Method for dynamic digital zooming and beam center positioning in an x-ray device

Abstract

This invention relates to a dynamic digital zoom and X-ray beam center positioning method in an X-ray device. The X-ray device includes a vision acquisition module, a depth sensing module, an X-ray emission module, an image receiving module, and a control module. First, a mapping relationship between the X-ray beam propagation path and the visual coordinate system is established through geometric calibration. Then, RGB color video streams and three-dimensional depth data of the area to be photographed are simultaneously acquired. An artificial intelligence model, combined with domain-specific rules, is used to identify the exposure area and determine the theoretical center of the X-ray beam. Based on the mapping relationship and the three-dimensional depth data, the SID value is calculated to estimate the actual center of the X-ray beam and obtain the offset. According to the offset, digital zoom processing or an automatic collimation mechanism is used to align the center of the RGB color video stream with the center of the X-ray beam. The correction result is output in real time, dynamically responding to changes in the shooting state. This invention solves the problem of X-ray beam offset under different shooting conditions through multi-source data fusion and dynamic correction, improving the positioning accuracy of the exposure area.

Description

Technical Field

[0001] This invention relates to the field of medical imaging equipment technology, and in particular to a method for dynamic digital zoom and beam center positioning in X-ray equipment. Background Technology

[0002] In the field of medical imaging, X-ray imaging equipment is widely used in clinical examinations. Its RGB video camera can provide real-time video streams to help users complete the alignment, positioning and exposure area settings of the part to be photographed, which is a key link to ensure image quality.

[0003] However, due to structural limitations, RGB cameras cannot be installed at the focal point of the X-ray source; they can only be placed in the equipment's adaptation space or auxiliary observation position, resulting in an inherent deviation between their viewing angle and the X-ray beam propagation path. When adjusting the source-to-image distance (SID) to accommodate patients of different body types, or changing the shooting angle to obtain images of specific body positions, the position of the X-ray beam center in the RGB image will shift with changes in SID and angle. If not corrected, this can lead to alignment errors, inaccurate exposure area settings, increased risk of repeated shooting, and decreased image quality.

[0004] While existing technologies attempt to optimize the imaging process through depth sensing devices—such as measuring tissue thickness to adjust exposure dose and constructing 3D models to estimate the beam range—their core focus is on optimizing a single parameter. They can only statically set parameters and cannot address the real-time correction needs arising from temporary SID adjustments or patient position changes during imaging. Furthermore, existing technologies suffer from the following limitations: a lack of real-time correction capabilities, relying heavily on manual calibration before imaging (such as laser marking), requiring re-calibration after parameter changes; insufficient positioning accuracy, calculating the beam range solely from 2D image features without linking depth data to the beam propagation path, making precise center coordinate location difficult; and low automation, requiring manual judgment of offsets and adjustments, resulting in complex operations and inconsistent performance. Summary of the Invention

[0005] This invention provides a method for dynamic digital zoom and beam center positioning in X-ray equipment to solve the above-mentioned technical problems.

[0006] To address the aforementioned technical problems, this invention provides a method for dynamic digital zoom and beam center positioning in an X-ray device. The X-ray device includes a vision acquisition module, a depth sensing module, an X-ray emission module, an image receiving module, and a control module. The method includes the following steps:

[0007] Step 1: Equipment calibration: The control module performs geometric calibration on the vision acquisition module and the depth sensing module to obtain their relative positional relationship and internal and external parameters. At the same time, it calibrates the relative geometric position of the X-ray source focus of the X-ray emission module with the vision acquisition module and the depth sensing module, and establishes the mapping relationship between the X-ray beam propagation path and the visual coordinate system.

[0008] Step 2: Real-time data acquisition: The vision acquisition module acquires the RGB color video stream of the part to be photographed in real time, the depth sensing module simultaneously measures the relative distance between the part to be photographed and the depth sensing module and generates three-dimensional depth data, and the control module receives and stores the RGB color video stream and the three-dimensional depth data in real time.

[0009] Step 3: Exposure area identification: The control module calls the preset artificial intelligence model to process the image frames in the RGB color video stream, and combines the human pose estimation model with professional rules in the field of medical imaging to automatically identify the target exposure area, calculate the geometric center of the target exposure area and use it as the theoretical center of the ray beam.

[0010] Step 4: Offset Detection: Based on the mapping relationship and the three-dimensional depth data, the control module calculates the SID value, combines the ray beam linear propagation model, estimates the actual center position of the ray beam in the RGB color video stream, and compares the actual center position with the theoretical center position to obtain the offset.

[0011] Step 5: Dynamic correction: If the offset is less than a preset threshold, the control module performs digital zoom processing on the RGB color video stream based on the offset direction and the magnitude of the offset, so that the center of the RGB color video stream is aligned with the center of the X-ray beam; if the offset is greater than or equal to the preset threshold, the control module first drives the automatic collimation mechanism of the X-ray emission module to adjust the spatial attitude of the X-ray emission module, and then performs digital zoom processing.

[0012] Step 6: Real-time optimization: Output the corrected RGB color video stream, current SID value and offset in real time. If movement of the part to be filmed or change of SID is detected, repeat steps 3 to 5.

[0013] Preferably, the visual acquisition module is an RGB video camera.

[0014] Preferably, the depth sensing module is a ToF camera, a binocular stereo vision camera, or a structured light camera.

[0015] Preferably, the ToF camera uses infrared light waves or laser ranging; the binocular stereo vision camera calculates parallax using the principle of stereo vision.

[0016] Preferably, the intrinsic and extrinsic parameters include focal length, principal point coordinates, distortion parameters, and coordinate system transformation matrix.

[0017] Preferably, the artificial intelligence model is an object detection model trained based on a convolutional neural network, the YOLO algorithm, or the ResNet architecture.

[0018] Preferably, the professional rules in the field of medical imaging include at least the position and size of the exposure area under different body positions.

[0019] Preferably, the digital zoom method includes at least adjusting the image scaling ratio, cropping edge regions, or panning the image.

[0020] Preferably, the X-ray device further includes a display module, which synchronously displays the corrected RGB color video stream, the current SID value, and the offset in real time.

[0021] Preferably, the display module is also used to display the irradiation range and offset direction of the ray beam.

[0022] Compared with the prior art, the dynamic digital zoom and beam center positioning method in X-ray equipment provided by the present invention has the following advantages:

[0023] 1. This invention integrates the RGB color video stream of the visual acquisition module and the three-dimensional depth data of the depth sensing module, realizing real-time correction of the center position of the ray beam, effectively avoiding alignment deviations under different SIDs and shooting angles, and ensuring the accuracy of exposure area positioning.

[0024] 2. This invention uses an artificial intelligence model combined with professional rules in the field of medical imaging to automatically identify the target exposure area without the need for manual calibration, which significantly improves the system's automation level and reduces the complexity of user operation and human error.

[0025] 3. This invention dynamically calculates the SID value based on three-dimensional depth data, and synchronously adjusts the digital zoom parameters to adapt to dynamic scenarios such as movement of the part to be photographed and changes in shooting conditions, thereby optimizing image quality, reducing the probability of repeated shooting, and improving the efficiency of clinical examination.

[0026] 4. This invention significantly improves the positioning accuracy of X-ray equipment through a dual correction mechanism of digital zoom processing and automatic collimation adjustment, combined with real-time data feedback and status response, adapting to the imaging needs of patients with complex positions and different body types, and broadening the application scenarios of the equipment. Attached Figure Description

[0027] Figure 1 This is a flowchart of a method for dynamic digital zoom and beam center positioning in an X-ray device according to a specific embodiment of the present invention. Detailed Implementation

[0028] To illustrate the technical solutions of the invention in more detail, specific embodiments are listed below to demonstrate the technical effects; it should be emphasized that these embodiments are used to illustrate the invention and not to limit the scope of the invention.

[0029] The present invention provides a method for dynamic digital zoom and beam center positioning in X-ray equipment, such as... Figure 1 As shown, the X-ray device includes a vision acquisition module, a depth sensing module, an X-ray emission module, an image receiving module, and a control module. The method includes the following steps:

[0030] Step 1: Equipment Calibration: The control module performs geometric calibration on the vision acquisition module and the depth sensing module, obtains their relative positional relationship (including rotation matrix and translation vector) and intrinsic and extrinsic parameters, and simultaneously calibrates the relative geometric position of the X-ray source focus of the X-ray emission module with the vision acquisition module and the depth sensing module, establishing the mapping relationship between the X-ray beam propagation path and the visual coordinate system.

[0031] Step 2: Real-time data acquisition: The vision acquisition module acquires the RGB color video stream of the part to be photographed in real time, the depth sensing module simultaneously measures the relative distance between the part to be photographed and the depth sensing module and generates three-dimensional depth data, and the control module receives and stores the RGB color video stream and the three-dimensional depth data in real time.

[0032] Step 3: Exposure Area Recognition: The control module calls a preset artificial intelligence model to process the image frames in the RGB color video stream. Combining the human pose estimation model with professional rules in the field of medical imaging, it automatically identifies the target exposure area, calculates the geometric center of the target exposure area and uses it as the theoretical center of the X-ray beam. The exposure area is defined as: the area that is expected to be irradiated by the X-ray beam under a specific inspection view.

[0033] Step 4: Offset Detection: Based on the mapping relationship and the three-dimensional depth data, the control module calculates the SID value, combines the ray beam linear propagation model, estimates the actual center position of the ray beam in the RGB color video stream, and compares the actual center position with the theoretical center position to obtain the offset.

[0034] Step 5: Dynamic correction: If the offset is less than a preset threshold, in some embodiments, the preset threshold ranges from 0.5 to 2 pixels. The control module performs digital zoom processing on the RGB color video stream based on the offset direction and the magnitude of the offset, so that the center of the RGB color video stream is aligned with the center of the X-ray beam. If the offset is greater than or equal to the preset threshold, the control module first drives the automatic collimation mechanism of the X-ray emission module to adjust the spatial attitude of the X-ray emission module, and then performs digital zoom processing.

[0035] Step 6: Real-time optimization: Output the corrected RGB color video stream, current SID value and offset in real time. If movement of the part to be filmed or change of SID is detected, repeat steps 3 to 5 to achieve dynamic optimization.

[0036] This invention integrates multi-source data and full-process dynamic correction, solving the problem of X-ray beam deviation under different shooting conditions; it can complete exposure area identification and correction without manual intervention, simplifying the operation process, reducing operational complexity and human error; it dynamically responds to changes in shooting status, reducing the rate of repeated shooting and improving examination efficiency; it is adaptable to patients of different body types and complex body positions, broadening the application scenarios of the equipment and meeting diverse clinical diagnostic needs.

[0037] In some embodiments, the visual acquisition module is an RGB video camera. The frame rate of the RGB video camera is adapted to the real-time positioning requirements of the X-ray equipment, and the resolution is not lower than the high-definition display standard, providing clear and real-time image data support for subsequent exposure area identification and offset detection.

[0038] In some embodiments, the depth sensing module is a ToF camera, a binocular stereo vision camera, or a structured light camera to meet different measurement distance, accuracy, and field of view requirements, ensuring the comprehensiveness and accuracy of three-dimensional depth data acquisition, and further providing reliable data support for SID calculation and offset detection.

[0039] In some embodiments, the ToF camera uses infrared light waves or laser ranging. For example, by emitting infrared light waves and receiving reflected waves, it calculates the time difference of light wave propagation, and combines this with the speed of light to deduce the relative distance between the part to be photographed and the ToF camera, generating three-dimensional depth data. The binocular stereo vision camera calculates parallax using the principle of stereo vision. For example, dual cameras simultaneously acquire images of the part to be photographed, and the control module calculates the image parallax using a stereo matching algorithm. Combined with the principle of triangulation, it derives the three-dimensional spatial coordinates and depth data.

[0040] In some embodiments, the intrinsic and extrinsic parameters include focal length, principal point coordinates, distortion parameters, and coordinate system transformation matrix. Parameter calibration is performed by acquiring multi-view images using a calibration board. Specifically, the calibration board can be placed at different positions and angles within the shooting area, and the vision acquisition module and depth sensing module can be controlled to simultaneously acquire multiple sets of images. By identifying the feature points of the calibration board, the focal length, principal point coordinates, distortion parameters, and coordinate system transformation matrix are calculated to complete geometric calibration.

[0041] In some embodiments, the artificial intelligence model is a target detection model trained based on a convolutional neural network, YOLO algorithm, or ResNet architecture. The control module calls the optimized target detection model. The training dataset for training the target detection model contains medical image samples of different body positions and different human body parts, thereby performing feature extraction and target recognition on the image frames of the RGB video stream and outputting the coordinates of the exposure area.

[0042] In some embodiments, the professional rules in the field of medical imaging include at least the position and size of the exposure area under different body positions, providing clear constraint boundaries for the artificial intelligence model and avoiding recognition deviations caused by training data bias.

[0043] In some embodiments, the digital zoom method includes at least adjusting the image scaling ratio, cropping edge regions, or panning the image. After processing, the effective visual information of the RGB color video stream is not lost. The present invention uses a variety of digital zoom methods in flexible combination to adapt to different offset scenarios and improve correction accuracy.

[0044] In some embodiments, the X-ray device further includes a display module, such as a medical high-definition display, on which the corrected RGB color video stream, the current SID value, and the offset are displayed synchronously in real time to ensure clear image details after correction. In some embodiments, the display module is also used to display the irradiation range and offset direction of the X-ray beam. For example, the control module can generate a schematic diagram (rectangle) of the irradiation range of the X-ray beam based on the X-ray beam propagation model and correction parameters, and generate an indicator arrow based on the offset direction (e.g., a rightward arrow is displayed if the beam is horizontally offset to the right). In this embodiment, an overlay method is used for display, such as the rectangle being overlaid on the corresponding position of the RGB color video stream, and the arrow being overlaid on the edge of the image, with the offset value marked, which is intuitive and clear, helping users to make adjustments quickly.

[0045] In summary, the present invention provides a dynamic digital zoom and beam center positioning method for X-ray equipment. The X-ray equipment includes a vision acquisition module, a depth sensing module, an X-ray emission module, an image receiving module, and a control module. The method includes the following steps: Step 1: Equipment calibration: The control module performs geometric calibration on the vision acquisition module and the depth sensing module, obtains their relative positional relationship and intrinsic and extrinsic parameters, and simultaneously calibrates the relative geometric position of the X-ray source focal point of the X-ray emission module with the vision acquisition module and the depth sensing module, establishing a mapping relationship between the beam propagation path and the visual coordinate system; Step 2: Real-time data acquisition: The vision acquisition module acquires the RGB color video stream of the area to be photographed in real time, and the depth sensing module simultaneously measures the relative distance between the area to be photographed and the depth sensing module and generates three-dimensional depth data. The control module receives and stores the RGB color video stream and the three-dimensional depth data in real time; Step 3: Exposure area recognition: The control module calls a preset artificial intelligence model to process the image frames in the RGB color video stream, combined with human posture. The estimation model and professional rules in the field of medical imaging are used to automatically identify the target exposure area, calculate the geometric center of the target exposure area and use it as the theoretical center of the X-ray beam; Step 4: Offset detection: The control module calculates the SID value based on the mapping relationship and three-dimensional depth data, and estimates the actual center position of the X-ray beam in the RGB color video stream by combining the X-ray beam linear propagation model. The actual center position is compared with the theoretical center position to obtain the offset; Step 5: Dynamic correction: If the offset is less than a preset threshold, the control module performs digital zoom processing on the RGB color video stream based on the offset direction and the magnitude of the offset, so that the center of the RGB color video stream is aligned with the center of the X-ray beam; If the offset is greater than or equal to the preset threshold, the control module first drives the automatic collimation mechanism of the X-ray emission module to adjust the spatial attitude of the X-ray emission module, and then performs digital zoom processing; Step 6: Real-time optimization: The corrected RGB color video stream, the current SID value and the offset are output in real time. If the movement of the part to be photographed or the SID change is detected, steps 3 to 5 are repeated to achieve dynamic optimization. This invention can measure distance in real time and dynamically adjust the position of the beam center to compensate for offset under different SID conditions. In turn, it can improve image alignment accuracy, exposure area setting accuracy and imaging efficiency through digital zoom / view correction.

[0046] Obviously, those skilled in the art can make various modifications and variations to the invention without departing from the spirit and scope of the invention. Therefore, if these modifications and variations fall within the scope of the claims of the invention and their equivalents, the invention is also intended to include these modifications and variations.

Claims

1. A method for dynamic digital zoom and beam center positioning in an X-ray device, characterized in that, The X-ray device includes a vision acquisition module, a depth sensing module, an X-ray emission module, an image receiving module, and a control module. The method includes the following steps: Step 1: Equipment calibration: The control module performs geometric calibration on the vision acquisition module and the depth sensing module to obtain their relative positional relationship and internal and external parameters. At the same time, it calibrates the relative geometric position of the X-ray source focus of the X-ray emission module with the vision acquisition module and the depth sensing module, and establishes the mapping relationship between the X-ray beam propagation path and the visual coordinate system. Step 2: Real-time data acquisition: The vision acquisition module acquires the RGB color video stream of the part to be photographed in real time, the depth sensing module simultaneously measures the relative distance between the part to be photographed and the depth sensing module and generates three-dimensional depth data, and the control module receives and stores the RGB color video stream and the three-dimensional depth data in real time. Step 3: Exposure area identification: The control module calls the preset artificial intelligence model to process the image frames in the RGB color video stream, and combines the human pose estimation model with professional rules in the field of medical imaging to automatically identify the target exposure area, calculate the geometric center of the target exposure area and use it as the theoretical center of the ray beam. Step 4: Offset Detection: Based on the mapping relationship and the three-dimensional depth data, the control module calculates the SID value, combines the ray beam linear propagation model, estimates the actual center position of the ray beam in the RGB color video stream, and compares the actual center position with the theoretical center position to obtain the offset. Step 5: Dynamic correction: If the offset is less than a preset threshold, the control module performs digital zoom processing on the RGB color video stream based on the offset direction and the magnitude of the offset, so that the center of the RGB color video stream is aligned with the actual center of the X-ray beam; if the offset is greater than or equal to the preset threshold, the control module first drives the automatic collimation mechanism of the X-ray emission module to adjust the spatial attitude of the X-ray emission module, and then performs digital zoom processing. Step 6: Real-time optimization: Output the corrected RGB color video stream, current SID value and offset in real time. If movement of the part to be filmed or change of SID is detected, repeat steps 3 to 5.

2. The dynamic digital zoom and beam center positioning method in an X-ray device as described in claim 1, characterized in that, The visual acquisition module is an RGB video camera.

3. The dynamic digital zoom and beam center positioning method in an X-ray device as described in claim 1, characterized in that, The depth sensing module is a ToF camera, a binocular stereo vision camera, or a structured light camera.

4. The dynamic digital zoom and beam center positioning method in an X-ray device as described in claim 3, characterized in that, The ToF camera uses infrared light waves or lasers for ranging; the binocular stereo vision camera calculates parallax based on the principle of stereo vision.

5. The dynamic digital zoom and beam center positioning method in an X-ray device as described in claim 1, characterized in that, The intrinsic and extrinsic parameters include focal length, principal point coordinates, distortion parameters, and coordinate system transformation matrix.

6. The dynamic digital zoom and beam center positioning method in an X-ray device as described in claim 1, characterized in that, The artificial intelligence model is an object detection model trained based on convolutional neural networks, YOLO algorithm, or ResNet architecture.

7. The dynamic digital zoom and beam center positioning method in an X-ray device as described in claim 1, characterized in that, The professional rules in the field of medical imaging include at least the location and size of the exposure area under different body positions.

8. The method for dynamic digital zoom and beam center positioning in an X-ray device as described in claim 1, characterized in that, The digital zoom method includes at least adjusting the image scaling ratio, cropping edge regions, or panning the image.

9. The dynamic digital zoom and beam center positioning method in an X-ray device as described in claim 1, characterized in that, The X-ray device also includes a display module, which displays the corrected RGB color video stream, the current SID value, and the offset in real time.

10. The method of claim 9, wherein the X-ray device is a computed radiography (CR) device. The display module is further configured to display a range of the radiation beam irradiation and a direction of the offset. The display module is further configured to display a range of the radiation beam irradiation and

Images