Limb volume detection method, device and equipment and storage medium

Abstract

The embodiment of the invention discloses a limb volume detection method, device and equipment and a storage medium, and is used for medical detection equipment which is at least provided with four cameras, the cameras are annularly arranged in a surrounding manner according to preset intervals, lenses of the cameras face the annular central area, and when it is detected that a to-be-detected limb penetrates through the annular interior defined by the cameras, the to-be-detected limb is detected to pass through the annular central area. Controlling the camera to carry out mobile scanning on the limb to be detected, and carrying out deduplication fusion on scanned point cloud data to obtain fused point cloud data; performing three-dimensional modeling according to the fused point cloud data to obtain a target three-dimensional limb model; performing slicing and gravity center analysis on the target three-dimensional limb model to determine a joint gravity center point; re-slicing according to the joint gravity center point to obtain target slice data; integrating the volume information of each target slice data to obtain overall volume information; the problems that existing limb volume measurement is low in working efficiency and large in error can be solved, the limb volume can be automatically scanned and recognized, and the working efficiency and accuracy of limb volume measurement are improved.

Description

Technical Field

[0001] The embodiments of the present application relate to the technical field of limb volume measurement, and in particular, to a limb volume detection method, device, equipment and storage medium. Background Art

[0002] Edema refers to the accumulation of excessive body fluid in the interstitial space outside blood vessels, which is a common pathological process. Edema usually indicates some internal diseases, such as hypertension, diabetes and congestive heart disease, so the detection of edema is of great clinical significance. At present, the presence of edema can be determined by measuring the volume information of the limbs and comparing the changes in the volume information.

[0003] The existing method of measuring edema mainly relies on medical staff to manually measure with a tape measure. Manual measurement is not only inefficient, but also due to the different tightness of the tape measure during each measurement, the corresponding error in judging whether there is edema is also relatively large. Summary of the invention

[0004] The embodiments of the present application provide a limb volume detection method, device, equipment and storage medium, which can solve the problems of low work efficiency and large errors in the existing limb volume measurement process, can automatically scan and identify limb volume, improve the work efficiency of limb volume measurement, and improve the accuracy of limb volume detection.

[0005] In a first aspect, an embodiment of the present application provides a limb volume detection method for a medical detection device, wherein the medical detection device is configured with at least four cameras, the cameras are arranged in a ring at preset intervals, the camera lenses are directed toward the central area of ​​the ring, and the ring is used for the limb to be detected to pass through, and the limb volume detection method specifically includes:

[0006] When it is detected that the limb to be tested passes through the ring formed by the cameras, all cameras are controlled to move along the limb to be tested and scan simultaneously, so as to obtain point cloud data scanned by multiple cameras;

[0007] The point cloud data is deduplicated and fused according to the preset spatial calibration relationship to obtain the fused point cloud data of all cameras. The preset spatial calibration relationship is the relative position relationship between every two cameras.

[0008] Perform three-dimensional modeling processing based on the fused point cloud data to obtain a target three-dimensional limb model;

[0009] Slice the target three-dimensional limb model and perform gravity center analysis to determine the joint gravity center;

[0010] Re-slicing the target three-dimensional limb model according to the joint center of gravity to obtain target slice data;

[0011] Perform parameter calculation processing according to the target slice data to obtain volume information of each target slice data, the volume information includes the thickness and cross-sectional area of ​​the slice;

[0012] The volume parameters of each target slice data are integrated and processed to obtain the overall volume information of the limb to be measured.

[0013] In one embodiment, the point cloud data is deduplicated and fused according to a preset spatial calibration relationship to obtain fused point cloud data of all cameras. The preset spatial calibration relationship is the relative position relationship between every two calibrated cameras, including:

[0014] Determine the overlapping area in the area corresponding to the point cloud data according to the preset spatial calibration relationship;

[0015] The point cloud data in the overlapping area is deduplicated to obtain fused point cloud data, which includes the point cloud data in the non-overlapping area and the deduplicated point cloud data in the overlapping area.

[0016] In one embodiment, three-dimensional modeling is performed based on the fused point cloud data to obtain a target three-dimensional limb model, including:

[0017] Create an initial voxel grid based on the resolution of the fused point cloud data;

[0018] voxelize the fused point cloud data according to the initial voxel grid to obtain first voxel data, where the first voxel data includes a vertex, a voxel index and a voxel value corresponding to each voxel;

[0019] The first voxel data is mapped to the point cloud grid, and color rendering is performed according to the mapping result and the projection superposition result to generate a target three-dimensional limb model.

[0020] In one embodiment, slicing and center of gravity analysis are performed on the target three-dimensional limb model to determine the center of gravity of the joint, including:

[0021] Slice the target three-dimensional limb model along a preset direction and at a preset distance to obtain reference slice data;

[0022] Calculate and determine the centroid of each reference slice data;

[0023] Connect the centroid of the first reference slice data and the centroid of the last reference slice data to obtain a first reference line;

[0024] The center of gravity of all reference slice data that is farthest from the first reference line is determined as the joint center of gravity.

[0025] In one embodiment, the target three-dimensional limb model is re-sliced ​​according to the joint center of gravity to obtain target slice data, including:

[0026] Connect the center of gravity of the first reference slice data and the center of gravity of the joint to obtain a second reference line; re-slice the first part of the target three-dimensional limb model along the second reference line at a preset interval to obtain first target slice data, wherein the first part is the part between the first point cloud of the first reference slice data and the last point cloud in the reference slice data corresponding to the center of gravity of the joint;

[0027] Connect the joint center of gravity with the center of gravity of the last reference slice data to obtain a third reference line;

[0028] The second part of the target three-dimensional limb model is re-sliced ​​along the third reference line at a preset interval to obtain second target slice data, wherein the second part is a part between the first point cloud in the reference slice data corresponding to the joint center of gravity point and the last point cloud in the last reference slice data;

[0029] The first target slice data and the second target slice data are fused and spliced ​​with the joint center point as the splicing point to obtain the target slice data of the entire target three-dimensional limb model.

[0030] In one embodiment, the target three-dimensional limb model is sliced ​​and subjected to center of gravity analysis, and before the joint center of gravity is determined, the following steps are included:

[0031] Scanning and three-dimensional modeling are performed twice on the limb to be tested to obtain a corresponding first three-dimensional limb model and a second three-dimensional limb model;

[0032] Slice the first three-dimensional limb model and the second three-dimensional limb model along a preset direction and at a preset interval to obtain first slice data corresponding to the first three-dimensional limb model and second slice data corresponding to the second three-dimensional limb model;

[0033] Performing gravity point calculation processing on the first slice data to obtain first gravity point data, the first gravity point data including gravity points of all the first slice data;

[0034] Performing gravity point calculation processing on the second slice data to obtain second gravity point data, the second gravity point data including gravity points of all the second slice data;

[0035] Mapping the first center of gravity point data and the second center of gravity point data to the same plane for curve drawing, and obtaining a first curve and a second curve correspondingly;

[0036] Performing distance calculation processing according to the first curve and the second curve to obtain the Euclidean distance between the first curve and the second curve;

[0037] When the Euclidean distance is below a preset value, determining the first three-dimensional limb model as a target three-dimensional limb model;

[0038] When the Euclidean distance exceeds a preset value, the limb to be measured is re-scanned and three-dimensionally modeled until the Euclidean distance is below the preset value, and the corresponding first three-dimensional limb model is determined as the target three-dimensional limb model.

[0039] In one embodiment, the limb volume detection method further includes:

[0040] When it is detected that the calibration object passes through the ring formed by the cameras, all cameras are controlled to move along the calibration object and scan at the same time to obtain the reference point cloud data;

[0041] Compare the reference point cloud data with the standard point cloud data to determine the first position relationship of each camera relative to the calibration object, where the standard point cloud data is the standard point cloud data after point cloud calibration and correction of the calibration object;

[0042] The spatial position relationship of the cameras is grouped and calibrated according to the first position relationship to determine a second position relationship between the cameras in the corresponding group;

[0043] Performing inter-group registration processing on the spatial position relationship of the cameras according to the first position relationship and the second position relationship to obtain a third relative position relationship between every two groups;

[0044] The spatial positional relationships of the cameras are calibrated in pairs according to the first positional relationship, the second positional relationship and the third relative positional relationship to obtain a preset spatial calibration relationship of all the cameras.

[0045] In a second aspect, an embodiment of the present application provides a limb volume detection device for use in a medical detection device, wherein the medical detection device is configured with at least four cameras, the cameras are arranged in a ring at preset intervals, the camera lenses are directed toward a central area of ​​the ring, and the ring is used for a limb to be detected to pass through, the device comprising:

[0046] The scanning module is used to control all cameras to move along the limb to be tested and scan simultaneously when it is detected that the limb to be tested passes through the ring formed by the cameras, so as to obtain point cloud data scanned by multiple cameras;

[0047] The point cloud fusion module is used to perform deduplication and fusion processing on the point cloud data according to the preset spatial calibration relationship to obtain the fused point cloud data of all cameras. The preset spatial calibration relationship is the relative position relationship between every two cameras.

[0048] A modeling module is used to perform three-dimensional modeling processing based on the fused point cloud data to obtain a target three-dimensional limb model;

[0049] A joint determination module is used to perform slicing processing and center of gravity analysis processing according to the target three-dimensional limb model to determine the center of gravity of the joint;

[0050] A re-slicing module is used to re-slice the target three-dimensional limb model according to the joint center of gravity to obtain target slice data;

[0051] A first calculation module is used to perform parameter calculation processing according to the target slice data to obtain volume information of each target slice data, wherein the volume information includes the thickness and cross-sectional area of ​​the slice;

[0052] The second calculation module is used to integrate and process the volume parameters of each target slice data to obtain the overall volume information of the limb to be measured.

[0053] In a third aspect, an embodiment of the present application provides a limb volume detection device, comprising:

[0054] memory and one or more processors;

[0055] The memory is used to store one or more programs;

[0056] When the one or more programs are executed by the one or more processors, the one or more processors implement the limb volume detection method as described in the first aspect.

[0057] In a fourth aspect, an embodiment of the present application provides a storage medium storing computer executable instructions, wherein the computer executable instructions, when executed by a computer processor, are used to perform the limb volume detection method as described in the first aspect.

[0058] The embodiment of the present application is used for a medical detection device equipped with at least four cameras, the cameras are arranged in a ring at preset intervals, the lenses of the cameras face the central area of ​​the ring, the ring is used for the limb to be tested to pass through, when it is detected that the limb to be tested passes through the ring formed by the cameras, all cameras are controlled to move and scan along the limb to be tested at the same time, de-duplication fusion and three-dimensional modeling are performed based on the scanned point cloud data to obtain a target three-dimensional limb model, slice processing and center of gravity analysis processing are performed according to the target three-dimensional limb model to determine the joint center of gravity point, the target three-dimensional limb model is re-sliced ​​based on the joint center of gravity point to obtain target slice data, and parameter calculation and data integration are performed based on the target slice data to obtain the overall volume information of the limb to be tested. By adopting the above technical means, the overall volume information of the limb to be tested can be obtained by scanning through the medical detection equipment, thereby avoiding the technical problems of low work efficiency and large errors caused by manual measurement, and the overall volume information of the limb to be tested can be obtained based on the camera scanning, thereby improving the work efficiency of limb volume detection; and firstly preliminarily slice to determine the joint center of gravity point, re-slice based on the joint center of gravity point, and slice separately with the joint center of gravity point as the boundary, thereby improving the accuracy of slicing, and then improving the accuracy of volume calculation. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 This is a schematic diagram of a single group of camera distribution provided in an embodiment of the present application;

[0060] Figure 2 It is a schematic diagram of the relationship between the position of a camera and a limb provided in an embodiment of the present application;

[0061] Figure 3 is a flow chart of a limb volume detection method provided in an embodiment of the present application;

[0062] Figure 4 is a schematic diagram of the position relationship between a camera and a calibration object provided in an embodiment of the present application;

[0063] Figure 5 is a first schematic diagram of a limb model slice provided in an embodiment of the present application;

[0064] Figure 6 is a second schematic diagram of a limb model slice provided in an embodiment of the present application;

[0065] Figure 7 is a structural schematic diagram of a limb volume detection device provided in an embodiment of the present application;

[0066] Figure 8 It is a structural schematic diagram of a limb volume detection device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0067] In order to make the purpose, technical scheme and advantages of the present application clearer, the specific embodiments of the present application are further described in detail below in conjunction with the accompanying drawings. It is understood that the specific embodiments described herein are only used to explain the present application, rather than to limit the present application. It should also be noted that, for the convenience of description, only the part related to the present application but not all the contents are shown in the accompanying drawings. Before discussing the exemplary embodiments in more detail, it should be mentioned that some exemplary embodiments are described as processes or methods depicted as flow charts. Although the flow chart describes each operation (or step) as a sequential process, many of the operations therein can be implemented in parallel, concurrently or simultaneously. In addition, the order of each operation can be rearranged. The process can be terminated when its operation is completed, but it can also have additional steps not included in the accompanying drawings. The process can correspond to a method, a function, a procedure, a subroutine, a subprogram, etc.

[0068] Edema refers to the accumulation of excessive body fluid in the interstitial space outside blood vessels, which is a common pathological process. Edema usually indicates some internal diseases, such as hypertension, diabetes, and congestive heart disease, so the detection of edema is of great clinical significance. At present, the volume information of the limbs can be measured and the presence of edema can be determined by comparing the changes in the volume information. The existing method of measuring edema mainly relies on medical staff to manually measure with a tape measure. Manual measurement is not only inefficient, but also based on the different tightness of the tape measure during each measurement, the corresponding error in judging whether there is edema is also relatively large.

[0069] There are also three-dimensional scanners that can be used to detect limb volume. However, the existing method is to slice directly along the axial direction of the three-dimensional scanner. However, due to the bending of the tester's limbs, taking the arm as an example, when measuring, the forearm and the upper arm are not in a straight line, or the arm is not placed in a parallel direction with the axial direction of the three-dimensional scanner. At this time, there is a certain error in measuring the slice parameters when slicing along the axial direction of the three-dimensional scanner, resulting in poor accuracy of the measurement results.

[0070] Based on this, a limb volume detection method of an embodiment of the present application is provided to solve the technical problems of low work efficiency and large errors in the existing limb volume detection process. The limb volume detection method provided by the present application can obtain the overall volume information of the limb to be measured by scanning with medical detection equipment, thereby avoiding the technical problems of low work efficiency and large errors caused by manual measurement. The overall volume information of the limb to be measured can be obtained based on camera scanning, which improves the work efficiency of limb volume detection compared with the existing manual measurement method; and firstly perform preliminary slicing to determine the center of gravity of the joint, re-slice based on the center of gravity of the joint, and slice separately with the center of gravity of the joint as the boundary, which improves the accuracy of slicing compared with the existing method of directly slicing according to the axial direction of the scanner, thereby improving the accuracy of volume calculation.

[0071] The limb volume detection method provided in the embodiment of the present application is used for medical detection equipment, which is equipped with at least four cameras, which are arranged in a ring at preset intervals, and the cameras in each ring form a camera group. Each camera is a binocular camera. Figure 1 This is a schematic diagram of a single camera group distribution provided in an embodiment of the present application, referring to Figure 1 The medical detection device is equipped with four cameras 10, and each four cameras form a camera group. Taking one of the camera groups as an example, the four cameras 10 in the group are arranged in a ring at a preset interval, and the lenses of the four cameras 10 are oriented toward the central area of ​​the ring H1, and the ring H1 is used for the limb to be tested to pass through. The medical detection device provided in this embodiment is provided with at least one camera group, and at least four cameras 10 are arranged in each camera group. Figure 2is a schematic diagram of the relationship between the camera and the limb position provided in an embodiment of the present application, referring to Figure 2 , taking two camera groups in a medical detection device as an example, the two camera groups are camera group 1 and camera group 2, and camera group 1 and camera group 2 are respectively as follows Figure 1 The four cameras 10 in the group shown are arranged in a ring at a preset interval, and the lenses of the four cameras 10 face the central area of ​​the ring H1, and the ring H1 is used for the limb to be tested to pass through. It should be noted that Figure 2 In the example, the limb to be tested is an arm. In practice, the limb to be tested may be other limbs besides an arm, such as a leg. Figure 2 As shown, camera group 1 and camera group 2 are arranged at a preset interval. The limb to be tested passes through the inside of the ring H1 surrounded by the cameras 10 of each camera group, approximately at the center line X position inside the ring H1. When it is detected that the limb to be tested passes through the inside of the ring H1 surrounded by the cameras 10, all cameras 10 are controlled to move along the limb to be tested and scan at the same time.

[0072] Figure 3 A flow chart of a limb volume detection method provided in an embodiment of the present application is given. The limb volume detection method provided in this embodiment can be executed by a limb volume detection device, which can be implemented by software and / or hardware. The limb volume detection device can be composed of two or more physical entities, or can be composed of one physical entity. Generally speaking, the limb volume detection device can be a medical detection device, such as a three-dimensional scanner.

[0073] The following description is made by taking a medical detection device as an example to implement the limb volume detection method. Figure 3 The limb volume detection method is used for medical detection equipment, which is equipped with at least four cameras, which are arranged in a ring at preset intervals, and the camera lenses are facing the central area of ​​the ring, and the ring is used for the limb to be detected to pass through. The limb volume detection method specifically includes:

[0074] S101. When it is detected that the limb to be measured passes through the ring formed by the cameras, all cameras are controlled to move along the limb to be measured and scan simultaneously, so as to obtain point cloud data scanned by multiple cameras.

[0075] The medical detection device provided in this embodiment is as follows Figure 1 and Figure 2 As shown, the medical detection device includes at least one camera group, and the four cameras 10 in each camera group are arranged in a ring at a preset interval, and the lenses of the four cameras 10 are oriented toward the central area of ​​the ring H1, and the ring H1 is used for the limb to be detected to pass through. When it is detected that the limb to be detected passes through the ring H1 surrounded by the cameras 10, for example Figure 2As shown, all cameras 10 are controlled to move and scan along the limb to be tested, and point cloud data scanned by multiple cameras 10 are obtained. It should be noted that, based on the visual overlap of the cameras 10, the point cloud data scanned by multiple cameras 10 also contain corresponding overlapping point cloud data.

[0076] As described above, when it is detected that a limb to be measured is inside the circle surrounded by the cameras, all cameras are controlled to move and scan along the limb to be measured at the same time to obtain the scanned point cloud data. Subsequently, three-dimensional modeling can be performed based on the point cloud data, and the limb volume can be detected based on the modeled three-dimensional limb model. Compared with the existing method of measuring with a soft tape, the method of this embodiment based on the three-dimensional limb model obtained by camera scanning has a higher accuracy rate, and automatically realizes scanning and volume measurement, thereby improving the degree of automation of limb volume measurement, and further improving the work efficiency of limb volume measurement.

[0077] S102, performing deduplication fusion processing on the point cloud data according to a preset spatial calibration relationship to obtain fused point cloud data of all cameras, wherein the preset spatial calibration relationship is the relative position relationship between every two cameras.

[0078] Before conducting a formal test, it is necessary to calibrate the spatial position of the camera to obtain the preset spatial calibration relationship of all cameras, that is, the relative position relationship between the cameras. When the formal test is subsequently conducted, the point cloud is deduplicated and fused based on the preset spatial calibration relationship of all cameras. This embodiment provides a calibration method for a preset spatial calibration relationship, and the calibration method specifically includes: when it is detected that the calibration object passes through the inside of the ring surrounded by the camera, all cameras are controlled to move along the calibration object at the same time and scan to obtain reference point cloud data. In three-dimensional modeling, a calibration object (also called a calibration part) is a specific object or device used to calibrate, verify or calibrate a measurement system, camera system, scanning equipment or other three-dimensional data acquisition equipment. The design and use of calibration parts are intended to ensure that these devices can accurately and reliably acquire and reconstruct the shape, size and position information of objects in three-dimensional space. Figure 4 is a schematic diagram of the position relationship between a camera and a calibration object provided in an embodiment of the present application, referring to Figure 4, when it is detected that the calibration object 3 passes through the inside of the ring H1 surrounded by the cameras 10, all cameras 10 are controlled to move and scan along the calibration object 3 at the same time to obtain the reference point cloud data. The reference point cloud data is compared with the standard point cloud data to determine the first position relationship of each camera 10 relative to the calibration object 3. The standard point cloud data is the standard point cloud data after the calibration object 3 is calibrated and corrected. The spatial position relationship of the cameras 10 is grouped and calibrated according to the first position relationship, and the second position relationship between the cameras 10 in the corresponding group is determined. The spatial position relationship of the cameras 10 is inter-grouped and registered according to the first position relationship and the second position relationship to obtain the third relative position relationship between each two groups. For example, as shown in 2, the relationship between camera group 1 and camera group 2 is registered, and the third position relationship between camera group 1 and camera group 2 is calibrated. The spatial position relationship of the cameras 10 is calibrated pairwise according to the first position relationship, the second position relationship and the third relative position relationship to obtain the preset spatial calibration relationship of all cameras 10. The final calibration result (i.e., the preset spatial calibration relationship) is stored in the system file. When the volume of the limb to be tested is subsequently detected, the pre-stored preset spatial calibration relationship can be directly called to deduplicate and fuse the point cloud data, thereby improving the overall work efficiency of limb volume detection.

[0079] The overlapping area in the area corresponding to the point cloud data is determined according to the preset spatial calibration relationship, and the point cloud data in the overlapping area is deduplicated to obtain fused point cloud data, which includes the point cloud data of the non-overlapping area and the point cloud data after deduplication in the overlapping area. The fused point cloud data can be understood as the fused point cloud data of the limb to be tested, and the 3D limb model of the limb to be tested can be obtained by subsequent 3D modeling based on the fused point cloud data.

[0080] As described above, after the limb to be measured is scanned to obtain point cloud data, the point cloud data is deduplicated and fused according to the preset spatial calibration relationship to obtain the fused point cloud data of all cameras, that is, the fused point cloud data of the limb to be measured, thereby achieving deduplication and fusion, and improving the accuracy of the point cloud data, thereby improving the accuracy of the three-dimensional limb model obtained by subsequent three-dimensional modeling, and further improving the accuracy of the limb volume measurement.

[0081] S103, performing three-dimensional modeling processing according to the fused point cloud data to obtain a target three-dimensional limb model.

[0082] The fused point cloud data obtained above is subjected to three-dimensional modeling processing to obtain a target three-dimensional limb model. Exemplarily, an initial voxel grid can be created according to the resolution of the fused point cloud data. The fused point cloud data is voxelized according to the initial voxel grid to obtain first voxel data, which includes vertices, voxel indices and voxel values ​​corresponding to each voxel. The first voxel data is mapped to the point cloud grid, and color rendering is performed according to the mapping result and the projection superposition result to generate a target three-dimensional limb model.

[0083] In the above, when it is detected that a limb to be measured is placed inside the circle formed by the cameras, the point cloud data obtained by scanning all the cameras are controlled, the point cloud data are deduplicated and fused to obtain the fused point cloud data, and three-dimensional modeling is performed according to the fused point cloud data to obtain the target three-dimensional limb model, that is, the three-dimensional limb model corresponding to the limb to be measured is obtained. Subsequently, the target three-dimensional limb model can be directly calculated and processed to obtain the corresponding volume information. Compared with the existing method of measuring with a soft ruler, the method of the three-dimensional limb model obtained by scanning with the camera in this embodiment has a higher accuracy rate, and automatically realizes scanning and volume measurement, thereby improving the degree of automation of limb volume measurement, and further improving the work efficiency of limb volume measurement.

[0084] Before measuring (calculating) the volume of the target 3D limb model, it is necessary to confirm whether the limb has severe shaking. If there is severe shaking, the 3D limb model obtained based on the point cloud data will have large errors and affect the accuracy of the subsequent calculation of the limb volume. Therefore, it is necessary to determine that the shaking range of the limb to be measured is within the threshold range (that is, within the range where the error caused by shaking can be ignored) before measuring the limb volume based on the target 3D limb. If it is determined that the shaking range of the limb to be measured exceeds the threshold range, it means that the limb to be measured is severely shaking during the scanning process and needs to be remeasured.

[0085] The present embodiment provides a limb shaking detection method, which specifically includes: scanning the limb to be tested twice and performing three-dimensional modeling processing to obtain the corresponding first three-dimensional limb model and second three-dimensional limb model. The specific scanning and modeling process of the first three-dimensional limb model and the second three-dimensional limb model is the same as the aforementioned S101-S103, and will not be repeated here. By scanning the same limb to be tested twice to obtain the first three-dimensional limb model and the second three-dimensional limb model, the first three-dimensional limb model and the second three-dimensional limb model are compared to confirm the shaking of the limb to be tested during measurement. By slicing the first three-dimensional limb model and the second three-dimensional limb model along a preset direction according to a preset interval (for example, 1 cm), the first slice data corresponding to the first three-dimensional limb model and the second slice data corresponding to the second three-dimensional limb model are obtained. The preset direction can be the direction corresponding to the center line of the ring surrounded by the camera (also referred to as the axial direction of the medical detection equipment), the placement direction of the limb to be tested, or the placement direction of the preset calibration object, etc. The first slice data is subjected to gravity calculation processing to obtain the first gravity point data, and the first gravity point data includes the gravity points of all the first slice data; the second slice data is subjected to gravity calculation processing to obtain the second gravity point data, and the second gravity point data includes the gravity points of all the second slice data. The first gravity point data and the second gravity point data are mapped to the same plane for curve drawing, and the first curve and the second curve are obtained accordingly, wherein the first curve corresponds to the change of the slice gravity point of the first three-dimensional limb model, and the second curve corresponds to the change of the slice gravity point of the second three-dimensional limb model. Distance calculation processing is performed based on the first curve and the second curve to obtain the Euclidean distance between the first curve and the second curve. The degree of overlap of the two curves is determined based on the Euclidean distance. When the Euclidean distance is below the preset value, it means that the shaking of the limb to be tested is within the normal range, and the first three-dimensional limb model is determined as the target three-dimensional limb model. When the Euclidean distance exceeds the preset value, it means that the limb to be tested has severe shaking during the scanning process, and the corresponding three-dimensional limb model obtained by scanning and modeling is not accurate. Therefore, the limb to be tested needs to be re-scanned and three-dimensionally modeled until the Euclidean distance is below the preset value, and the corresponding first three-dimensional limb model is determined as the target three-dimensional limb model.

[0086] As mentioned above, the limb shaking detection method is used to determine that the three-dimensional limb model corresponding to the shaking condition within the normal range is the target three-dimensional limb model, so as to avoid directly analyzing and calculating the three-dimensional limb model obtained based on the severe shaking of the limb to be tested, resulting in inaccurate volume information finally obtained. This embodiment determines that the three-dimensional limb model corresponding to the shaking condition within the normal range is the target three-dimensional limb model, which can improve the accuracy of the target three-dimensional limb model, and thereby improve the accuracy of the volume information obtained by subsequent volume detection based on the target three-dimensional limb model.

[0087] S104, performing slicing processing and center of gravity analysis processing according to the target three-dimensional limb model to determine the center of gravity of the joint.

[0088] After determining the target three-dimensional limb model corresponding to the limb to be tested, the target three-dimensional limb model can be decomposed into slices by slicing, and volume detection is performed for each slice. The volume parameters corresponding to all slices are integrated to obtain the complete volume information corresponding to the limb to be tested. However, during the scanning process, the limb to be tested may not be placed completely straight, and there may be a certain angle. Take the limb to be tested as an arm as an example for explanation. Based on the joint position of the arm, it is not completely straight during scanning, and there is a certain bending based on the joint position. Therefore, the existing method of directly slicing the three-dimensional limb model according to the direction of arm placement to obtain the limb volume will cause corresponding errors in the slice data due to the bending of the arm. Therefore, this embodiment provides a two-slicing method, the first slicing is to slice along a preset direction to obtain slice data, analyze and process the slice data to determine the joint position, re-slice based on the joint position, and slice in the corresponding direction with the joint position as the boundary, so as to reduce the influence of the bending based on the joint position on the accuracy of the slice data, thereby improving the accuracy of the slice data finally obtained.

[0089] This embodiment provides an implementation method for determining the joint positions in a target three-dimensional limb model, specifically comprising: slicing the target three-dimensional limb model along a preset direction and at a preset distance to obtain reference slice data. Calculating and determining the center of gravity of each reference slice data, connecting the center of gravity of the first reference slice data and the center of gravity of the last reference slice data to obtain a first reference line. Determine the center of gravity of all the reference slice data that is farthest from the first reference line as the joint center of gravity. Exemplarily, Figure 5 is a first schematic diagram of a limb model slice provided in an embodiment of the present application, referring to Figure 5 , slice the target three-dimensional limb model along a preset direction (assuming the X1 (not shown in the figure) axis direction corresponding to the first reference line x1 in the figure) at a preset distance (assuming 1 cm) to obtain reference slice data. Assume that the first reference slice is slice A and the last reference slice is slice B. Calculate and determine the center of gravity of each reference slice data. Assume that the center of gravity point a of slice A and the center of gravity point b of slice B are calculated. Connect the center of gravity point a and the center of gravity point b to obtain the first reference line x1. Determine the center of gravity point farthest from the first reference line x1 among the center of gravity points of all reference slice data as the joint center of gravity point. Assume that the center of gravity point farthest from the first reference line x1 is point c, then determine the joint center of gravity point as point c.

[0090] As described above, slicing processing and center of gravity analysis processing are performed based on the target three-dimensional limb model to determine the joint center of gravity points, and then secondary slicing is performed based on the joint center of gravity points to reduce the impact of bending based on joint position on the accuracy of the slice data, thereby improving the accuracy of the slice data, and further improving the accuracy of the volume information calculated based on the slice data.

[0091] S105, re-slicing the target three-dimensional limb model according to the joint center of gravity points to obtain target slice data.

[0092] After the joint gravity point is determined as mentioned above, the target three-dimensional limb model is divided into two parts with the joint gravity point as the boundary, and the part between the first point cloud of the first reference slice data and the last point cloud in the reference slice data corresponding to the joint gravity point is defined as the first part, such as the forearm of the arm; the part between the first point cloud in the reference slice data corresponding to the joint gravity point and the last point cloud in the last reference slice data is defined as the second part, such as the upper arm of the arm. The first part and the second part can be re-sliced ​​respectively, so that the first part and the second part are sliced ​​in the direction corresponding to their own axis, so as to avoid bending the complete target three-position limb according to one direction to affect the accuracy of the slice data based on the joint position. Specifically, the gravity point of the first reference slice data and the joint gravity point are connected to obtain the second reference line. The first part of the target three-dimensional limb model is re-sliced ​​along the second reference line at a preset interval to obtain the first target slice data. The joint gravity point and the center point of the last reference slice data are connected to obtain the third reference line. The second part of the target three-dimensional limb model is re-sliced ​​along the third reference line at a preset interval to obtain the second target slice data. The first target slice data and the second target slice data are fused and spliced ​​with the joint center point as the splicing point to obtain the target slice data of the entire target three-dimensional limb model.

[0093] For example, Figure 6 The second schematic diagram of a limb model slice provided in the embodiment of the present application is referred to as Figure 6, determine the joint center of gravity as point c, and the reference slice of the joint center of gravity as slice C. Define slice A to slice C as the first part, and define slice C to slice B as the second part. Connect the center of gravity point a of slice A and the joint center of gravity point c to obtain a second reference line X2. Re-slice the first part of the target three-dimensional limb model (the part from slice A to slice C) along the second reference line X2 at a preset interval (for example, 1 cm) to obtain the first target slice data. Connect the joint center of gravity point c and the center of gravity point b of the reference slice B to obtain a third reference line X3. Re-slice the second part of the target three-dimensional limb model (the part from slice C to slice B) along the third reference line X3 at a preset interval (for example, 1 cm) to obtain the second target slice data. The first target slice data and the second target slice data are fused and spliced ​​with the joint center of gravity as the splicing point to obtain the target slice data of the entire target three-dimensional limb model.

[0094] S106 , performing parameter calculation processing according to the target slice data to obtain volume information of each target slice data, where the volume information includes the thickness and cross-sectional area of ​​the slice.

[0095] After the target slice data is obtained by re-slicing based on the joint center of gravity, parameter calculation is performed according to the target slice data to obtain the slice thickness and cross-sectional area corresponding to each target slice data, that is, to obtain the volume information of each target slice data. Since the limb to be tested is not a regular cylindrical object, this embodiment slices the target three-dimensional limb model corresponding to the limb to be tested into slices that are approximately cylindrical, and calculates the volume information of each slice, making the calculation more convenient and accurate, thereby improving the overall work efficiency and accuracy of volume detection.

[0096] S107, integrating and processing the volume parameters of each target slice data to obtain the overall volume information of the limb to be measured.

[0097] Based on slicing the target three-dimensional limb model to obtain the corresponding target slice data, after detecting and calculating the volume parameters corresponding to each target slice data, the volume parameters of each target slice data are integrated to obtain the overall volume parameters of the target three-dimensional limb model, that is, the overall volume information of the limb to be tested. Subsequently, the overall volume information of the limb to be tested can be obtained based on different time detections for comparative analysis to determine whether the corresponding limb to be tested has limb volume changes such as edema.

[0098] The above is used for a medical detection device equipped with at least four cameras, the cameras are arranged in a ring at preset intervals, the camera lenses face the central area of ​​the ring, the ring is used for the limb to be tested to pass through, when it is detected that the limb to be tested passes through the ring formed by the cameras, all cameras are controlled to move and scan along the limb to be tested at the same time, de-duplication fusion and three-dimensional modeling are performed based on the scanned point cloud data to obtain a target three-dimensional limb model, slice processing and center of gravity analysis processing are performed based on the target three-dimensional limb model to determine the joint center of gravity point, the target three-dimensional limb model is re-sliced ​​based on the joint center of gravity point to obtain target slice data, and parameter calculation and data integration are performed based on the target slice data to obtain the overall volume information of the limb to be tested. By adopting the above technical means, the overall volume information of the limb to be tested can be obtained by scanning with the medical detection equipment, thereby avoiding the technical problems of low work efficiency and large errors caused by manual measurement, and the overall volume information of the limb to be tested can be obtained based on camera scanning, thereby improving the work efficiency of limb volume detection; and firstly preliminarily slice to determine the joint center of gravity point, re-slice based on the joint center of gravity point, and slice separately with the joint center of gravity point as the boundary, thereby improving the accuracy of slicing, and then improving the accuracy of volume calculation.

[0099] Based on the above embodiments, Figure 7 This is a schematic diagram of the structure of a limb volume detection device provided in an embodiment of the present application. Figure 7 The limb volume detection device provided in this embodiment specifically includes: a scanning module 21, a point cloud fusion module 22, a modeling module 23, a joint determination module 24, a re-slicing module 25, a first calculation module 26 and a second calculation module 27.

[0100] The scanning module 21 is used to control all cameras to move along the limb to be tested and scan simultaneously when it is detected that the limb to be tested passes through the ring formed by the cameras, so as to obtain point cloud data scanned by multiple cameras;

[0101] The point cloud fusion module 22 is used to perform deduplication fusion processing on the point cloud data according to a preset spatial calibration relationship to obtain fused point cloud data of all cameras, where the preset spatial calibration relationship is the relative position relationship between every two cameras;

[0102] A modeling module 23 is used to perform three-dimensional modeling processing according to the fused point cloud data to obtain a target three-dimensional limb model;

[0103] The joint determination module 24 is used to perform slicing processing and gravity center analysis processing according to the target three-dimensional limb model to determine the joint gravity center point;

[0104] A re-slicing module 25 is used to re-slice the target three-dimensional limb model according to the joint center of gravity to obtain target slice data;

[0105] A first calculation module 26 is used to perform parameter calculation processing according to the target slice data to obtain volume information of each target slice data, where the volume information includes the thickness and cross-sectional area of ​​the slice;

[0106] The second calculation module 27 is used to integrate the volume parameters of each target slice data to obtain the overall volume information of the limb to be measured.

[0107] In one embodiment, the modeling module 23 includes a voxel grid construction submodule, a voxelization processing submodule, and a voxel mapping submodule;

[0108] A voxel grid construction submodule is used to create an initial voxel grid according to the resolution of the fused point cloud data;

[0109] A voxel processing submodule, used for voxelizing the fused point cloud data according to the initial voxel grid to obtain first voxel data, wherein the first voxel data includes a vertex, a voxel index and a voxel value corresponding to each voxel;

[0110] The voxel mapping submodule is used to map the first voxel data into the point cloud grid, perform color rendering processing according to the mapping result and the projection superposition result, and generate a target three-dimensional limb model.

[0111] In one embodiment, the joint determination module 24 includes a first slicing submodule, a first center of gravity calculation submodule, a first reference line confirmation submodule, and a joint determination submodule;

[0112] The first slicing submodule is used to slice the target three-dimensional limb model along a preset direction and at a preset distance to obtain reference slice data;

[0113] A first centroid calculation submodule is used to calculate and determine the centroid point of each reference slice data;

[0114] A first reference line confirmation submodule, used to connect the centroid point of the first reference slice data and the centroid point of the last reference slice data to obtain a first reference line;

[0115] The joint determination submodule is used to determine the center of gravity point farthest from the first reference line among the center of gravity points of all reference slice data as the joint center of gravity point.

[0116] In one embodiment, the reslicing module 25 includes a second reference line determination submodule, a first reslicing submodule, a third reference line determination submodule, a second reslicing submodule, and a stitching submodule;

[0117] A second reference line determination submodule is used to connect the center of gravity of the first reference slice data and the center of gravity of the joint to obtain a second reference line;

[0118] A first re-slicing submodule is used to re-slice a first part of the target three-dimensional limb model along a second reference line at a preset interval to obtain first target slice data, wherein the first part is a part between a first point cloud of the first reference slice data and a last point cloud of the reference slice data corresponding to a joint centroid point;

[0119] The third reference line determination submodule is used to connect the joint center of gravity point with the center of gravity point of the last reference slice data to obtain the third reference line;

[0120] A second re-slicing submodule is used to re-slice the second part of the target three-dimensional limb model along the third reference line at a preset interval to obtain second target slice data, where the second part is a part between the first point cloud in the reference slice data corresponding to the joint center of gravity point and the last point cloud in the last reference slice data;

[0121] The splicing submodule is used to fuse and splice the first target slice data and the second target slice data with the joint center of gravity as the splicing point to obtain the target slice data of the entire target three-dimensional limb model.

[0122] In one embodiment, the limb volume detection device further includes a slicing module, a first center of gravity calculation module, a second center of gravity calculation module, a center of gravity point mapping module, a distance calculation module, and a target model determination module;

[0123] The modeling module 23 is further used to perform two scans and three-dimensional modeling processes on the limb to be tested to obtain a corresponding first three-dimensional limb model and a second three-dimensional limb model;

[0124] A slicing module, used to slice the first three-dimensional limb model and the second three-dimensional limb model along a preset direction and at a preset interval to obtain first slice data corresponding to the first three-dimensional limb model and second slice data corresponding to the second three-dimensional limb model;

[0125] A first centroid calculation module, used to perform centroid calculation processing on the first slice data to obtain first centroid data, where the first centroid data includes centroids of all first slice data;

[0126] A second centroid calculation module, used to perform centroid calculation processing on the second slice data to obtain second centroid data, where the second centroid data includes centroids of all second slice data;

[0127] A centroid mapping module, used for mapping the first centroid data and the second centroid data to the same plane for curve drawing, and obtaining a first curve and a second curve correspondingly;

[0128] A distance calculation module, used for performing distance calculation processing according to the first curve and the second curve to obtain the Euclidean distance between the first curve and the second curve;

[0129] A target model determination module, used for determining the first three-dimensional limb model as a target three-dimensional limb model when the Euclidean distance is below a preset value;

[0130] The target model determination module is also used to re-scan and perform three-dimensional modeling on the limb to be tested when the Euclidean distance exceeds a preset value, until the Euclidean distance is below the preset value, and then determine that the corresponding first three-dimensional limb model is the target three-dimensional limb model.

[0131] In one embodiment, the limb volume detection device further includes a calibration scanning module 21, a first position calibration module, a second position calibration module, a third position calibration module and a space position calibration module;

[0132] The calibration scanning module 21 is used to control all cameras to move along the calibration object and scan it simultaneously when the calibration object is detected to pass through the ring formed by the cameras, so as to obtain reference point cloud data;

[0133] A first position relationship calibration module is used to compare the reference point cloud data with the standard point cloud data to determine a first position relationship of each camera relative to the calibration object, wherein the standard point cloud data is the standard point cloud data after point cloud calibration and correction of the calibration object;

[0134] A second position calibration module, used to calibrate the spatial position relationship of the cameras in groups according to the first position relationship, and determine a second position relationship between the cameras in the corresponding group;

[0135] A third position calibration module is used to perform inter-group registration processing on the spatial position relationship of the cameras according to the first position relationship and the second position relationship to obtain a third relative position relationship between every two groups;

[0136] The spatial position calibration module is used to perform pairwise calibration processing on the spatial position relationship of the cameras according to the first position relationship, the second position relationship and the third relative position relationship to obtain a preset spatial calibration relationship of all cameras.

[0137] The limb volume detection device provided in the embodiment of the present application can be used to execute the limb volume detection method provided in the above embodiment, and has corresponding functions and beneficial effects.

[0138] The present application embodiment provides a limb volume detection device, referring to Figure 8The limb volume detection device includes: a processor 31, a memory 32, a communication module 33, an input device 34 and an output device 35. The number of processors in the limb volume detection device can be one or more, and the number of memories in the limb volume detection device can be one or more. The processor, memory, communication module, input device and output device of the limb volume detection device can be connected via a bus or other means.

[0139] The memory 32, as a computer-readable storage medium, can be used to store software programs, computer executable programs and modules, such as program instructions / modules corresponding to the limb volume detection method described in any embodiment of the present application (for example, the scanning module, point cloud fusion module, modeling module, joint determination module, re-slicing module, first calculation module and second calculation module in the limb volume detection device). The memory may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system and an application required for at least one function; the data storage area may store data created according to the use of the device, etc. In addition, the memory may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device or other non-volatile solid-state storage device. In some instances, the memory may further include a memory remotely arranged relative to the processor, and these remote memories may be connected to the device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network and a combination thereof.

[0140] The communication module 33 is used for data transmission.

[0141] The processor 31 executes various functional applications and data processing of the device by running the software programs, instructions and modules stored in the memory, that is, realizes the above-mentioned limb volume detection method.

[0142] The input device 34 may be used to receive input digital or character information and generate key signal input related to user settings and function control of the device. The output device 35 may include a display device such as a display screen.

[0143] The limb volume detection device provided above can be used to execute the limb volume detection method provided in the above embodiment, and has corresponding functions and beneficial effects.

[0144] The embodiment of the present application also provides a storage medium storing computer executable instructions, which are used to execute a limb volume detection method when executed by a computer processor. The limb volume detection method includes: when it is detected that the limb to be measured passes through the inside of the ring surrounded by cameras, controlling all cameras to move and scan along the limb to be measured at the same time to obtain point cloud data scanned by multiple cameras; performing de-duplication fusion processing on the point cloud data according to a preset spatial calibration relationship to obtain fused point cloud data of all cameras, and the preset spatial calibration relationship is the relative position relationship between every two cameras; performing three-dimensional modeling processing on the fused point cloud data to obtain a target three-dimensional limb model; performing slicing processing and center of gravity analysis processing on the target three-dimensional limb model to determine the center of gravity points of the joints; re-slicing the target three-dimensional limb model according to the center of gravity points of the joints to obtain target slice data; performing parameter calculation processing on the target slice data to obtain volume information of each target slice data, and the volume information includes the thickness and cross-sectional area of ​​the slice; integrating the volume parameters of each target slice data to obtain the overall volume information of the limb to be measured.

[0145] Storage medium - any of various types of memory devices or storage devices. The term "storage medium" is intended to include: installation media, such as CD-ROM, floppy disk or tape device; computer system memory or random access memory, such as DRAM, DDR RAM, SRAM, EDO RAM, Rambus RAM, etc.; non-volatile memory, such as flash memory, magnetic media (such as hard disk or optical storage); registers or other similar types of memory elements, etc. Storage media may also include other types of memory or combinations thereof. In addition, the storage medium may be located in the first computer system in which the program is executed, or may be located in a different second computer system, which is connected to the first computer system via a network (such as the Internet). The second computer system can provide program instructions to the first computer for execution. The term "storage medium" may include two or more storage media residing in different locations (for example, in different computer systems connected by a network). The storage medium may store program instructions (for example, embodied as a computer program) that can be executed by one or more processors.

[0146] Of course, the computer executable instructions of a storage medium storing computer executable instructions provided in an embodiment of the present application are not limited to the limb volume detection method described above, and can also execute related operations in the limb volume detection method provided in any embodiment of the present application.

[0147] The limb volume detection device, storage medium and limb volume detection equipment provided in the above embodiments can execute the limb volume detection method provided in any embodiment of the present application. For technical details not described in detail in the above embodiments, please refer to the limb volume detection method provided in any embodiment of the present application.

[0148] The above are only preferred embodiments of the present application and the technical principles used. The present application is not limited to the specific embodiments described herein, and various obvious changes, readjustments and substitutions that can be made by those skilled in the art will not deviate from the scope of protection of the present application. Therefore, although the present application is described in more detail through the above embodiments, the present application is not limited to the above embodiments, and may include more other equivalent embodiments without departing from the concept of the present application, and the scope of the present application is determined by the scope of the claims.

Claims

1. A method for detecting limb volume, characterized in that: Used in medical detection equipment, the medical detection equipment is equipped with at least four cameras, the cameras are arranged in a ring at preset intervals, the lenses of the cameras are oriented towards the central area of ​​the ring, the ring is used for a limb to be detected to pass through, the method comprising: When it is detected that the limb to be tested passes through the ring formed by the cameras, all cameras are controlled to move and scan along the limb to be tested at the same time, so as to obtain point cloud data scanned by multiple cameras; De-duplication and fusion processing is performed on the point cloud data according to a preset spatial calibration relationship to obtain fused point cloud data of all cameras, wherein the preset spatial calibration relationship is a relative position relationship between every two cameras; Performing three-dimensional modeling processing according to the fused point cloud data to obtain a target three-dimensional limb model; Performing slicing processing and center of gravity analysis processing according to the target three-dimensional limb model to determine the center of gravity of the joint; Re-slicing the target three-dimensional limb model according to the joint center of gravity to obtain target slice data; Performing parameter calculation processing according to the target slice data to obtain volume information of each target slice data, wherein the volume information includes the thickness and cross-sectional area of ​​the slice; The volume parameters of each target slice data are integrated to obtain the overall volume information of the limb to be measured.

2. The method according to claim 1, characterized in that: The deduplication fusion processing is performed on the point cloud data according to the preset spatial calibration relationship to obtain the fused point cloud data of all cameras, wherein the preset spatial calibration relationship is the relative position relationship between every two calibrated cameras, including: Determine the overlapping area in the area corresponding to the point cloud data according to a preset spatial calibration relationship; The point cloud data in the overlapping area is deduplicated to obtain fused point cloud data, wherein the fused point cloud data includes the point cloud data in the non-overlapping area and the deduplicated point cloud data in the overlapping area.

3. The method according to claim 1, characterized in that The three-dimensional modeling process is performed according to the fused point cloud data to obtain a target three-dimensional limb model, including: Creating an initial voxel grid according to the resolution of the fused point cloud data; voxelize the fused point cloud data according to the initial voxel grid to obtain first voxel data, wherein the first voxel data includes a vertex, a voxel index and a voxel value corresponding to each voxel; The first voxel data is mapped into a point cloud grid, and color rendering is performed according to the mapping result and the projection superposition result to generate the target three-dimensional limb model.

4. The method according to claim 1, characterized in that: The step of performing slicing and gravity center analysis processing according to the target three-dimensional limb model to determine the joint gravity center point includes: Slicing the target three-dimensional limb model along a preset direction and at a preset distance to obtain reference slice data; Calculate and determine the centroid of each reference slice data; Connect the centroid of the first reference slice data and the centroid of the last reference slice data to obtain a first reference line; The center of gravity of all the reference slice data that is farthest from the first reference line is determined as the joint center of gravity.

5. The method according to claim 4, characterized in that The step of reslicing the target three-dimensional limb model according to the joint center of gravity to obtain target slice data includes: Connecting the center of gravity of the first reference slice data and the center of gravity of the joint to obtain a second reference line; Reslicing the first part of the target three-dimensional limb model along the second reference line at a preset interval to obtain first target slice data, wherein the first part is a part between the first point cloud of the first reference slice data and the last point cloud of the reference slice data corresponding to the joint centroid point; Connect the joint center of gravity point and the center of gravity point of the last reference slice data to obtain a third reference line; Re-slicing the second part of the target three-dimensional limb model along the third reference line at a preset interval to obtain second target slice data, wherein the second part is a part between the first point cloud in the reference slice data corresponding to the joint center of gravity point and the last point cloud in the last reference slice data; The first target slice data and the second target slice data are fused and spliced ​​with the joint center point as the splicing point to obtain the target slice data of the entire target three-dimensional limb model.

6. The method according to claim 1, characterized in that The method of performing slicing processing and center of gravity analysis processing according to the target three-dimensional limb model to determine the center of gravity of the joint includes: Scanning and three-dimensional modeling are performed twice on the limb to be tested to obtain a corresponding first three-dimensional limb model and a second three-dimensional limb model; Slice the first three-dimensional limb model and the second three-dimensional limb model along a preset direction and at a preset interval to obtain first slice data corresponding to the first three-dimensional limb model and second slice data corresponding to the second three-dimensional limb model; Performing a gravity point calculation process on the first slice data to obtain first gravity point data, wherein the first gravity point data includes gravity points of all the first slice data; Performing a gravity point calculation process on the second slice data to obtain second gravity point data, wherein the second gravity point data includes gravity points of all the second slice data; Mapping the first center of gravity point data and the second center of gravity point data to the same plane for curve drawing, and obtaining a first curve and a second curve correspondingly; Performing distance calculation processing according to the first curve and the second curve to obtain a Euclidean distance between the first curve and the second curve; When the Euclidean distance is below a preset value, determining the first three-dimensional limb model as the target three-dimensional limb model; When the Euclidean distance exceeds a preset value, the limb to be tested is re-scanned and three-dimensionally modeled until the Euclidean distance is below the preset value, and the corresponding first three-dimensional limb model is determined as the target three-dimensional limb model.

7. The method according to claim 1, characterized in that The method further comprises: When it is detected that the calibration object passes through the ring formed by the cameras, all cameras are controlled to move along the calibration object and scan at the same time to obtain reference point cloud data; Comparing the reference point cloud data with standard point cloud data to determine a first position relationship of each camera relative to the calibration object, wherein the standard point cloud data is standard point cloud data after point cloud calibration and correction is performed on the calibration object; Grouping and calibrating the spatial positional relationship of the cameras according to the first positional relationship to determine a second positional relationship between the cameras in the corresponding group; Performing inter-group registration processing on the spatial positional relationship of the cameras according to the first positional relationship and the second positional relationship to obtain a third relative positional relationship between every two groups; The spatial positional relationships of the cameras are calibrated in pairs according to the first positional relationship, the second positional relationship and the third relative positional relationship to obtain preset spatial calibration relationships of all cameras.

8. A limb volume detection device, characterized in that: Used in medical detection equipment, the medical detection equipment is equipped with at least four cameras, the cameras are arranged in a ring at preset intervals, the lenses of the cameras are oriented towards the central area of ​​the ring, the ring is used for the limb to be detected to pass through, the device comprises: A scanning module, for controlling all cameras to move along the limb to be tested and scan simultaneously when detecting that the limb to be tested passes through the ring formed by the cameras, so as to obtain point cloud data scanned by multiple cameras; A point cloud fusion module is used to perform deduplication fusion processing on the point cloud data according to a preset spatial calibration relationship to obtain fused point cloud data of all cameras, wherein the preset spatial calibration relationship is the relative position relationship between every two cameras; A modeling module, used for performing three-dimensional modeling processing according to the fused point cloud data to obtain a target three-dimensional limb model; A joint determination module, used for performing slicing processing and center of gravity analysis processing according to the target three-dimensional limb model to determine the center of gravity of the joint; A re-slicing module is used to re-slice the target three-dimensional limb model according to the joint center of gravity point to obtain target slice data; A first calculation module, used for performing parameter calculation processing according to the target slice data to obtain volume information of each target slice data, wherein the volume information includes the thickness and cross-sectional area of ​​the slice; The second calculation module is used to integrate the volume parameters of each target slice data to obtain the overall volume information of the limb to be measured.

9. A limb volume detection device, characterized in that: include: memory and one or more processors; The memory is used to store one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1 to 7.

10. A storage medium storing computer executable instructions, characterized in that: The computer executable instructions are used to perform the method according to any one of claims 1 to 7 when executed by a processor.

Images