Object positioning method, radiation delivery system, and storage medium

Abstract

The invention relates to an object positioning method, a radiation delivery system and a storage medium. The method comprises the following steps: acquiring a first posture image of a first positioning posture when a target object is positioned at a first positioning position and a second posture image of a second positioning posture when the target object is positioned at a second positioning position after the first positioning position; and determining a first position of one or more key points of the target object in the first attitude image and a second position of one or more key points in the second attitude image, and determining positioning information of the target object according to the first position and the second position. According to the method, the target object is automatically guided to be positioned by using the radiation delivery system, and particularly, the positioning state or the positioning deviation of the target object between different batches is quickly found and corrected by comparing the postures of the target object under different positioning positions in real time; the target object can automatically adjust the posture according to the prompt or guidance of the radiation delivery system, and the positioning efficiency is improved.

Description

Technical Field

[0001] The present application relates to the field of medical technology, and in particular to an object positioning method, a radiation delivery system, an object positioning device, a radiation delivery equipment, a storage medium, and a program product. Background Art

[0002] With the rapid development of medical equipment, users have increasingly higher requirements for the accuracy of treatment using medical equipment. For example, when using radiotherapy equipment to treat patients between fractions, the patient needs to be positioned multiple times during the fractionated treatment so that the medical equipment can obtain the accurate dose of the target area.

[0003] In related technologies, radiotherapy positioning usually relies on laser lights, and medical staff use experience to make manual adjustments with the help of laser lines and body marker lines. This method relies heavily on the accuracy of laser light placement, which places high demands on installation and service engineers. In addition, the repeatability of positioning between fractions is limited by the skills and experience of medical staff, and there are also deviations in the positioning of fractions performed by different medical staff, which limits the use of radiotherapy equipment and fails to achieve optimal medical performance.

[0004] Therefore, how to provide an automatic positioning method has become a technical problem that needs to be solved urgently in the current medical technology field. Summary of the invention

[0005] Based on this, it is necessary to provide an object positioning method, radiation delivery system, object positioning device, radiation delivery equipment, storage medium, and program product that can automatically guide users to position themselves during radiotherapy in response to the above-mentioned technical problems.

[0006] In a first aspect, the present application provides an object positioning method, comprising:

[0007] Acquire a first posture image of a first position posture of a target object in a first position, and a second posture image of a second position posture of the target object in a second position after the first position;

[0008] Determining first positions of one or more key points of the target object in the first pose image and second positions of the one or more key points in the second pose image;

[0009] The positioning information of the target object is determined according to the first position and the second position.

[0010] In one embodiment, determining the positioning information of the target object according to the first position and the second position includes:

[0011] According to the first position and the second position, a positioning state and / or a positioning deviation of the second positioning posture of the target object relative to the first positioning posture is determined as positioning information.

[0012] In one of the embodiments, the positioning information is output to the target object in the form of voice and / or image, so as to guide the target object to perform positioning.

[0013] In one embodiment, determining a first position of one or more key points of the target object in the first posture image and a second position of the one or more key points in the second posture image includes:

[0014] Identify one or more first key points in the first posture image to obtain first position coordinates of each of the first key points;

[0015] One or more second key points in the second posture image are identified to obtain second position coordinates of each second key point; wherein the second key points are consistent in number and type with the first key points, and the first position coordinates of each first key point and the second position coordinates of each second key point are used to determine the positioning deviation.

[0016] In one embodiment, the positioning deviation includes a position deviation and a direction deviation, and a method for determining the positioning deviation according to the first position coordinates of each of the first key points and the second position coordinates of each of the second key points includes:

[0017] Performing distance calculation on each of the first position coordinates and the corresponding each of the second position coordinates to determine a key point position deviation between each of the first key points and the corresponding each of the second key points;

[0018] Calculating the angle between each of the first position coordinates and the corresponding second position coordinates to determine the key point direction deviation between each of the first key points and the corresponding second key points;

[0019] The positioning deviation is determined according to the key point position deviation and the key point direction deviation.

[0020] In one embodiment, determining the positioning deviation according to the key point position deviation and the key point direction deviation includes:

[0021] Performing a mean operation on the position deviations of all or part of the key points to determine an average position deviation of the second positioning posture relative to the first positioning posture;

[0022] Perform a mean operation on all or part of the key point directional deviations to determine an average directional deviation of the second positioning posture relative to the first positioning posture.

[0023] In one embodiment, the positioning deviation includes a depth deviation, and the object positioning method further includes:

[0024] Acquire acquisition parameters for acquiring the second posture image and iris information of the target object;

[0025] A depth deviation of the second positioning posture relative to the first positioning posture is determined according to the acquisition parameter and the iris information.

[0026] In one embodiment, the first position is an initial position of the target object during or before radiation delivery planning, and the second position is a delivery position after radiation delivery planning and during or before radiation delivery.

[0027] In a second aspect, a radiation delivery system comprises: at least one image acquisition device, a positioning device, and a support platform;

[0028] The image acquisition device is used to acquire a first posture image of a first positioning posture of the target object when the target object is in a first positioning on the support platform, and a second posture image of a second positioning posture of the target object when the target object is in a second positioning after the first positioning on the support platform;

[0029] The positioning device is used to determine a first position of one or more key points of the target object in the first posture image and a second position of the one or more key points in the second posture image, and determine positioning information according to the first position and the second position.

[0030] In a third aspect, an object positioning device is provided, the object positioning device comprising:

[0031] an acquisition module, configured to acquire a first posture image of a first position posture of a target object in a first position, and a second posture image of a second position posture of the target object in a second position after the first position;

[0032] A first determination module is used to determine a first position of one or more key points of the target object in the first posture image and a second position of the one or more key points in the second posture image, wherein the first position and the second position are used to determine the positioning information of the target object;

[0033] The second determining module is used to determine the positioning information of the target object according to the first position and the second position.

[0034] In a fourth aspect, a radiation delivery device comprises a memory and a processor, wherein the memory stores a computer program, and the processor implements the object positioning method described in the first aspect when executing the computer program.

[0035] In a fifth aspect, a computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the object positioning method described in the first aspect is implemented.

[0036] In a sixth aspect, a computer program product includes a computer program, and when the computer program is executed by a processor, the object positioning method described in the first aspect is implemented.

[0037] The above-mentioned object positioning method, radiation delivery system, object positioning device, radiation delivery equipment, storage medium and program product obtain a first posture image of a first positioning posture of a target object in a first positioning, and a second posture image of a second positioning posture of the target object in a second positioning after the first positioning; determine the first position of one or more key points of the target object in the first posture image and the second position of one or more key points in the second posture image, and determine the positioning information of the target object according to the first position and the second position. The above-mentioned method realizes a method for automatically guiding the positioning of the target object using a radiation delivery system, specifically by comparing the postures of the target object in different positioning in real time, quickly discovering and correcting the positioning state or positioning deviation of the target object between different fractions, so that the target object can adjust its posture according to the prompt or guidance of the radiation delivery system, thereby improving the positioning efficiency; moreover, the above-mentioned method determines the positioning information by identifying the positions of key points in different posture images, and can accurately determine the positioning deviation between different positioning postures of the target object, and then accurately perform positioning guidance, which solves the repeatability problem of the target object in multiple positioning to a certain extent. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the drawings required for use in the embodiments of the present application or related technical descriptions will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0039] Figure 1 is one of the structural schematic diagrams of a radiation delivery system in one embodiment;

[0040] Figure 2 is one of the flowcharts of the object positioning method in one embodiment;

[0041] Figure 3 The second flowchart of the object positioning method in one embodiment is shown;

[0042] Figure 4 The third flowchart of the object positioning method in one embodiment;

[0043] Figure 5 FIG4 is a fourth flow chart of an object positioning method in one embodiment;

[0044] Figure 6 FIG5 is a fifth flow chart of an object positioning method in one embodiment;

[0045] Figure 7 FIG6 is a sixth flow chart of an object positioning method in one embodiment;

[0046] Figure 8 is a schematic diagram of a positioning image in one embodiment;

[0047] Fig. 9 is a schematic diagram of a display interface in one embodiment;

[0048] Fig.10 FIG7 is a flowchart of an object positioning method according to an embodiment;

[0049] Fig.11 is a second structural schematic diagram of a radiation delivery system in one embodiment;

[0050] Fig.12 is a structural block diagram of an object positioning device in one embodiment;

[0051] Fig.13 FIG. 1 is a diagram of the internal structure of a radiation delivery device in one embodiment. DETAILED DESCRIPTION

[0052] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0053] The positioning method provided in the embodiment of the present application can be applied to Figure 1In the radiation delivery system shown. The radiation delivery system includes: at least one image acquisition device 101, an object positioning device 102 and a support platform 103; the object positioning device 102 is connected to the image acquisition device 101 and the support platform 103 by wire or wireless respectively; the image acquisition device 101 is used to acquire a posture image of the target object on the support platform 103; the object positioning device 102 is used to analyze the posture image acquired by the image acquisition device 101, so as to guide the target object to be positioned on the support platform 103 according to the analysis result; the support platform 103 is used to carry the target object and drive the target object to move during the positioning process. Among them, the image acquisition device 101 can be a camera, or a camera, etc.; the object positioning device 101 can store the image acquired by the image acquisition device 101. Among them, the object positioning device 101 can be, but not limited to, various personal computers, laptops, smart phones, tablet computers, Internet of Things devices, and projection devices. It should be noted that the positioning method provided in the embodiment of the present application can be used in the radiation delivery system as mentioned above, and can also be applied to other imaging systems or devices with positioning requirements. The above-mentioned radiation delivery system is only an explanation of an application requirement of the positioning method described in the embodiment of the present application, and is not a limitation thereto.

[0054] Those skilled in the art will understand that Figure 1 The structure shown in the figure is only a block diagram of a part of the structure related to the present application scheme, and does not constitute a limitation on the radiation delivery system to which the present application scheme is applied. The specific radiation delivery system may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components. In this article, the terms "positioning" and "positioning" are used interchangeably.

[0055] In an exemplary embodiment, Figure 2 As shown, a method for positioning an object is provided, and the method is applied to Figure 1 The present invention is described as an example of an object positioning device in a radiation delivery system, including:

[0056] S201, acquiring a first posture image of a target object in a first position and a second posture image of the target object in a second position after the first position.

[0057] Among them, the first positioning can be the initial positioning of the target object during or before the formulation of the radiation delivery plan. Correspondingly, the first posture image can be a posture image of the target object when it is positioned under the guidance of medical staff during or before the formulation of the radiation delivery plan, or it can be a posture image of the target object when it is positioned under the guidance of the radiation delivery system during or before the formulation of the radiation delivery plan. For example, the first posture image can be a posture image of the initial positioning, or it can be a posture image of any of the posture images of multiple positioning steps. The second positioning is the delivery positioning after the formulation of the radiation delivery plan and before the radiation delivery. Correspondingly, the second posture image is a posture image of the target object when it is positioned after the formulation of the radiation delivery plan and before the radiation delivery, for example, the second posture image is a posture image of any of the first, second, third, ..., Nth steps of the target object after the first positioning.

[0058] In an embodiment of the present application, during or before the formulation of the radiation delivery plan, when the target object performs the first positioning and assumes the first positioning posture, the image acquisition device is started to acquire the posture image of the target object as the first posture image. For example, after the first positioning and after the formulation of the radiation delivery plan, during or before the formulation of the radiation delivery plan, when the target object needs to perform the second positioning, the image acquisition device is controlled to acquire the current image of the target object that can reflect the posture of the target object, and the acquired image is used as the second posture image. Specifically, the above-mentioned positioning and posture image acquisition process can be a cyclic process. For example, when the target object performs the first positioning, the image acquisition device is started to collect the posture image of the target object as the first posture image of the first time; after the first positioning, when the target object performs the first second positioning, the image acquisition device is started to collect the posture image of the target object as the second posture image of the first time; after the first second positioning, when the target object performs the second second positioning, the image acquisition device is started to collect the posture image of the target object as the second posture image of the second time; after the second second positioning, when the target object performs the third second positioning, the image acquisition device is started to collect the posture image of the target object as the third posture image of the third time; and so on and so forth until the last second posture image is collected. It should be noted that when the first posture image is obtained in the first positioning, the first posture image can be saved for use in the analysis of the positioning information in the next time. The positioning information can include various types of information associated with the positioning, which is not limited here. As a non-limiting example, the positioning information can include the positioning posture or state of the target object, the positioning error, the recommended positioning direction, the recommended positioning distance, whether the positioning result or process meets the specified standard, whether the positioning is in progress, and so on. Optionally, each time the second posture image is acquired at the second position, the second posture images of all the second positions may be saved so that the user can view them at any time.

[0059] S202: Determine first positions of one or more key points of the target object in the first posture image and second positions of one or more key points in the second posture image.

[0060] Key points may include reference sites (e.g., part points or joint points, etc.) that are strongly related to the posture, motion state, identity, etc. of the target object, or key points are reference sites that need to be focused on or are essential for determining the posture, motion state, identity, etc. of the target object. Taking the human body as an example, body key points, facial key points, etc. may be included. Taking the target object as a person as an example, key points may include body key points and facial key points, wherein body key points may include joint points on the body, points on the body symmetry axis, etc. Facial key points may include feature points on the facial features, points on the facial symmetry axis, etc. Taking the target object as an object as an example, center points, vertices, etc. may be included. Key points may include position points on the surface of the object, or position points in or inside the object that can be predicted or inferred based on points on the surface of the object.

[0061] In an embodiment of the present application, an optional method is that when the object positioning device obtains the first posture image based on the aforementioned steps, it can perform position recognition on one or more key points contained in the first posture image, identify the position coordinates of the first position of each key point, and save the position coordinates of the first position of each key point identified to the database; when the object positioning device obtains the second posture image of the next segment based on the aforementioned steps, it can perform position recognition on one or more key points contained in the second posture image, identify the position coordinates of the second position of each key point, and then extract the position coordinates of the first position from the database, and analyze the positioning information of the target object based on the position coordinates of the first position and the position coordinates of the second position. The above method is that each time the object positioning device obtains a posture image, it can perform position recognition on the key points in the posture image, and save the position coordinates of the key points after identifying them so that they can be called at any time when used. Another option is that when the object positioning device obtains the first posture image based on the above steps, the first posture image can be saved in a database, and then when the object positioning device obtains the second posture image of the next division based on the above steps, the first posture image can be extracted from the database, and the key points in the first posture image and the second posture image can be positionally identified to obtain the position coordinates of the first position of the key point in the first posture image and the position coordinates of the second position of the key point in the second posture image, and then, according to the position coordinates of the first position and the position coordinates of the second position, the positioning information of the target object is analyzed. The above method is that each time the object positioning device obtains a posture image, it is first saved, and when it is necessary to analyze the positioning information between different divisions, the posture images of different divisions are extracted from the database to perform position identification of key points to determine the positioning information of the second position of the target object relative to the first position each time.

[0062] S203: Determine positioning information of the target object according to the first position and the second position.

[0063] The positioning information may include the positioning state and / or positioning deviation of the second positioning posture of the target object relative to the first positioning posture. Optionally, the positioning information is used to guide the target object to position. The positioning information is output to the target object in the form of voice and / or image, and is used to guide the target object to position. The positioning information may include the positioning direction in which the target object needs to be positioned and the positioning distance in which the target object needs to be positioned, wherein the positioning direction may be expressed using conventional terms such as "left", "right", "up", "down", "forward", and "backward" to indicate directions. The positioning information also includes a guidance image, and the guidance image includes an image or outline of the target object, and a positioning mark, wherein the positioning mark includes a direction mark and a distance mark, and the direction mark is used to indicate that the target object is positioned in the direction of the direction mark. The distance mark is used to indicate that the target object is positioned in accordance with the distance mark.

[0064] In an embodiment of the present application, when the object positioning device obtains the first position of one or more key points of the target object in the first posture image and the second position of one or more key points in the second posture image based on the aforementioned steps, the position deviation between the first position of each key point in the first posture image and the second position of each corresponding key point in the second posture image can be analyzed, and the positioning state and / or positioning deviation of the second positioning posture of the target object relative to the first positioning posture can be further determined based on the position deviation corresponding to each key point, and then based on the positioning state and / or positioning deviation of the second positioning posture of the target object relative to the first positioning posture, positioning information including positioning direction and positioning distance is generated, or positioning information including positioning image is generated, or positioning information including positioning direction, positioning distance and positioning image is generated, so that the positioning information can guide the user to perform positioning.

[0065] The method described in the above embodiment is applied to a radiation delivery system, which obtains a first posture image of a first positioning posture of a target object in a first positioning position, and a second posture image of a second positioning posture of the target object in a second positioning position after the first positioning position; determines a first position of one or more key points of the target object in the first posture image and a second position of one or more key points in the second posture image, and determines the positioning information of the target object according to the first position and the second position. The above method realizes a method of automatically guiding the positioning of a target object using a radiation delivery system, specifically by comparing the postures of the target object in different positioning positions in real time, quickly discovering and correcting the positioning state or positioning deviation of the target object between different fractions, so that the target object can adjust its posture according to the prompt or guidance of the radiation delivery system, thereby improving the positioning efficiency; moreover, the above method determines the positioning information by identifying the positions of key points in different posture images, and can accurately determine the positioning deviation between different positioning postures of the target object, and then accurately provide positioning guidance, which solves the repeatability problem of the target object in multiple positioning to a certain extent.

[0066] In an exemplary embodiment, a method for determining positioning information is provided, the method comprising: determining a positioning state and / or a positioning deviation of a second positioning posture of a target object relative to a first positioning posture as positioning information based on a first position and a second position.

[0067] Among them, the positioning state can be used to indicate whether the second positioning posture of the target object is consistent with the first positioning posture, or whether the second positioning posture of the target object meets the requirements. Position deviation can include at least one of distance deviation, direction deviation and depth deviation, wherein the distance deviation refers to the position of the target object in the second positioning on the three axes of the three-dimensional space (for example, the X axis, the Y axis and the Z axis, where the directions of the three axes can be set arbitrarily, as long as the three axes are perpendicular to each other), relative to the distance deviation of the initial position of the target object in the first positioning between the three axes of the three-dimensional space. Direction deviation refers to the deviation between the direction of the target object in the second positioning and the direction of the initial position of the target object in the first positioning. Depth deviation refers to the distance between the target object and the image acquisition device in the Z direction in the image.

[0068] In the embodiment of the present application, when the object positioning device determines the first position of the key point in the first posture image and the second position of the key point in the second posture image, the positioning information of the target object can be further determined based on the first position and the second position. Specifically, the positioning state of the second positioning posture of the target object relative to the first positioning posture can be analyzed, and the positioning deviation of the second positioning posture of the target object relative to the first positioning posture can also be analyzed, and then the above-mentioned positioning state and / or positioning deviation are used as positioning information. For example, through the difference between the first position and the second position, at least one of the distance deviation, direction deviation, and depth deviation between the target object in the first positioning and the second positioning is analyzed. And / or, it is analyzed whether the posture of the target object in the first positioning and the second positioning is consistent, or whether the corresponding posture state of the target object in the second positioning meets the positioning requirements. Furthermore, the object positioning device can also analyze the distance value that the target object needs to move based on the distance deviation in the positioning deviation, and analyze the specific direction that the target object needs to move based on the direction deviation in the positioning deviation, and then convert the distance value that the target object needs to move and the specific direction that the target object needs to move into a terminology expression for guiding movement, and perform positioning guidance based on the terminology expression. Optionally, when the object positioning device determines the positioning information of the target object based on the aforementioned steps, it can also generate a guidance image based on the positioning state and / or positioning deviation. Continuing with the above, when the positioning information includes a distance value and a direction, the positioning device can output the positioning information by voice broadcast or by display, so that the target object can swing according to the positioning information; when the positioning information includes a guidance image, the positioning device can output the positioning information by display, so that the target object can clearly see the positioning mark and position according to the positioning mark.

[0069] In an exemplary embodiment, a method for identifying key points is provided, that is, the above S202 "determines a first position of one or more key points of the target object in the first posture image and a second position of one or more key points in the second posture image", such as Figure 3 As shown, the method includes:

[0070] S301, identifying one or more key points in a first posture image to obtain first position coordinates of each key point.

[0071] The key points in the first posture image may be key points of all regions of the target object in the first posture image, such as key points of the whole body; optionally, the key points in the first posture image may also be several key points of a specified region of the target object, such as key points of the upper body, key points of the face, etc. The first position coordinates represent the coordinates of the X-axis, Y-axis, and Z-axis in the three-dimensional space.

[0072] In the embodiment of the present application, the radiation delivery system can realize the recognition of key points in the first posture image based on any type of neural network model to obtain the first position coordinates of each key point; specifically, the radiation delivery system can use the neural network model in the existing model library to recognize the key points in the image, or it can obtain the neural network model based on sample data training online, and use the trained neural network model to recognize the key points in the first posture image to obtain the first position coordinates of each key point. For example, the radiation delivery system is based on the mediapipe built on the BlazePose algorithm (the BlazePose algorithm is a lightweight convolutional neural network architecture for real-time human posture estimation), and uses a lightweight CNN model that can be deployed on a low-cost CPU device to identify the position of key points of the human body. Optionally, when the key points are several key points in a designated area on the target object, the radiation delivery system needs to first determine the designated area in the first posture image, and then identify the key points in the designated area to obtain the first position coordinates of each key point. The above process can be implemented using two neural network models. For example, a segmentation model or a classification model based on a neural network architecture is used to segment the image of the designated area in the first posture image, and then a recognition model of a neural network architecture is further used to recognize the image of the designated area to identify the first position coordinates of each key point.

[0073] S302: Identify one or more key points in the second posture image to obtain second position coordinates of each key point.

[0074] Among them, the key points in the second posture image can be the key points of all areas on the target object in the second posture image, such as the key points of the whole body; optionally, the key points in the second posture image can also be several key points of a specified area on the target object, such as the key points of the upper body or the key points of the face, etc. If the key points in the first posture image and the key points in the second posture image are both several key points of the specified area, the number and type of the key points in the first posture image are consistent with those in the second posture image. The second position coordinates represent the coordinates of the X-axis, Y-axis, and Z-axis in three-dimensional space.

[0075] In the embodiment of the present application, the radiation delivery system can realize the recognition of key points in the second posture image based on any type of neural network model to obtain the second position coordinates of each key point; specifically, the radiation delivery system can use the neural network model in the existing model library to recognize the key points in the second posture image, or it can obtain the neural network model based on sample data training online, and use the trained neural network model to recognize the key points in the second posture image to obtain the second position coordinates of each key point. For example, the radiation delivery system is based on MediaPipe built on the BlazePose algorithm (the BlazePose algorithm is a lightweight convolutional neural network architecture for real-time human posture estimation), and uses a lightweight CNN model that can be deployed on a low-cost CPU device to identify the position of key points of the human body. Optionally, when the key points in the second posture image are several key points of a designated area on the target object, the radiation delivery system needs to first determine the designated area in the second posture image, and then identify the key points in the designated area to obtain the second position coordinates of each key point. The above process can be implemented using two neural network models. For example, a segmentation model or a classification model based on a neural network architecture is used to segment the image of the designated area in the fractionated posture image, and then a recognition model of a neural network architecture is further used to recognize the image of the designated area to identify the second position coordinates of each key point.

[0076] Following the above steps, when the object positioning device determines the first position coordinates of the key point in the first posture image and the second position coordinates of the key point in the second posture image, the positioning information of the target object can be further determined based on the first position coordinates and the second position coordinates. Specifically, since the position coordinates of the key point can be the position coordinates of the X-axis, Y-axis, and Z-axis in the three-dimensional space, when calculating for a key point in a first posture image and a key point in a second posture image, the difference between the position coordinates of the key point in the first posture image and the key point in the second posture image on the X-axis, the difference between the position coordinates of the key point in the first posture image and the key point in the second posture image on the Y-axis, and the difference between the position coordinates of the key point in the first posture image and the key point in the second posture image on the Z-axis can be calculated respectively. Here, the difference between the position coordinates of each key point can be calculated simultaneously, or the difference between the position coordinates of all key points can be calculated sequentially. When the position coordinate differences between all the key points in the first posture image and the key points in the second posture image on the three axes are determined, for each axis, the position coordinate differences between all the key points in the first posture image and the key points in the second posture image on each axis can be averaged to obtain the position deviation of the second posture of the target object relative to the first posture on each axis. For example, the first posture image includes three key points: A (x1, y1, z1), B (x2, y2, z2), and C (x3, y3, z3); ​​the second posture image includes three key points: C (x4, y4, z4), D (x5, y5, z5), and E (x6, y6, z6), key point A corresponds to key point C, and the position coordinate deviation between key point A and key point C on the X-axis is calculated to be x4-x1, the position coordinate deviation on the Y-axis is y4-y1, and the position coordinate deviation on the Z-axis is calculated to be The position coordinate deviation is z4-z1; key point B corresponds to key point D, and the position coordinate deviation of key point B and key point D on the X axis is calculated to be x5-x2, the position coordinate deviation on the Y axis is y5-y2, and the position coordinate deviation on the Z axis is z5-z2; key point C corresponds to key point E, and the position coordinate deviation of key point C and key point E on the X axis is calculated to be x6-x3, the position coordinate deviation on the Y axis is y6-y3, and the position coordinate deviation on the Z axis is z6-z3. Finally, calculate the mean of the position coordinates of all key points on the X-axis: ((x4-x1)+(x5-x2)+(x6-x3)) / 3; calculate the mean of the position coordinates of all key points on the Y-axis: ((y4-y1)+(y5-y2)+(y6-y3)) / 3; calculate the mean of the position coordinates of all key points on the Z-axis: ((z4-z1)+(z5-z2)+(z6-z3)) / 3, and then use the mean of the position coordinates of these three axes as the positioning deviation on the three axes.

[0077] The method described in the above embodiment determines the positioning deviation by identifying the key points in the first posture image and the second posture image, which can accurately determine the positioning deviation and then accurately provide positioning guidance, and to a certain extent solves the repeatability problem of the target object in multiple positioning, thereby improving the accuracy of the projection dose of the later radiation delivery system.

[0078] In an exemplary embodiment, when the positioning deviation includes a position deviation and a direction deviation, another implementation method for determining the positioning information is provided, such as Figure 4 As shown, the method for determining the positioning information according to each first position coordinate and each second position coordinate includes:

[0079] S401, performing distance calculation on each first position coordinate and each corresponding second position coordinate to determine a key point position deviation between a key point corresponding to each first position coordinate and a key point corresponding to each second position coordinate.

[0080] Among them, the key point position deviation actually represents the distance deviation between two key points.

[0081] In the embodiment of the present application, the radiation delivery system can use the Euclidean distance calculation method or other distance calculation methods to calculate the distance between each first position coordinate and the corresponding second position coordinate, obtain the distance deviation between the key point corresponding to each first position coordinate and the key point corresponding to each second position coordinate, and use the distance deviation as the key point position deviation. For example, the first posture image includes three key points: A (x1, y1, z1), B (x2, y2, z2), C (x3, y3, z3); ​​the second posture image includes three key points: C (x4, y4, z4), D (x5, y5, z5), E (x6, y6, z6), then the following relationship (1) can be used to calculate the Euclidean distance D1 between key point A and key point C, the following relationship (2) can be used to calculate the Euclidean distance D2 between key point B and key point D, and the following relationship (3) can be used to calculate the Euclidean distance D3 between key point C and key point E:

[0082] (1);

[0083] (2);

[0084] (3);

[0085] S402, performing cosine similarity calculation on each first position coordinate and each corresponding second position coordinate to determine a key point direction deviation between a key point corresponding to each first position coordinate and a key point corresponding to each second position coordinate.

[0086] The key point direction deviation actually represents the direction deviation between two key points, for example, whether the directions of the two key points are consistent or opposite.

[0087] In the embodiment of the present application, the radiation delivery system can use the cosine similarity calculation method or other similarity calculation methods to calculate the angle between each first position coordinate and the corresponding second position coordinate, obtain the direction deviation between the key point corresponding to each first position coordinate and the key point corresponding to each second position coordinate, and use the direction deviation as the key point direction deviation. For example, the first posture image includes three key points: A (x1, y1, z1), B (x2, y2, z2), C (x3, y3, z3); ​​the second posture image includes three key points: C (x4, y4, z4), D (x5, y5, z5), E (x6, y6, z6), then the following relationship (4) can be used to calculate the direction deviation F1 between key point A and key point C, the direction deviation F2 between key point B and key point D, and the direction deviation F3 between key point C and key point E:

[0088]

[0089]

[0090]

[0091] (4) ;

[0092] in, Represents the position coordinates of key points in any first posture image; Represents the position coordinates of the key points in any second pose image.

[0093] S403: Determine positioning information according to the key point position deviation and the key point direction deviation.

[0094] In an embodiment of the present application, the positioning information may include the position deviation of each key point on the target object, and the directional deviation of each key point; when the radiation delivery system obtains the key point position deviation of the key points corresponding to all the first position coordinates based on the aforementioned steps, the key point position deviation of each key point can be used as the position deviation in the positioning information; when the radiation delivery system obtains the key point directional deviation of the key points corresponding to all the second position coordinates based on the aforementioned steps, the key point directional deviation of each key point can be used as the directional deviation in the positioning deviation.

[0095] The method described in the above embodiment evaluates the deviation between the initial posture and the successive postures of the target object from the two dimensions of distance and direction, and can comprehensively evaluate the posture changes of the target object, thereby accurately guiding the positioning of the target object and improving the positioning efficiency.

[0096] In an exemplary embodiment, a method for determining the positioning information according to the key point position deviation of the key point corresponding to the first position coordinate and the key point direction deviation corresponding to the second position coordinate is provided, that is, the above S403 "determine the positioning information according to the key point position deviation and the key point direction deviation", such as Figure 5 As shown, the method includes:

[0097] S501, performing a mean operation on all or part of the key point position deviations to determine an average position deviation of the second positioning posture relative to the first positioning posture.

[0098] In the embodiment of the present application, when the radiation delivery system obtains the key point position deviations of all key points or the key point position deviations of some key points based on the above steps, the key point position deviations of all key points can be averaged, or the key point position deviations of some key points can be averaged to obtain the average position deviation of the second positioning posture of the target object relative to the first positioning posture; for example, based on the above Figure 4 In an example of an embodiment, the key point position deviations D1, D2 and D3 of the three key points may be averaged, and the obtained distance average is the position deviation of the second positioning posture of the target object relative to the first positioning posture.

[0099] S502, performing a mean operation on all or part of the key point direction deviations to determine an average direction deviation of the second positioning posture relative to the first positioning posture.

[0100] In the embodiment of the present application, when the radiation delivery system obtains the key point direction deviations of all key points or the key point direction deviations of some key points based on the above steps, the key point direction deviations of all key points can be averaged, or the key point direction deviations of some key points can be averaged to obtain the direction deviation of the second position posture of the target object relative to the first position posture; for example, based on the above Figure 4 In an example of an embodiment, the key point direction deviations F1, F2 and F3 of the three key points may be averaged, and the obtained direction average is the direction deviation of the second positioning posture of the target object relative to the first positioning posture.

[0101] The method described in the above embodiment determines the position deviation of the second positioning posture of the target object relative to the first positioning posture by calculating the mean of the key point position deviations of all or part of the key points, and determines the direction deviation of the second positioning posture of the target object relative to the first positioning posture by calculating the mean of the key point direction deviations of all or part of the key points. This can achieve an overall evaluation of the deviations between different postures of the target object from two dimensions of distance and direction, and can comprehensively evaluate the posture changes of the target object, thereby accurately guiding the positioning of the target object and improving the positioning efficiency.

[0102] In an exemplary embodiment, the positioning information includes a depth deviation, based on which, Figure 4 The method described in the embodiment, such as Figure 6 As shown, it also includes:

[0103] S601, acquiring acquisition parameters for acquiring a second posture image and iris information of a target object.

[0104] The acquisition parameters include the field of view angle parameter and the image width; the iris information includes the actual size of the iris of the target object and the pixel size of the iris. The image acquisition device can be a device for capturing images, such as a camera or the like.

[0105] In the embodiment of the present application, when the radiation delivery system needs to calculate the depth deviation, the radiation delivery system can first obtain the configuration parameters of the image acquisition device used to acquire the second posture image, and obtain the acquisition parameters including the field angle parameter and the image width from the configuration parameters. In practical applications, the object positioning device and the image acquisition device in the radiation delivery system can be connected in communication, so when the object positioning device needs to acquire the acquisition parameters, it can read the acquisition parameters of the image acquisition device, and the image acquisition device can also transmit the acquisition parameters to the object positioning device. Correspondingly, the radiation delivery system can perform iris recognition based on the first posture image or the second posture image of the target object, detect the actual size of the iris of the target object and the pixel size of the iris, and obtain the iris information of the target object. Optionally, the radiation delivery system can also obtain the iris information of the target object from the case database corresponding to the target object; optionally, since the actual size and pixel size of the irises of different people are not much different, the iris information of the target object can be obtained from the standard indicator data. For example, in general, the actual size of the iris is 11.7mm.

[0106] S602: Determine a depth deviation of the second positioning posture relative to the first positioning posture according to the acquisition parameters and the iris information.

[0107] The embodiment of the present application relates to a specific method for calculating the depth deviation. When the radiation delivery system obtains the acquisition parameters of the image acquisition device and the iris information of the target object based on the above steps, the depth distance corresponding to the first posture image and the depth distance corresponding to the second posture image can be calculated using the following relationship (5), and then the depth deviation can be obtained by performing a difference operation on the two depth distances:

[0108]

[0109]

[0110] (5);

[0111] in, represents the field of view angle parameter, Indicates the field of view angle parameter after radian conversion. represents the image width of the image acquired by the image acquisition device, represents the focal length of the image acquisition device, Indicates the actual size of the iris. represents the pixel size of the iris, Indicates the depth distance.

[0112] The method described in the above embodiment also evaluates the deviation of the second positioning posture of the target object relative to the first positioning posture based on the depth distance between the image acquisition device and the target object, and more comprehensively evaluates the posture change of the target object, thereby accurately guiding the positioning of the target object and improving the positioning efficiency.

[0113] In an exemplary embodiment, during each positioning process, if the positioning information includes the positioning state and / or the positioning deviation, the positioning state and / or the positioning deviation may also be evaluated to determine whether the positioning state and / or the positioning deviation are correct or meet the positioning requirements. Based on this, Figure 2 The method described in the embodiment, such as Figure 7 As shown, it also includes:

[0114] S701, determine whether the positioning state and / or the positioning deviation meets the preset positioning requirement. If the positioning state and / or the positioning deviation meets the preset positioning requirement, execute step S702; if the positioning state and / or the positioning deviation does not meet the preset positioning requirement, execute step S703.

[0115] S702, performing identity authentication on the target object, and ending the current positioning if the identity authentication passes.

[0116] S703, output the positioning information, obtain the current posture image of the target object, and use the current posture image as a new second posture image, and return to execute the step of determining the first position of one or more key points of the target object in the first posture image and the second position of one or more key points in the second posture image.

[0117] Among them, the preset positioning requirement may include a requirement that the positioning deviation is less than a preset deviation threshold, or a requirement that the positioning state is consistent with the preset positioning state, which may be determined in advance according to actual evaluation requirements.

[0118] In the embodiment of the present application, when the radiation delivery system obtains the positioning information based on the above steps, it can further determine whether the positioning information meets the preset positioning requirements. If the positioning information meets the preset positioning requirements, it means that the deviation between the first positioning posture and the second positioning posture of the target object is very small and the repeatability is very strong, that is, it meets the positioning requirements. Therefore, the radiation delivery system does not need to instruct the target object to adjust the posture at this time, and can further use the face or iris detection method to authenticate the identity of the target object. Here, the face detection can be the radiation delivery system based on the face image of the target object to identify the identity information of the target object; the iris detection here can be the radiation delivery system based on the face image or posture image of the target object to identify the identity information of the target object. When the identity information of the target object is determined, the legitimacy or consistency of the target object can be authenticated according to the identity information, and when the identity authentication is passed, the current positioning is terminated, and then the radiation device in the radiation delivery system can be started to project the dose to the target target area of ​​the target object. If the positioning information does not meet the preset positioning requirements, the positioning information can be output, that is, the positioning information is output in the form of voice or display image to guide the target object to perform correct positioning. Then, when the target object adjusts its current posture according to the positioning information, the radiation delivery system can acquire the current posture image of the target object in real time, and use the current posture image as a new second posture image, and return to execute the step of "determining the first position of one or more key points of the target object in the first posture image and the second position of one or more key points in the second posture image" in S202, essentially detecting the current posture of the target object in real time, and then returning to the aforementioned steps to redetermine the positioning information until the current posture of the target object can meet the preset positioning requirements.

[0119] The method described in the above embodiment continuously detects the current posture of the target object in real time during the positioning process, and compares the initial posture of the target object, and automatically adjusts the posture of the target object in real time, thereby providing a method for guiding the positioning of the target object through a radiation delivery system, thereby improving the positioning efficiency and accuracy.

[0120] In an exemplary embodiment, a method for outputting positioning information is provided, the method comprising: outputting to a target object in the form of voice to guide the target object to perform positioning, wherein the positioning information can be converted into voice positioning information, and the voice positioning information can be broadcasted using a voice broadcaster in a radiation delivery system.

[0121] In the embodiment of the present application, this is a voice broadcast method, that is, when the positioning device in the radiation delivery system obtains the positioning information based on the aforementioned steps, since the positioning information is a text expression or professional terminology, in order to enable the target object to understand the voice broadcast, the positioning device can convert the positioning information into voice positioning information, that is, convert it into ordinary terminology, so as to clearly and efficiently guide the target object to perform positioning.

[0122] Optionally, another method of outputting positioning information is provided, which includes: outputting to the target object in the form of an image to guide the target object to position. Wherein, the positioning image is generated according to the positioning information, the first posture image and the second posture image, and the positioning image is displayed to the target object using a display in the radiation delivery system. Wherein, the positioning image includes a positioning indicator, and the positioning indicator includes a standard indicator and an actual indicator. The standard indicator indicates the first positioning posture (including a distance indicator and a direction indicator) of the target object in the first posture image, and the actual indicator indicates the second positioning posture (including a distance indicator and a direction indicator) of the target object in the second posture image. This voice broadcast method allows the target object to clearly obtain the information to be moved, and move according to the voice broadcast information, which can improve the positioning efficiency.

[0123] In the embodiment of the present application, this is an image broadcasting method, that is, when the object positioning device in the radiation delivery system obtains the positioning information based on the aforementioned steps, and obtains the positioning deviation, the first posture image, and the second posture image, it can identify the first positioning posture in the first posture image to determine the first positioning position of the target object, and then identify the second positioning posture in the second posture image to determine the second positioning position of the target object. Since the second posture image is the current posture image of the target object, the object positioning device can generate a standard indication mark and an actual indication mark based on the positioning information, the first positioning position of the target object, and the second positioning position, and display the standard indication mark and the actual indication mark on the second posture image, or obtain the current posture image of the target object in real time, and display the standard indication mark and the actual indication mark on the current posture image. For example, see Figure 8The schematic diagram of the positioning image shown in FIG. 1 includes a standard indicator 0' and an actual indicator 0. The standard indicator 0' includes a position indicator and a direction indicator; the actual indicator 0 includes a position indicator and a direction indicator. It should be noted that the positioning image may include standard indicators and actual indicators corresponding to multiple key points. Figure 8 Only one standard indicator and one actual indicator are displayed. This is just an example. There is no limit on the number of standard indicators and actual indicators. This image broadcasting method allows the target object to more intuitively observe the deviation of its position, and then adjust its posture according to the guidance of the picture, which can improve the efficiency of positioning.

[0124] In an exemplary embodiment, a method for positioning an object is also provided, namely Figure 2 The method described in the embodiment also includes the step of controlling the support platform in the radiation delivery system to drive the target object to move according to the instructions of the positioning information to complete the positioning.

[0125] In the embodiment of the present application, when the radiation delivery system obtains the positioning information based on the above steps, on the one hand, it can guide the target object to position based on the positioning information, and can also generate an instruction according to the positioning information, and send the instruction to the controller of the support platform connected to the positioning device. After receiving the instruction, the controller can control the support platform to move according to the instruction of the instruction, so as to assist the target object to automatically position. This method can improve the efficiency and accuracy of positioning by issuing instructions to the support platform to move the support platform to position the target object.

[0126] In an exemplary embodiment, a display interface of a display in an object positioning device is also provided, which can realize interaction with medical staff and operate the display of positioning images, see Fig. 9 The display interface shown includes an image display area, an information display area and an interactive area, wherein the image display area displays a positioning image, the information display area displays a positioning deviation, and the interactive area displays operation controls; the operation controls include at least one of an image acquisition selection control, a browsing control, an import video stream control, a positioning start control, a positioning indication mark switching control, and a save positioning control.

[0127] In the embodiment of the present application, the display interface of the display is oriented to the target object, so see Figure 1 Schematic diagram shown, wherein the device closest to the support platform 103 of the object positioning device 102 is the display, so that when the target object is located on the support platform 103, the display interface of the display 102 can be intuitively seen for positioning. Fig. 9A schematic diagram of the display interface, wherein the display interface displays a positioning image, which is acquired in real time and is an image with a positioning indication mark added in real time. When the image acquisition selection control is triggered by the user, it is used to select the corresponding image acquisition device. For example, a radiation delivery system is provided with multiple cameras. When the user triggers the image acquisition selection control, the corresponding camera can be selected to acquire images. When the browse control is triggered by the user, it is used to acquire any second posture image or the first first posture image of the target object from the image database, and display it in the image display area for the user to browse and view. When the import video stream control is triggered by the user, the image collection collected by other image acquisition devices is used to acquire the first posture image or the second posture image of the target object. Here, other image acquisition devices refer to image acquisition devices on non-radiation delivery systems. For example, it can be a camera in the room where the radiation delivery system is located, and the camera can also acquire the first posture image or the second posture image of the target object. When the positioning start control is triggered by the user, it is used to display the positioning image in the image display area. When the positioning indicator switch control is triggered by the user, it is used to switch the display of the positioning indicator on the positioning image, for example, only displaying the standard indicator or the actual indicator on the positioning image, or displaying both indicators. Optionally, when the positioning indicator switch control is triggered by the user, it can also be used to switch the display of the indicator containing the position deviation on the positioning image or not, for example, Fig. 9 The skeleton view switching control in refers to switching to display or not displaying the mark containing the position deviation; and is also used to switch to display or not display the indicator mark containing the direction deviation on the positioning image, for example, Fig. 9 The direction view switching control in refers to switching the display or non-display of the indicator containing the direction deviation. When the save position control is triggered by the user, it is used to save the current posture image of the target object as the current sub-posture image.

[0128] The method described in the above embodiment provides an interactive interface, so that the medical staff can interact with the radiation delivery system to assist the target object in positioning during the automatic positioning of the target object according to the guidance of the radiation delivery system, thereby enhancing the intelligence of the entire radiation delivery system. In addition, in addition to the image guiding the positioning, the display interface also displays information about the positioning deviation to the target object, so that the target object can more intuitively and clearly understand his current posture and the position and direction to be moved, thereby improving the positioning efficiency and accuracy.

[0129] Based on the methods described in all the above embodiments, a positioning method is provided, such as Fig.10 As shown, the method includes:

[0130] S801, obtaining a first posture image of a target object in a first posture at a first position.

[0131] S802: Identify one or more key points of the target object in the first posture image, obtain first position coordinates of each key point, and save them.

[0132] S803, after the first positioning, obtaining a second posture image of the target object in any second positioning, and identifying one or more key points in the second posture image to obtain second position coordinates of each key point.

[0133] S804, performing distance calculation on each first position coordinate and the corresponding second position coordinate, determining the key point position deviation between each key point in the first posture image and each key point in the second posture image, and performing mean operation on all or part of the key point position deviations, to determine the position deviation of the second positioning posture of the target object relative to the first positioning posture.

[0134] S805, calculate the angles of each first position coordinate and the corresponding second position coordinate, determine the key point direction deviation between each key point in the first posture image and each key point in the second posture image, and perform a mean operation on all or part of the key point direction deviations to determine the direction deviation of the second posture of the target object relative to the first posture.

[0135] S806, acquiring acquisition parameters of an image acquisition device for acquiring a second posture image and iris information of the target object, and determining a depth deviation of the second posture of the target object relative to the first posture according to the acquisition parameters and the iris information.

[0136] S807, determine whether the position deviation, direction deviation and depth deviation meet the preset positioning requirements; if they meet the preset positioning requirements, execute step S808; if they do not meet the preset positioning requirements, execute step S809.

[0137] S808, performing identity authentication on the target object, and ending the current positioning if the identity authentication passes.

[0138] S809, outputting the positioning information, reacquiring the current posture image of the target object, and reusing the current posture image as the sub-posture image, and returning to execute step S803.

[0139] The above steps have been explained above. Please refer to the above instructions for details and will not be repeated here.

[0140] In an exemplary embodiment, based on the positioning methods described in all the above embodiments, another radiation delivery system is provided, which can be used for radiation processing, radiation verification using a phantom before radiotherapy, etc. In the embodiment of the present application, the radiation delivery system is used to guide the initial positioning of the target object during or before the formulation of the radiation delivery plan, and to guide the delivery positioning of the target object after the formulation of the radiation delivery plan and during or before the radiation delivery. Fig.11 As shown, the radiation delivery system includes: at least one image acquisition device 101, an object positioning device 102 and a support platform 103. The image acquisition device 101 is used to acquire a first posture image of the target object in the first positioning on the support platform 103, and a second posture image of the target object in any second positioning on the support platform 103. The object positioning device 102 is used to identify key points of the first posture image and the second posture image, determine the positioning deviation of the second positioning posture of the target object relative to the first positioning posture, and determine and output positioning information according to the positioning deviation.

[0141] The radiation delivery system described in the embodiment of the present application is applied to the aforementioned Figures 2 to 10 For the method described in any embodiment, please refer to the contents described in the aforementioned embodiment for the specific method implementation, which will not be repeated here.

[0142] In an exemplary embodiment, Fig.11 The object positioning device 102 shown also includes a processing component 1021, a display 1022 and / or a voice announcer 1023. The processing component 1021 is used to perform key point recognition on the first posture image and the second posture image, determine the position coordinates of the first position of the key point in the first posture image, and the position coordinates of the second position of the key point in the second posture image, so as to obtain the positioning information of the second positioning posture of the target object relative to the first positioning posture; the voice announcer 1023 is used to convert the positioning information into voice positioning information and announce the voice positioning information; the processing component 1021 is also used to generate a positioning image according to the positioning information, the first posture image and the second posture image; the positioning image includes a positioning indication mark; the display 1022 is used to display the positioning image to the target object.

[0143] The methods of using each component or module in the above-mentioned radiation delivery system have been described above. Please refer to the above description for details and will not be repeated here.

[0144] In an exemplary embodiment, Fig.11 The radiation delivery device shown also includes: a rack 104, the image acquisition device 101 is arranged on the rack 104, and the support platform 103 is located within the shooting range of the image acquisition device 101. The rack 104 can be a rack of a ring structure ( Fig.111041 in the schematic diagram shown), or a non-annular frame (such as Fig.11 1042 in the schematic diagram). Fig.11 The rack 104 in the figure includes two parts. One part is the rack (1041) connected to the support platform 103, which is a ring-shaped structure and on which a treatment component 105 is arranged. After the target object is positioned, the support platform 103 drives the target object to move into the ring of the rack (1041), and the treatment component 105 is started to project a dose to the target object. The other part is a rack (1042) located near the support platform 103, on which an image acquisition device 101 is arranged to acquire the posture image of the target object in real time. If the positioning device includes a display, the rack (1042) is also provided with a display 102, and its display screen faces the target object so that the target object can position itself based on the instructions on the screen. If the positioning device includes a voice announcer, a voice announcer can also be arranged on the rack 1042 to voice-broadcast the positioning information to the target object so that the target object can clearly hear the voice announcement at a close distance to position itself. It should be noted that the image acquisition device 101 can also be arranged on the ring-shaped rack ( Fig.11 1042 ) in the support platform 103 to facilitate image acquisition of the target object.

[0145] The methods of using each component or module in the above-mentioned radiation delivery system have been described above. Please refer to the above description for details and will not be repeated here.

[0146] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0147] Based on the same inventive concept, the embodiment of the present application also provides a positioning device for implementing the positioning method involved above. The implementation solution provided by the device to solve the problem is similar to the implementation solution recorded in the above method, so the specific limitations in one or more positioning device embodiments provided below can refer to the limitations on the positioning method above, and will not be repeated here.

[0148] In an exemplary embodiment, Fig.12As shown, an object positioning device is provided, comprising:

[0149] The acquisition module 121 is used for a first posture image of a target object in a first position and a second posture image of the target object in a second position after the first position.

[0150] A first determination module 122, configured to determine a first position of one or more key points of the target object in the first posture image and a second position of the one or more key points in the second posture image;

[0151] The second determining module 123 is configured to determine the position information of the target object according to the first position and the second position.

[0152] In an exemplary embodiment, the second determination module 123 is used to determine, based on the first position and the second position, a positioning state and / or a positioning deviation of the second positioning posture of the target object relative to the first positioning posture as the positioning information.

[0153] In an exemplary embodiment, the first determining module 122 includes:

[0154] A first recognition unit, used to recognize one or more key points in the first posture image to obtain first position coordinates of each of the key points;

[0155] A second recognition unit is used to identify one or more key points in the second posture image to obtain second position coordinates of each key point; wherein the number and type of key points in the first posture image are consistent with those in the second posture image, and each first position coordinate and each second position coordinate are used to determine the positioning information.

[0156] In an exemplary embodiment, the object positioning device further includes:

[0157] A first calculation module, used for performing distance calculation between each of the first position coordinates and the corresponding second position coordinates, and determining a key point position deviation between a key point corresponding to each of the first position coordinates and a key point corresponding to each of the second position coordinates;

[0158] A second calculation module, used for performing angle calculation on each of the first position coordinates and the corresponding second position coordinates, and determining a key point direction deviation between a key point corresponding to each of the first position coordinates and a key point corresponding to each of the second position coordinates;

[0159] The first positioning module is used to determine the positioning information according to the key point position deviation and the key point direction deviation.

[0160] In an exemplary embodiment, the above-mentioned determination subunit is specifically used to perform a mean operation on all or part of the key point position deviations to determine the average position deviation of the second positioning posture relative to the first positioning posture, and perform a mean operation on all or part of the key point direction deviations to determine the average direction deviation of the second positioning posture relative to the first positioning posture.

[0161] In an exemplary embodiment, the above-mentioned object positioning device further includes:

[0162] An information acquisition module, used to acquire acquisition parameters for acquiring the second posture image and iris information of the target object; the acquisition parameters include field of view angle parameters and image width; the iris information includes the actual size of the iris of the target object and the pixel size of the iris;

[0163] The second positioning module is used to determine a depth deviation of the second positioning posture relative to the first positioning posture according to the acquisition parameters and the iris information.

[0164] Each module in the above-mentioned object positioning device can be implemented in whole or in part by software, hardware or a combination thereof. Each module can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute the operations corresponding to each module.

[0165] In an exemplary embodiment, a radiation delivery device is provided. The radiation delivery device may be a terminal, and its internal structure diagram may be as shown in FIG. Fig.13As shown. The radiation delivery device includes a processor, a memory, an input / output interface, a communication interface, a display unit and an input device. The processor, the memory and the input / output interface are connected through a system bus, and the communication interface, the display unit and the input device are connected to the system bus through the input / output interface. The processor of the radiation delivery device is used to provide computing and control capabilities. The memory of the radiation delivery device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The input / output interface of the radiation delivery device is used to exchange information between the processor and an external device. The communication interface of the radiation delivery device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be implemented through WIFI, a mobile cellular network, near field communication (NFC) or other technologies. When the computer program is executed by the processor, a positioning method is implemented. The display unit of the radiation delivery device is used to form a visually visible picture, which can be a display screen, a projection device or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen, and the input device of the radiation delivery device can be a touch layer covering the display screen, or a button, trackball or touchpad set on the shell of the radiation delivery device, or an external keyboard, touchpad or mouse.

[0166] Those skilled in the art will understand that Fig.13 The structure shown in the figure is merely a block diagram of a portion of the structure related to the present application scheme, and does not constitute a limitation on the radiation delivery device to which the present application scheme is applied. The specific radiation delivery device may include more or fewer components than shown in the figure, or combine certain components, or have a different arrangement of components.

[0167] In an exemplary embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, the following steps are implemented:

[0168] Acquire a first posture image of a first position posture of a target object in a first position, and a second posture image of a second position posture of the target object in a second position after the first position;

[0169] Determining first positions of one or more key points of the target object in the first pose image and second positions of the one or more key points in the second pose image;

[0170] The positioning information of the target object is determined according to the first position and the second position.

[0171] The computer device provided in the above embodiment has an implementation principle and technical effects similar to those of the above method embodiment, which will not be described in detail here.

[0172] In one embodiment, a computer readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the following steps are implemented:

[0173] Acquire a first posture image of a first position posture of a target object in a first position, and a second posture image of a second position posture of the target object in a second position after the first position;

[0174] Determining first positions of one or more key points of the target object in the first pose image and second positions of the one or more key points in the second pose image;

[0175] The positioning information of the target object is determined according to the first position and the second position.

[0176] The above embodiment provides a computer-readable storage medium, whose implementation principle and technical effect are similar to those of the above method embodiment, and will not be repeated here.

[0177] In one embodiment, a computer program product is provided, comprising a computer program, which, when executed by a processor, implements the following steps:

[0178] Acquire a first posture image of a first position posture of a target object in a first position, and a second posture image of a second position posture of the target object in a second position after the first position;

[0179] Determining first positions of one or more key points of the target object in the first pose image and second positions of the one or more key points in the second pose image;

[0180] The positioning information of the target object is determined according to the first position and the second position.

[0181] The above embodiment provides a computer program product, whose implementation principle and technical effect are similar to those of the above method embodiment, and will not be repeated here.

[0182] A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, an artificial intelligence (AI) processor, etc., but are not limited to this.

[0183] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0184] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. An object positioning method, characterized in that: The method comprises: Acquire a first posture image of a first position posture of a target object in a first position, and a second posture image of a second position posture of the target object in a second position after the first position; Determining first positions of one or more key points of the target object in the first pose image and second positions of the one or more key points in the second pose image; The positioning information of the target object is determined according to the first position and the second position.

2. The object positioning method according to claim 1, characterized in that: The determining the positioning information of the target object according to the first position and the second position includes: According to the first position and the second position, a positioning state and / or a positioning deviation of the second positioning posture of the target object relative to the first positioning posture is determined as positioning information.

3. The object positioning method according to claim 1, characterized in that: The positioning information is output to the target object in the form of voice and / or image, so as to guide the target object to perform positioning.

4. The object positioning method according to claim 1, characterized in that: The determining of a first position of one or more key points of the target object in the first posture image and a second position of the one or more key points in the second posture image comprises: Identify one or more key points in the first posture image to obtain first position coordinates of each of the key points; One or more key points in the second posture image are identified to obtain second position coordinates of each of the key points; wherein the number and type of key points in the first posture image are consistent with those in the second posture image, and each of the first position coordinates and each of the second position coordinates are used to determine the positioning information.

5. The object positioning method according to claim 4, characterized in that: The positioning information includes a position deviation and a direction deviation. The method for determining the positioning information according to each of the first position coordinates and each of the second position coordinates includes: Performing distance calculation on each of the first position coordinates and the corresponding second position coordinates to determine a key point position deviation between a key point corresponding to each of the first position coordinates and a key point corresponding to each of the second position coordinates; Calculating the angles between each of the first position coordinates and each of the corresponding second position coordinates to determine a key point direction deviation between a key point corresponding to each of the first position coordinates and a key point corresponding to each of the second position coordinates; The positioning information is determined according to the key point position deviation and the key point direction deviation.

6. The object positioning method according to claim 5, characterized in that: The determining the positioning information according to the key point position deviation and the key point direction deviation includes: Performing a mean operation on all or part of the key point position deviations to determine an average position deviation of the second positioning posture relative to the first positioning posture; Perform a mean operation on all or part of the key point directional deviations to determine an average directional deviation of the second positioning posture relative to the first positioning posture.

7. The object positioning method according to claim 1, characterized in that: The positioning information includes a depth deviation, and the object positioning method further includes: Acquire acquisition parameters for acquiring the second posture image and iris information of the target object; A depth deviation of the second positioning posture relative to the first positioning posture is determined according to the acquisition parameter and the iris information.

8. The object positioning method according to claim 1, characterized in that: The first position is an initial position of the target object during or before radiation delivery planning, and the second position is a delivery position after the radiation delivery planning is formulated and during or before radiation delivery.

9. A radiation delivery system, characterized in that The radiation delivery system includes: at least one image acquisition device, a positioning device, and a support platform; The image acquisition device is used to acquire a first posture image of a first positioning posture of the target object when the target object is in a first positioning on the support platform, and a second posture image of a second positioning posture of the target object when the target object is in a second positioning after the first positioning on the support platform; The positioning device is used to determine a first position of one or more key points of the target object in the first posture image and a second position of the one or more key points in the second posture image, and determine positioning information according to the first position and the second position.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 8 are implemented.

Images