt connection points to obtain the interference region and the candidate occlusion region.

[0094] In step S602, interference regions of a preset shape are removed from the ray projection image.

[0095] For example, such as Figure 7 As shown, this is the original ray projection image acquired. The grid fill indicates that the background color is not a pure, single color, and the black dots correspond to candidate occlusion areas. Figure 8 The image shown is the ray projection image obtained after binarization. The white triangular regions in the image represent interference regions. Candidate connected regions 801 and 802 are regions that may have adhesion issues. Candidate occlusion regions p1 and p2 are adhered to the interference regions. The number of white pixels in candidate connected region 801 is counted, as shown below. Figure 9 As shown in the graph, the horizontal axis represents the row number, and the vertical axis represents the number of white pixels. The graph shows that starting from row 50, the number of white pixels decreases as the row number increases. However, a local abrupt change occurs around row 55, which is identified as the abrupt connection point. After removing the interference regions in candidate connected regions 801 and 802, the resulting ray projection image containing only the candidate occluded regions is shown below. Figure 10 As shown.

[0096] The technical solution of this disclosure can not only identify and eliminate large-area complex artifacts, but also peel out the occlusion area corresponding to the effective blocking body that is stuck in the artifacts, ensuring the accurate output of calibration information in complex interference environment.

[0097] Furthermore, the calibration accuracy of the ray occlusion area directly affects the quality of scattering correction. During actual imaging, slight shifts in the focal point of the ray source or minor mechanical vibrations in the BSA can cause pixel-level deviations between the actual shadow position and the calibration position in the ray projection image. If the calibrated occlusion range is exactly equal to the original size of the physical shadow, any slight offset will cause the sampling point to slip out of the shadow area. Since the intensity of rays outside the shadow area is much higher than the intensity of scattered rays inside the shadow area, any leakage of high-intensity rays into the scattering sampling area will lead to a significant deviation in the estimated scattering distribution. Therefore, in some embodiments, before outputting calibration information, it is necessary to enlarge the feature size based on the center coordinates and feature size of the standard occlusion area corresponding to the target occlusion area, using a preset error tolerance, to obtain the updated feature size.

[0098] The error tolerance is a pre-set size magnification ratio to account for factors such as BSA installation error, imaging noise, and geometric parameter calculation error. The specific tolerance can be adjusted according to calibration accuracy requirements and the actual error level. For example, when calibration accuracy requirements are high and the actual error is small, the error tolerance can be set to 5%; when the error is large, the error tolerance can be set to 15%. The error tolerance can be determined experimentally. For example, by obtaining the actual size of the target occlusion area through multiple manual calibrations and comparing it with the feature size obtained from automatic calibration, the error ratio can be calculated, and a reasonable error tolerance can be set.

[0099] The updated feature size is the current feature size plus the product of the current feature size and the error tolerance. For example, combining... Figure 10 ,like Figure 11 As shown, the target occlusion area is circular. After the radius is enlarged, the center remains unchanged. This is a schematic diagram of the enlarged target occlusion area.

[0100] The technical solution of this embodiment creates a safety buffer by performing amplification processing on the feature size. After the update, the feature size will be slightly larger than the physical shadow formed by the real obstruction. Even if unexpected mechanical displacement or focus drift occurs during operation, the sampling point of the scattered signal can still be locked within the occlusion range due to the existence of the safety buffer.

[0101] Due to factors such as BSA installation errors, tooling deformation caused by temperature changes, and imaging noise fluctuations, the initial binarization parameters used when binarizing the acquired raw ray projection image may not accurately distinguish between occluded areas and the background (non-occluded areas). This results in the spatial distribution of the final target occluded areas not conforming to the expected arrangement rules, such as insufficient number of occluded areas, the presence of redundant interference areas, or irregular arrangement. Therefore, this embodiment adjusts the binarization parameters through a feedback adjustment mechanism, such as... Figure 12 As shown, in the calibration method for the ray-blocking area, the adjustment process of the binarization parameter includes steps S1201 and S1202.

[0102] In step S1201, based on the prior layout information of BSA, when the spatial distribution of the verification target occlusion area does not conform to the expected layout rules, the binarization parameters are adjusted.

[0103] In step S1202, the ray projection image is updated based on the adjusted binarization parameters.

[0104] The prior layout information of the BSA refers to the preset layout rules of the BSA's barriers, including the number of barriers, the layout method (such as uniform rectangular array, hexagonal array), row spacing, column spacing, etc. For example, the BSA is a 10×10 uniform rectangular array with 100 barriers and a row spacing and column spacing of 50mm.

[0105] The prior arrangement information also includes the expected arrangement rules of the occluded areas, such as odd columns being consistent with odd columns, that is, the center coordinates of the obstructions in the odd columns are the same in the y direction, even columns being consistent with even columns, that is, the center coordinates of the obstructions in the even columns are the same in the y direction, and the pixel values ​​of the row spacing and column spacing in the ray projection image.

[0106] The computing device verifies the spatial distribution of the target occlusion areas based on prior layout information. Verification may include: quantity verification, determining whether the number of target occlusion areas matches the number of obstructions in the prior layout information. For example, if the BSA is a 10×10 array with 100 obstructions, the number of target occlusion areas should be close to 100 (within the allowable error range, as incomplete areas at the edges may be removed). If the number of target occlusion areas is much less than 100, it indicates missed detections; if it is much greater than 100, it indicates false detections.

[0107] The layout rule verification determines whether the layout of the target occlusion area conforms to the expected layout rule. For example, in a uniform rectangular array BSA, the row spacing (the difference in the y-direction coordinates of the center coordinates of two adjacent rows) of the target occlusion area should be close to the expected row spacing (the pixel value calculated based on prior layout information), and the column spacing (the difference in the x-direction coordinates of the center coordinates of two adjacent columns) should be close to the expected column spacing. The center coordinates of the target occlusion area in odd-numbered columns should be consistent in the y-direction, and the same applies to even-numbered columns. If the row spacing of two rows is much larger than the expected row spacing, it indicates that there is a missed detection; if the center coordinates of a column deviate too much from the expected position, it indicates that there is a calibration error.

[0108] For example, the BSA is a 3×3 uniform rectangular array with 9 obstructions. The expected row and column spacing is 50 pixels. The expected center coordinates in the y-direction of the odd-numbered columns (columns 1 and 3) are 300 pixels, and the expected center coordinates in the y-direction of the even-numbered column (column 2) are also 300 pixels. During verification, it was found that there were 8 target occlusion areas (1 missing), and the center coordinates in the y-direction of the 1st and 3rd columns were 300 pixels and 305 pixels respectively (a difference of 5 pixels, exceeding the allowable error of 3 pixels), indicating that the spatial distribution does not conform to the expected arrangement rules.

[0109] If the verification result shows that the spatial distribution of the target occlusion area does not conform to the expected arrangement rule, the binarization parameter (grayscale threshold) is adjusted, and the ray projection image is updated based on the adjusted binarization parameter (re-binarization processing is performed). Then, the process returns to determine the geometric parameters of the candidate occlusion area and subsequent steps until the spatial distribution conforms to the expected arrangement rule.

[0110] The specific rules for adjusting the binarization parameters can be as follows: If the number of target occluded areas is insufficient (missed detections): This indicates that the initial binarization threshold is set too high, causing the grayscale values ​​of some real occluded areas to be higher than the threshold and misclassified as background pixels. In this case, the binarization threshold should be lowered. For example, if the initial threshold is 80, it can be adjusted to 70, so that more pixels with lower grayscale values ​​are identified as foreground pixels, thereby identifying more candidate occluded areas.

[0111] If there are too many target occlusion areas (false detections): This indicates that the initial binarization threshold is set too low, causing some background noise to be misidentified as foreground pixels, forming false candidate occlusion areas. In this case, the binarization threshold should be increased, for example, if the initial threshold is 80, adjust it to 90 to remove low grayscale noise pixels.

[0112] If the arrangement of the target occlusion area does not conform to expectations (e.g., the center coordinates deviate), it indicates that the initial binarization threshold caused some edge pixels in the occlusion area to be misjudged, affecting the accuracy of geometric parameter calculation. In this case, the threshold can be adjusted according to the direction of deviation. For example, if the center coordinates are generally biased to the left, it may be that the pixels on the left edge are misjudged as background. The threshold can be appropriately reduced to include the pixels on the left edge in the occlusion area.

[0113] This embodiment verifies the spatial distribution of the target occlusion area through prior arrangement information and dynamically adjusts the binarization parameters based on the verification results. This allows for timely detection and correction of missed or false detections caused by improper binarization parameter settings, ensuring that the number and arrangement of the target occlusion area conform to the actual situation. Consequently, the calibration method for the ray occlusion area can adapt to various complex working conditions such as BSA installation errors, tooling deformation, and imaging noise fluctuations. Through parameter adjustment, the calibration method maintains a high level of accuracy.

[0114] See Figure 13 This illustrates a structural block diagram of a computing device provided in an exemplary embodiment of the present disclosure. In some examples, Figure 13 The computing device shown can be implemented as Figure 1 The image calibration system 100 shown includes a computing device 104. In some examples, the computing device 104 can receive the X-ray projection image transmitted by the detector 102 based on the accessed wired or wireless network. It is understood that the computing device 104 undertakes the calculation and processing work of the technical solution of this disclosure, and this disclosure does not limit it.

[0115] like Figure 13 As shown, the computing device in this disclosure may include one or more of the following components: processor 1310 and memory 1320.

[0116] Optionally, the processor 1310 connects various parts within the computing device using various interfaces and lines, and performs various functions and processes data by running or executing instructions, programs, code sets, or instruction sets stored in the memory 1320, and by calling data stored in the memory 1320. Optionally, the processor 1310 can be implemented using at least one hardware form of Digital Signal Processing (DSP), Field-Programmable Gate Array (FPGA), or Programmable Logic Array (PLA). The processor 1310 can integrate one or a combination of several of the following: Central Processing Unit (CPU), Graphics Processing Unit (GPU), Neural-network Processing Unit (NPU), and baseband chip. Specifically, the CPU primarily handles the operating system, user interface, and applications; the GPU is responsible for rendering and drawing the content required to be displayed on the touch screen; the NPU is used to implement Artificial Intelligence (AI) functions; and the baseband chip is used to handle wireless communication. It is understandable that the aforementioned baseband chip may not be integrated into the processor 1310, but may be implemented using a separate chip.

[0117] The memory 1320 may include random access memory (RAM) or read-only memory (ROM). Optionally, the memory 1320 may include a non-transitory computer-readable storage medium. The memory 1320 may be used to store instructions, programs, code, code sets, or instruction sets. The memory 1320 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as touch function, sound playback function, image playback function, etc.), instructions for implementing the various method embodiments described above, etc.; the data storage area may store data created according to the use of the computing device, etc.

[0118] In addition, those skilled in the art will understand that the structure of the computing device 104 shown in the above figures does not constitute a limitation on the computing device 104. The computing device 104 may include more or fewer components than shown, or combine certain components, or have different component arrangements. For example, the computing device 104 may also include a display screen, a camera assembly, a microphone, a speaker, radio frequency circuitry, an input unit, sensors (such as accelerometers, angular velocity sensors, light sensors, etc.), audio circuitry, a WiFi module, a power supply, a Bluetooth module, and other components, which will not be described in detail here.

[0119] This disclosure also provides a computer-readable storage medium storing at least one instruction, which is executed by a processor to implement the ray shielding region calibration method as described in the above embodiments.

[0120] This disclosure also provides a computer program product including computer instructions stored in a computer-readable storage medium; a processor of an electronic device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the electronic device to perform the calibration method for the radiation shielding area described in the above embodiments.

[0121] This disclosure also provides a chip, which includes a processor and a communication interface. The communication interface is coupled to the processor. The processor is used to run programs or instructions to implement the various processes of the above-described method embodiment for calibrating the ray-shielding area, and can achieve the same technical effect. To avoid repetition, it will not be described again here.

[0122] It should be understood that the chip mentioned in the embodiments of this disclosure may also be referred to as a system-on-a-chip, system chip, chip system, or system-on-a-chip, etc.

[0123] In the several embodiments provided in this disclosure, it should be understood that the disclosed systems, apparatuses, servers, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection between apparatuses or units through some interfaces, and may be electrical, mechanical, or other forms.

[0124] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0125] Furthermore, the functional units in the various embodiments of this disclosure can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0126] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this disclosure, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this disclosure. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, ROM, RAM, magnetic disks, or optical disks.

[0127] Those skilled in the art will recognize that the functions described in this disclosure in one or more of the examples above can be implemented using hardware, software, firmware, or any combination thereof. When implemented in software, these functions can be stored in a computer-readable medium or transmitted as one or more instructions or code on a computer-readable medium. Computer-readable media include computer storage media and communication media, wherein communication media include any medium that facilitates the transfer of a computer program from one place to another. Storage media can be any available medium accessible to a general-purpose or special-purpose computer.

[0128] It should be noted that the technical solutions described in this disclosure can be combined arbitrarily as long as they do not conflict.

[0129] The above description is merely a specific embodiment of this disclosure, but the scope of protection of this disclosure is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this disclosure should be included within the scope of protection of this disclosure.

Claims

1. A method for calibrating a radiation shielding area, characterized in that, The method includes: Acquire ray projection images including the target occlusion area formed by the ray blocking array; Based on the ray projection image, determine the geometric parameters of the candidate occlusion region in the ray projection image; Based on the geometric parameters of the candidate occlusion region, the candidate occlusion region is subjected to noise reduction processing to obtain the target occlusion region; Output the calibration information of the target occlusion area, the calibration information including the center coordinates of the target occlusion area.

2. The method for calibrating the radiation shielding area according to claim 1, characterized in that, The geometric parameters include: size and center coordinates; the noise reduction process performed on the candidate occlusion region based on the geometric parameters of the candidate occlusion region to obtain the target occlusion region includes: Based on the geometric parameters of the candidate occlusion region, the candidate occlusion region that meets the constraints is determined as the target occlusion region. The constraints include at least one of the following: area ratio constraint, size threshold constraint, and boundary position constraint.

3. The method for calibrating the radiation shielding area according to claim 2, characterized in that, The step of determining the candidate occlusion region as the target occlusion region based on the geometric parameters of the candidate occlusion region includes: In response to the constraint including an area ratio constraint, based on the size of the candidate occlusion region and the size of the corresponding standard occlusion region, the ratio of the area of ​​the candidate occlusion region to that of the corresponding standard occlusion region is determined; candidate occlusion regions whose area ratio is greater than or equal to a preset ratio are determined as the target occlusion regions; and / or, In response to the constraint condition including a size threshold constraint, a dynamic size threshold is determined based on the feature size of the standard occlusion region corresponding to all candidate occlusion regions; candidate occlusion regions corresponding to the standard occlusion regions with feature sizes greater than the dynamic size threshold are determined as the target occlusion region; and / or, In response to the constraints including boundary position constraints, based on the center coordinates, feature size, average feature size of the standard occlusion region corresponding to the candidate occlusion region and the boundary coordinates of the ray projection image, the candidate occlusion region corresponding to the standard occlusion region with a complete shape is determined as the target occlusion region.

4. The method for calibrating the radiation shielding area according to claim 1, characterized in that, When the imaging mode of the ray projection image is half-field imaging, before determining the geometric parameters of the candidate occlusion region in the ray projection image based on the ray projection image, the method further includes: Identify interference regions of a preset shape in the ray projection image; Remove the interference region of the preset shape from the ray projection image.

5. The method for calibrating the radiation shielding area according to claim 4, characterized in that, The identification of interference regions of a preset shape in the ray projection image includes: Based on at least one of spatial location constraints, size constraints, and boundary consistency constraints, candidate connected regions are identified in the ray projection image. Based on the edge gradient features of the candidate connected regions, the abrupt connection points of the candidate connected regions are determined; The candidate connected region is segmented at the abrupt connection to obtain the interference region and the candidate occlusion region.

6. The method for calibrating the radiation shielding area according to claim 1, characterized in that, The ray projection image is a binary image processed by binarization parameters; the method further includes: Based on the prior arrangement information of the ray blocking array, when it is verified that the spatial distribution of the target occlusion area does not conform to the expected arrangement rules, the binarization parameters are adjusted, and the ray projection image is updated based on the adjusted binarization parameters.

7. The method for calibrating the radiation shielding area according to claim 1, characterized in that, The method further includes: Based on the center coordinates and feature size of the standard occlusion region corresponding to the target occlusion region, the feature size is enlarged through a preset error tolerance to obtain the updated feature size corresponding to the target occlusion region. The step of outputting the calibration information of the target occlusion region includes: Output the updated feature size and center coordinates of the target occlusion region.

8. The method for calibrating the radiation shielding area according to any one of claims 1 to 7, characterized in that, The calibration information also includes a template image containing the target occlusion area.

9. A computing device, characterized in that, It includes a processor, a memory, and a program or instructions stored in the memory and executable on the processor, wherein the program or instructions, when executed by the processor, implement the steps of the calibration method for the radiation shielding area as described in any one of claims 1 to 8.

10. An image calibration system, characterized in that, Includes a radiation source, a detector, a radiation blocking array, and a computing device as described in claim 9. The radiation source is configured to emit radiation; The detector, positioned opposite the radiation source, is configured to receive radiation emitted by the radiation source and generate a radiation projection image. The ray blocking array is disposed on the ray path between the ray source and the detector and is configured to block part of the ray to form a target blocking area in the ray projection image generated by the detector.

Images