Method and device for measuring number of spherical branchlets of caulerpa lentillifera

Abstract

The invention discloses a method and a device for measuring the number of spherical branchlets of caulerpa lentillifera, and relates to the technical field of caulerpa lentillifera detection, and the method comprises the following steps: collecting three-dimensional point cloud data of each caulerpa lentillifera sample by adopting a small animal living microCT imager; performing spherical branchlet labeling on the three-dimensional point cloud data to construct a data set; training the point cloud segmentation network by adopting the data set to obtain a target segmentation model; and detecting the number of the spherical branchlets of the caulerpa lentillifera to be detected by adopting the target segmentation model. According to the invention, lossless high-throughput measurement of the number of spherical branchlets can be realized.

Description

Technical Field

[0001] The present application relates to the technical field of detection of Caulerpa lentillifera, and particularly relates to a method and device for measuring the number of spherical branches of Caulerpa lentillifera. Background Art

[0002] The measurement of Caulerpa lentillifera is generally carried out manually, that is, using tweezers to remove spherical branches one by one and count. Measuring the traits of upright branches manually is not only time-consuming and laborious, but also causes the upright branches to wilt and die. Moreover, due to factors such as human subjective errors and complex data management, the data accuracy is relatively low. In addition, due to the special grape-like morphological structure of the thallus itself, extracting the phenotypic characteristics of upright branches from two-dimensional images has problems of inaccurate data and slow processing speed due to the mutual occlusion between spherical branches. Summary of the Invention

[0003] The purpose of the present application is to provide a method and device for measuring the number of spherical branches of Caulerpa lentillifera, which can improve the accuracy of statistics of spherical upright branches.

[0004] To achieve the above purpose, the present application provides the following solutions: In a first aspect, the present application provides a method for measuring the number of spherical branches of Caulerpa lentillifera, and the method for measuring the number of spherical branches of Caulerpa lentillifera includes: Collecting three-dimensional point cloud data of each Caulerpa lentillifera sample by using a small animal in-vivo microCT imager; Constructing a data set by performing spherical branch annotation on each three-dimensional point cloud data; Training a point cloud segmentation network by using the data set to obtain a target segmentation model; Detecting the number of spherical branches of the Caulerpa lentillifera to be detected by using the target segmentation model.

[0005] Optionally, collecting three-dimensional point cloud data of each Caulerpa lentillifera sample by using a small animal in-vivo microCT imager specifically includes: Starting from one end of the sample bed groove, using a polyethylene film to wind along the radial contour of the sample bed groove, and stretching the polyethylene film to form a support plane, where the support plane makes the Caulerpa lentillifera sample not in contact with the bottom of the sample bed groove, so as to obtain a processed Caulerpa lentillifera sample; After putting the processed Caulerpa lentillifera sample into the chamber of the small animal in-vivo microCT imager, setting the working parameters of the small animal in-vivo microCT imager according to preset working parameter conditions, where the preset working parameter conditions include that the voltage and current meet the set conditions, the field of view is adjusted to the minimum, the camera rotation speed is the set rotation speed, and the camera rotation angle is the set angle; Obtain the three-dimensional point cloud data obtained after the small animal in vivo microCT imager performs CT scanning on the Caulerpa lentillifera sample.

[0006] Optionally, the condition that the voltage and current satisfy is that the voltage Current ≤ 8W, the set rotation speed is 21° / s, and the set angle is 384°.

[0007] Optionally, the preset working parameter conditions further include that the diameter of the X-ray scanning area is set to 86 mm and the voxel size is set to 172 μm.

[0008] Optionally, the point cloud segmentation network adopts the PointNet++ network.

[0009] Optionally, the target segmentation model is used to detect the number of spherical branches of the Caulerpa lentillifera to be detected, specifically including: Use a small animal in vivo microCT imager to collect the three-dimensional point cloud data of the Caulerpa lentillifera to be detected; Input the three-dimensional point cloud data of the Caulerpa lentillifera to be detected into the target segmentation model to obtain the total number of spherical branches of the Caulerpa lentillifera to be detected.

[0010] Optionally, the three-dimensional point cloud data is three-dimensional point cloud data in ply format.

[0011] Optionally, a data set is constructed by performing spherical branch annotation on each three-dimensional point cloud data, specifically including: Perform boundary annotation on the spherical branches of each three-dimensional point cloud data, and use the bounding box of each spherical branch and the total number of spherical branches as label data.

[0012] Optionally, performing boundary annotation on the spherical branches of each three-dimensional point cloud data specifically includes: Use CloudCompare software to perform boundary annotation on the spherical branches of each three-dimensional point cloud data.

[0013] In a second aspect, the present application provides a device for measuring the number of spherical branches of Caulerpa lentillifera, characterized in that the device for measuring the number of spherical branches of Caulerpa lentillifera applies the method for measuring the number of spherical branches of Caulerpa lentillifera described in any one of the above, and the device for measuring the number of spherical branches of Caulerpa lentillifera includes: A three-dimensional point cloud data acquisition module, configured to use a small animal in vivo microCT imager to collect the three-dimensional point cloud data of each Caulerpa lentillifera sample; A data set construction module, configured to construct a data set by performing spherical branch annotation on each three-dimensional point cloud data; A model training module, configured to use the data set to train a point cloud segmentation network to obtain a target segmentation model; A detection module, configured to detect the number of spherical branches of the Caulerpa lentillifera to be detected by using the target segmentation model.

[0014] According to the specific embodiments provided in this application, the following technical effects are disclosed in this application: This application provides a method and device for measuring the number of spherical branches of Caulerpa lentillifera. The three-dimensional (3D) point cloud data of each Caulerpa lentillifera sample is collected by using a small animal in vivo microCT imager (Quantum GX2). The obtained point cloud data has high resolution and fast scanning speed. The target segmentation model is used to detect the number of upright branches (spherical branches) of the Caulerpa lentillifera to be detected. Compared with manual statistics, this application improves the accuracy and efficiency of spherical upright branch statistics based on CT-3D imaging and the target segmentation module, and realizes non-destructive high-throughput measurement of upright branches, providing an effective tool for measuring the morphological traits of Caulerpa lentillifera and a scientific method for variety breeding and improvement. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the following will briefly introduce the drawings required in the embodiments. Obviously, the drawings in the following description are only some embodiments of this application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0016] Figure 1 It is a schematic flowchart of a method for measuring the number of spherical branches of Caulerpa lentillifera provided in an embodiment of this application.

[0017] Figure 2 It is a schematic diagram of the positional relationship between a Caulerpa lentillifera sample and a sample bed provided in an embodiment of this application.

[0018] Figure 3 It is a schematic diagram of the interface for separating the spherical branch part during the point cloud annotation process provided in an embodiment of this application.

[0019] Figure 4 It is a schematic diagram of the interface for selecting the segmented point cloud during the point cloud annotation process provided in an embodiment of this application.

[0020] Figure 5 It is a schematic diagram of the interface for annotating the target in the point cloud during the point cloud annotation process provided in an embodiment of this application.

[0021] Figure 6 It is a schematic diagram of the interface after saving during the point cloud annotation process provided in an embodiment of this application.

[0022] Figure 7Schematic diagram of manual threshing and counting provided by an embodiment of the present application.

[0023] Figure 8 Schematic diagram of the functional modules of a measuring device for the number of spherical branches of Caulerpa lentillifera provided by an embodiment of the present application. Detailed implementation manners

[0024] Next, the technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present application without creative efforts shall fall within the protection scope of the present application.

[0025] To make the above objects, features, and advantages of the present application more obvious and understandable, the present application will be further described in detail below in conjunction with the accompanying drawings and specific implementation manners.

[0026] In an exemplary embodiment, the present application provides a method for measuring the number of spherical branches of Caulerpa lentillifera, as Figure 1 shown, the method for measuring the number of spherical branches of Caulerpa lentillifera includes steps 101 to 104.

[0027] Step 101: Use a small animal in vivo microCT imager to collect three-dimensional point cloud data of each Caulerpa lentillifera sample.

[0028] Step 102: Construct a data set by labeling spherical branches for each three-dimensional point cloud data.

[0029] Step 103: Use the data set to train a point cloud segmentation network to obtain a target segmentation model.

[0030] Step 104: Use the target segmentation model to detect the number of spherical branches of the Caulerpa lentillifera to be detected.

[0031] The present application uses a small animal in vivo microCT imager to collect three-dimensional point cloud data of each Caulerpa lentillifera sample. The obtained point cloud data has high resolution and fast scanning speed. The target segmentation model is used to detect the number of spherical branches (upright branches) of the Caulerpa lentillifera to be detected. Compared with manual statistics, the accuracy and efficiency of the statistics of spherical upright branches are improved.

[0032] In an exemplary embodiment, step 101 specifically includes steps 201 to 202.

[0033] Step 201: Starting from one end of the sample bed groove, use a polyethylene film to wind along the radial contour of the sample bed groove, stretch the polyethylene film to form a support plane, and the support plane keeps the Caulerpa lentillifera sample from contacting the bottom of the sample bed groove, obtaining a processed Caulerpa lentillifera sample.

[0034] Step 202: After placing the processed Caulerpa lentillifera sample into the chamber of the small animal in vivo microCT imager, set the working parameters of the small animal in vivo microCT imager according to the preset working parameter conditions, and the preset working parameter conditions include that the voltage and current meet the set conditions, the field of view is adjusted to the minimum, the camera rotation speed is the set rotation speed, and the camera rotation angle is the set angle.

[0035] Step 203: Obtain the three-dimensional point cloud data obtained after the small animal in vivo microCT imager performs CT scanning on the Caulerpa lentillifera sample.

[0036] The voltage and current meeting the set conditions means that the voltage current ≤ 8 W, the set rotation speed is 21° / s, and the set angle is 384°.

[0037] The preset working parameter conditions also include that the diameter of the X-ray scanning area is set to 86 mm and the voxel size is set to 172 μm.

[0038] In an exemplary embodiment, step 104 specifically includes: using a small animal in vivo microCT imager to collect the three-dimensional point cloud data of the Caulerpa lentillifera to be detected; inputting the three-dimensional point cloud data of the Caulerpa lentillifera to be detected into the target segmentation model to obtain the total number of spherical branches of the Caulerpa lentillifera to be detected.

[0039] The three-dimensional point cloud data is three-dimensional point cloud data in ply format.

[0040] Each Caulerpa lentillifera sample includes Caulerpa lentillifera samples of multiple germplasms. Before measuring with a small animal in vivo microCT imager, select relatively complete upright branches of each germplasm, dry the moisture and wrap them with plastic wrap to prevent the spherical branches from sticking to each other and interfering.

[0041] In an exemplary embodiment, the operation of the small animal in vivo microCT imager includes the following content.

[0042] (1) Turn on the machine and preheat. After the preheating is completed, select the appropriate sample chamber and sample bed, and wrap the middle of the sample bed with plastic wrap to form a support plane to separate the sample from the sample bed groove for convenient subsequent image processing and data processing. During the operation of the instrument, close and lock the X-ray chamber door, and it is strictly prohibited to open the X-ray chamber door during operation to avoid X-ray radiation leakage.

[0043] Specifically, the density of the Caulerpa lentillifera sample is similar to that of the stage (groove shape), and the difference in the linear attenuation coefficient (μ value) of the two for X-rays is extremely small, resulting in a reduced contrast in the CT image and making it difficult to distinguish the boundaries. The stage is the sample bed. Therefore, the Caulerpa lentillifera sample and the stage are physically separated. Material selection: A diaphragm plate is made of a low-attenuation coefficient material - polyethylene plastic wrap to reduce its density similarity to the sample. Starting from one end of the groove, the plastic wrap is slowly wound along the groove contour, and the plastic wrap is appropriately stretched during the process to make it closely adhere to the groove, forming a support plane. The support is fixed in suspension to avoid direct contact, separating the sample from the groove of the sample bed for convenient subsequent image processing and data processing. Figure 2 In part A, it shows the imaging effect of the Caulerpa lentillifera sample and the stage before using the polyethylene film, and in part B, it shows the imaging effect of the Caulerpa lentillifera sample and the stage after using the polyethylene film. It can be seen that the polyethylene plastic wrap can fully separate the groove of the sample bed from the Caulerpa lentillifera sample.

[0044] (2) Place the processed sample and close the chamber door.

[0045] (3) Establish a data storage path and sample description. Establish a database (Database), establish a sample description, and perform parameter settings. Set the voltage to 90V and the current to 88uA on the shooting interface (ensuring that the voltage current ≤ 8W is sufficient), adjust the field of view to the minimum (FOV: 86mm; Voxel Size: 172um), rotate the camera 384°, and adjust the camera rotation speed to 21° / s. FOV represents the field of view angle, and Voxel Size represents the voxel resolution.

[0046] (4) Data acquisition.

[0047] After obtaining the vox file output by Quantum GX2, it is converted to the ply format through a script. After obtaining the vox file, a specially designed script is used to perform the conversion work. This script can efficiently convert the vox format file to the ply format, facilitating subsequent processing and analysis. The ply format is widely used in the field of 3D modeling and visualization due to its flexibility and wide support. Through such a conversion, the originally difficult-to-directly-use vox data can be read and manipulated by various 3D processing software, greatly expanding the application scope and possibilities of the data.

[0048] Through Quantum GX2, the sample processing speed can reach 600 per hour, far exceeding the current manual counting speed.

[0049] Specifically, the parameter settings include: adjusting the position of the stage and the radiation device according to the width and height of the upright branches of the long-stemmed grape fern. In the CT shooting test, the long-stemmed grape fern is a special research sample with unique physical and chemical properties, which puts forward specific requirements for CT scanning imaging. The water content of the algae is extremely high, up to 97%, and during the sample preparation process, it is difficult to completely remove the water attached to its upright branches.

[0050] 1. Given that water has special absorption characteristics for X-rays, especially low-energy X-rays, its photoelectric absorption effect is significant, which will lead to a strong attenuation effect. When performing CT scanning on long-stem grape fern algae, considering the actual situation that the sample thickness is greater than 1 cm, if too low a tube voltage (such as less than 70kV) is used to excite X-rays, due to the high absorption of water to low-energy X-rays, the energy of X-rays will decay too quickly in the process of penetrating the sample, and it is very likely that they will not be able to effectively penetrate the sample, resulting in the loss of imaging information or serious artifacts, affecting the accurate observation of the internal structure of the sample. On the contrary, moderately increasing the tube voltage to a higher level can, on the one hand, ensure that the X-rays have sufficient energy to penetrate algae samples containing a large amount of water and a certain thickness, and ensure that the ray projection data required for imaging is fully acquired; in addition, the morphological structure of long-stem grape fern algae is relatively special, and it contains many spherical branches, and a large amount of water is often accumulated between these spherical branches. Higher tube voltage helps to reduce the negative impact of this part of water on imaging, and by enhancing the penetration of X-rays, reducing the scattering and attenuation caused by the accumulation of water layers between spherical twigs, further improving the clarity and accuracy of the image, and avoiding the problem of blurred images due to interference from local water layers. Therefore, in order to achieve both X-rays penetrating the test sample and ensuring clear imaging, the voltage is set to 90kV and the current is set to 88μA. The relatively high voltage of 90kV can produce X-rays with stronger energy and better penetration; the current of 88μA can ensure the intensity of the X-rays, and cooperate with the voltage to obtain clear images at a suitable dose.

[0051] 2. The diameter of the X-ray scanning area was set to 86 mm, the voxel size was set to 172 μm, and the camera was rotated to scan and obtain the vox 3D model file format of the upright branches. It took about 18 seconds, the rotation step was 384°, and the camera speed was adjusted to 21° / s. The filter 3D-Soft: This filter can smooth the reconstructed image and reduce noise, which helps to better observe the morphology and boundaries of the upright branches. X-ray filter Cu 0.06+Al 0.5: This filter can filter out low-energy X-rays, optimize the X-ray energy spectrum, improve the contrast and clarity of the imaging, and reduce unnecessary radiation doses, reducing radiation hazards to small animals and operators.

[0052] 3. During the CT scanning process, to ensure image quality, attention should also be paid to the fixation and positioning of the samples. The upright branches of Caulerpa lentillifera should be firmly placed on the stage to avoid movement or deformation during scanning, which may affect the final imaging effect. During positioning, it is necessary to ensure that the center of the upright branch coincides with the center of the X-ray beam to ensure the integrity of the scanning area and the symmetry of the image.

[0053] In an exemplary embodiment, step 102 specifically includes: performing boundary annotation on the spherical branches of each three-dimensional point cloud data, and taking the bounding boxes of each spherical branch and the total number of spherical branches as label data.

[0054] Each sample in the dataset includes input data and label data. The input data is three-dimensional point cloud data, and the label data is the bounding boxes of each spherical branch and the total number of spherical branches.

[0055] The point cloud segmentation network adopts the PointNet++ network.

[0056] In an exemplary embodiment, performing boundary annotation on the spherical branches of each three-dimensional point cloud data specifically includes: using CloudCompare software to perform boundary annotation on the spherical branches of each three-dimensional point cloud data.

[0057] More specifically, first perform three-dimensional point cloud segmentation to remove the grooves. Then, perform the following similar annotation on the three-dimensional point cloud dataset, and assign a label, such as "1", to each target (spherical branch), while the other background is 0. Finally, convert the point cloud data.ply to a.txt file one by one and divide it into a training set and a test set.

[0058] (1) Open the open-source software Cloudcompare, click the open button, that is, "Open" to select the point cloud to be annotated. In the display window, click and drag the left mouse button to rotate the point cloud, and click and drag the right mouse button to move the viewing angle to adjust the point cloud to a suitable viewing angle for annotation.

[0059] (2) Select the point cloud file to be annotated, as Figure 3 shown, use the scissors button to use the polygon selection method to select and separate the part of the spherical branch; as Figure 4 shown, cancel the selection of the previous point cloud and select the segmented point cloud; as Figure 5 shown, enter "1" in the value category option box of the spherical branch part.

[0060] (3) After annotating all the targets, enter "0" in the value category option box of the remaining upright branch axis (without spherical branches).

[0061] (4) Click on the "merge" icon, click on "do not generate a scalar field with the original cloud index", which means the marking is completed. After adjustment, click on the save button, i.e., "Save" to save. As Figure 6 shown, save it in ply format.

[0062] In an exemplary embodiment, the three-dimensional point cloud data of the upright branches of Caulerpa lentillifera is converted into txt format and divided into a semantic segmentation training set, a validation set, and a test set according to a ratio of 7:2:1, which are respectively used for the training and evaluation of the subsequent semantic segmentation model.

[0063] (1) Production of data export format.

[0064] (2) Production of dataset structure.

[0065] First, write the dataset division file list.

[0066] Secondly, re-place the data of the present application according to the structure of its KITTI dataset.

[0067] Finally, write the dataset loading.yaml file according to the input method of the PointNet++ network.

[0068] (3) Divide the training set, validation set, and test set according to a ratio of 7:2:1.

[0069] In an exemplary embodiment, the training environment is set up as follows: (1) Install basic operating software: Pycharm + software and Anaconda software.

[0070] (2) Install the Graphics Processing Unit (GPU) operating environment: Compute Unified Device Architecture (CUDA) and CUDA Deep Neural Network library (cuDNN).

[0071] (3) Set up an Anaconda virtual operating environment with Pycharm as the carrier: which includes the deep learning library pytorch and other basic libraries. The Python interpreter in the Anaconda environment uses Interpreter.

[0072] As shown in Table 1, the algorithm running task of this application is executed on the Windows 10 system, with the Python version being 3.11. The PyTorch 1.6.0 framework and CUDA 10.1 are adopted. The server selects an Intel (R) Core (TM) i7-9750H CPU @2.60GHz processor and an NVIDIA GeForce RTX 2060 graphics card.

[0073] Table 1 Training Environment

[0074] After downloading the code of the PointNet++ network in this application and opening it with Pycharm, modify the path of the training file and the type of the dataset.

[0075] During the training process of the PointNet++ network, the PointNet++ network receives the original three-dimensional point cloud data as input. Then, the network starts to process the data layer by layer. Through a series of transformation and feature extraction operations, the original three-dimensional point cloud data is gradually transformed into a higher-level feature representation, learning and identifying the points belonging to the spherical branches; after being processed by the network, each point will be assigned a semantic label, so as to know which points belong to the spherical branches and which points belong to other parts; according to the semantic segmentation result of the PointNet++ network, the points belonging to the spherical branches are extracted, and then the extracted points are segmented by Euclidean distance clustering.

[0076] Using the PointNet++ network for 3D point cloud object detection, after data preprocessing (converting to txt files and dataset division), 3D object detection training can be carried out. By adjusting the hyperparameters, after several hundred rounds of training, the trained PointNet++ network model reaches the optimal effect, and an object segmentation model is obtained.

[0077] Load the trained pth file and successfully infer and detect the number of Caulerpa lentillifera.

[0078] The verification result of this application shows that the segmentation accuracy of the trained PointNet++ network is about 96.68%, indicating that the object segmentation model can meet the segmentation requirements of this application. And a higher precision rate means that the possibility of misjudgment of this object segmentation model is lower. A higher mean intersection over union of this object segmentation model means that the model has better performance in dealing with the annotation problem of the spherical branches of Caulerpa lentillifera with long stems. For scenarios with multiple similar objects, it can more accurately evaluate the ability of the model to distinguish and segment each instance.

[0079] This application applies the non-invasive in-vivo imaging technology of small animals to the determination of algal traits, which is innovative. With the Quantum GX2, the sample processing speed can far exceed the current manual counting speed. Experiments have found that the 3D imaging function of the Quantum GX2 system can be used to quickly and non-destructively obtain the three-dimensional data of upright branches and convert them into point cloud data, so as to achieve the counting of spherical branches and conduct the phenotypic determination of upright branches. Under the condition of using an extremely low X-ray dose to ensure real long-time microCT imaging, the Quantum GX2 imaging system can still provide high-resolution images and faster scanning speed, and can achieve high-throughput counting of the number of spherical branches of upright branches, providing an ideal platform for the evaluation of upright branch traits.

[0080] To evaluate the accuracy of the target segmentation model measured after training the PointNet++ network, 20 upright branches were used as samples in this experiment. The number of spherical branches was obtained by manual counting and the target segmentation model respectively. After the CT system completed the scanning, manual spherical branch counting was performed on all upright branches. The counting method was to use tweezers to remove the spherical branches on the upright branches one by one, as Figure 7 shown, and the time consumed by each was recorded respectively. Common indicators for evaluating measurement accuracy include the coefficient of determination (R 2 ), mean absolute percentage error (MAPE), and root mean square error (RMSE).

[0081] The R 2 between the manual value and the model calculated value of the number of spherical branches of Mitracula sp. is 0.9276. The closer the R² value is to 1, the stronger the ability of the target segmentation model to interpret the data and the better the fitting effect. RMSE and MAPE are important indicators for evaluating the prediction error of the target segmentation model. In this experiment, RMSE is 4.23 and MAPE is 6.37%, indicating that the fitting effect of the target segmentation model of Mitracula sp. is good.

[0082] Based on the same inventive concept, the embodiment of this application also provides a measuring device for the number of spherical branches of Caulerpa lentillifera for implementing the measuring method of the number of spherical branches of Caulerpa lentillifera involved above. The implementation solutions provided by this device to solve problems are similar to the implementation solutions recorded in the above method. Therefore, the specific limitations in one or more embodiments of the measuring device for the number of spherical branches of Caulerpa lentillifera provided below can refer to the limitations on the measuring method of the number of spherical branches of Caulerpa lentillifera in the above text and will not be elaborated here.

[0083] In an exemplary embodiment, as Figure 8 shown, a measuring device for the number of spherical branches of Caulerpa lentillifera is provided. The measuring device for the number of spherical branches of Caulerpa lentillifera applies any one of the measuring methods for the number of spherical branches of Caulerpa lentillifera. The measuring device for the number of spherical branches of Caulerpa lentillifera includes the following modules.

[0084] A three-dimensional point cloud data acquisition module for acquiring three-dimensional point cloud data of each Caulerpa lentillifera sample by using a small animal in vivo microCT imager.

[0085] A data set construction module for constructing a data set by performing spherical branch annotation on each three-dimensional point cloud data.

[0086] A model training module for training a point cloud segmentation network by using the data set to obtain a target segmentation model.

[0087] A detection module for detecting the number of spherical branches of the Caulerpa lentillifera to be detected by using the target segmentation model.

[0088] The technical features of the above embodiments can be combined arbitrarily. For the sake of concise description, 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, it should be considered as the scope described in this specification.

[0089] In this article, specific examples are used to elaborate on the principle and implementation manner of the present application. The description of the above embodiments is only used to help understand the method and its core idea of the present application; at the same time, for those of ordinary skill in the art, according to the idea of the present application, there will be changes in the specific implementation manner and application scope. In summary, the content of this specification should not be construed as a limitation to the present application.

Claims

1. A method for measuring the number of spherical branches of Caulerpa lentillifera, characterized in that, The method for measuring the number of spherical branches of Caulerpa lentillifera includes: Collecting three-dimensional point cloud data of each Caulerpa lentillifera sample using a small animal in-vivo microCT imager; Constructing a data set by annotating spherical branches for each three-dimensional point cloud data; Training a point cloud segmentation network using the data set to obtain a target segmentation model; Detecting the number of spherical branches of the Caulerpa lentillifera to be detected using the target segmentation model.

2. The method for measuring the number of spherical branches of Caulerpa lentillifera according to claim 1, characterized in that, Collecting three-dimensional point cloud data of each Caulerpa lentillifera sample using a small animal in-vivo microCT imager, specifically including: Starting from one end of the sample bed groove, winding a polyethylene film along the radial contour of the sample bed groove, and stretching the polyethylene film to form a support plane, where the support plane keeps the Caulerpa lentillifera sample from contacting the bottom of the sample bed groove, to obtain a processed Caulerpa lentillifera sample; After placing the processed Caulerpa lentillifera sample into the chamber of the small animal in-vivo microCT imager, setting the working parameters of the small animal in-vivo microCT imager according to preset working parameter conditions; the preset working parameter conditions include that the voltage and current meet the set conditions, the field of view is adjusted to the minimum, the camera rotation speed is the set rotation speed, and the camera rotation angle is the set angle; Obtaining the three-dimensional point cloud data obtained after the small animal in-vivo microCT imager performs CT scanning on the Caulerpa lentillifera sample.

3. The method for measuring the number of spherical branches of Caulerpa lentillifera according to claim 2, characterized in that The voltage and current satisfy the set conditions, where the voltage current ≤ 8W, the set rotational speed is 21° / s, and the set angle is 384°.

4. The method for measuring the number of spherical branches of Caulerpa lentillifera according to claim 2, characterized in that The preset working parameter conditions further include that the diameter of the X-ray scanning area is set to 86 mm and the voxel size is set to 172 μm.

5. The method for measuring the number of spherical branches of Caulerpa lentillifera according to claim 1, characterized in that, The point cloud segmentation network uses the PointNet++ network.

6. The method for measuring the number of spherical branches of Caulerpa lentillifera according to claim 1, characterized in that, Detecting the number of spherical branches of the Caulerpa lentillifera to be detected using the target segmentation model, specifically including: Collecting three-dimensional point cloud data of the Caulerpa lentillifera to be detected using a small animal in-vivo microCT imager; Inputting the three-dimensional point cloud data of the Caulerpa lentillifera to be detected into the target segmentation model to obtain the total number of spherical branches of the Caulerpa lentillifera to be detected.

7. The method for measuring the number of spherical branches of Caulerpa lentillifera according to claim 1, characterized in that The three-dimensional point cloud data is three-dimensional point cloud data in ply format.

8. The method for measuring the number of spherical branches of Caulerpa lentillifera according to claim 1, wherein Constructing a data set by annotating spherical branches for each three-dimensional point cloud data, specifically including: Performing boundary annotation on the spherical branches of each three-dimensional point cloud data, and taking the bounding box of each spherical branch and the total number of spherical branches as label data.

9. The method for measuring the number of spherical branches of Caulerpa lentillifera according to claim 8, wherein Performing boundary annotation on the spherical branches of each three-dimensional point cloud data, specifically including: Using CloudCompare software to perform boundary annotation on the spherical branches of each three-dimensional point cloud data.

10. A measuring device for the number of spherical branches of Caulerpa lentillifera, characterized in that, The device for measuring the number of spherical branches of Caulerpa lentillifera applies the method for measuring the number of spherical branches of Caulerpa lentillifera according to any one of claims 1-9, and the device for measuring the number of spherical branches of Caulerpa lentillifera includes: A three-dimensional point cloud data acquisition module for collecting three-dimensional point cloud data of each Caulerpa lentillifera sample using a small animal in-vivo microCT imager; A data set construction module for constructing a data set by annotating spherical branches for each three-dimensional point cloud data; A model training module for training a point cloud segmentation network using the data set to obtain a target segmentation model; A detection module, configured to detect the number of spherical branches of the Caulerpa lentillifera to be detected by using the target segmentation model.

Images