Respirator mask manufacturing parameter determination method and device, computer equipment and readable storage medium

Abstract

The invention relates to a respirator mask manufacturing parameter determination method and device, computer equipment, a computer readable storage medium and a computer program product. The method comprises the following steps: acquiring a facial image of a target object, and performing facial three-dimensional reconstruction based on the facial image to obtain a facial three-dimensional model; face area segmentation is carried out on the face three-dimensional model, and a mask area in the face three-dimensional model and a nose mask area and a mouth shield area in the mask area are determined; based on the three-dimensional coordinates of the facial feature points in the areas covered by the mask area, the nose mask area and the mouth shield area, target size parameters of the mask area, the nose mask area and the mouth shield area are determined respectively; according to the target size parameters of the mask area, the nose mask area and the mouth shield area, manufacturing parameters of the respirator mask are determined. The respirator mask manufactured through the method can be attached to the face of the target object, so that air leakage of the respirator mask during use is avoided, and the use effect of a respirator is improved.

Description

Technical Field

[0001] The present application relates to the technical field of mask manufacturing, and in particular to a method, device, computer equipment, computer-readable storage medium and computer program product for determining manufacturing parameters of a ventilator mask. Background Art

[0002] During the non-invasive use of a ventilator, the user needs to wear a ventilator mask (such as an oral and nasal mask or a nasal mask) connected to the breathing circuit for assisted ventilation. Current ventilator masks generally have several fixed sizes for users to choose from, such as large, medium, and small. However, due to the different facial shapes of users, fixed-size ventilator masks are difficult to fit the faces of all users. Inappropriate-sized ventilator masks may leak during use, affecting the effectiveness of the ventilator. Therefore, there is an urgent need for ventilator masks that can fit the user's face and improve the effectiveness of the ventilator. Summary of the invention

[0003] Based on this, it is necessary to provide a method, device, computer equipment, computer-readable storage medium and computer program product for determining the manufacturing parameters of a ventilator mask, which can make the ventilator mask fit the user's face and thus improve the use effect of the ventilator, in order to address the above technical problems.

[0004] In a first aspect, the present application provides a method for determining manufacturing parameters of a ventilator mask, comprising:

[0005] Acquire a facial image of a target object, perform three-dimensional facial reconstruction based on the facial image, and obtain a three-dimensional facial model; the three-dimensional facial model includes three-dimensional coordinates of a plurality of facial feature points in the facial image;

[0006] Performing facial region segmentation on the three-dimensional facial model to determine the mask region in the three-dimensional facial model, as well as the nose mask region and the mouth guard region within the mask region;

[0007] Determine the target size parameters of the mask area, the nasal mask area and the mouth stop area respectively based on the three-dimensional coordinates of the facial feature points in the areas covered by the mask area, the nasal mask area and the mouth stop area;

[0008] The manufacturing parameters of the ventilator mask are determined according to the respective target size parameters of the mask area, the nasal mask area and the mouth stop area.

[0009] In one embodiment, performing facial region segmentation on the three-dimensional facial model to determine the mask region in the three-dimensional facial model, and the nose mask region and the mouth guard region in the mask region include:

[0010] According to the facial three-dimensional structure formed by facial feature points in the facial three-dimensional model, the facial three-dimensional model is segmented into facial regions to obtain a facial region segmentation result;

[0011] According to the facial region segmentation result, a mask region and a nose mask region within the mask region are determined in the three-dimensional facial model;

[0012] Based on the three-dimensional facial structure formed by facial feature points in the mask area, mouth stop area detection is performed in the mask area to determine the mouth stop area.

[0013] In one of the embodiments, according to the facial three-dimensional structure formed by facial feature points in the facial three-dimensional model, the facial three-dimensional model is segmented into facial regions to obtain a facial region segmentation result, including:

[0014] Inputting the facial 3D model into the facial region segmentation model, so that the facial region segmentation model performs facial region segmentation on the facial 3D model according to the facial 3D structure formed by facial feature points in the facial 3D model, and outputs the facial region segmentation result;

[0015] The facial region segmentation model is generated in the following way:

[0016] Acquire a data set including a plurality of preset three-dimensional facial models; each preset three-dimensional facial model has a mask area and a nose mask area marked;

[0017] Based on the data set, a facial region segmentation model is trained.

[0018] In one embodiment, based on the three-dimensional facial structure formed by facial feature points in the mask area, performing mouth stop area detection in the mask area to determine the mouth stop area includes:

[0019] Get the basic outline of the mouth block area;

[0020] According to the facial three-dimensional structure formed by the basic contour of the mouth stop area and the facial feature points in the mask area, contour point detection is performed on the facial feature points in the mask area to determine the target feature points in the mask area that belong to the edge of the mouth stop area;

[0021] Based on the target feature points, the mouth blocking area is determined.

[0022] In one embodiment, based on the three-dimensional coordinates of facial feature points in the areas covered by the mask area, the nasal mask area and the mouth stop area, target size parameters of the mask area, the nasal mask area and the mouth stop area are determined respectively, including:

[0023] Based on the three-dimensional coordinates of the facial feature points in the areas covered by the mask area, the nose mask area and the mouth stop area, respectively, the initial size parameters of the mask area, the nose mask area and the mouth stop area are determined;

[0024] Determining a size conversion ratio between the three-dimensional facial model and the actual face of the target subject;

[0025] According to the size conversion ratio, the initial size parameters of the mask area, the nasal mask area and the mouth stop area are converted to obtain the target size parameters of the mask area, the nasal mask area and the mouth stop area.

[0026] In one embodiment, the manufacturing parameters of the ventilator mask are determined according to the target size parameters of the mask area, the nasal mask area and the mouth stop area, including:

[0027] Get the preset material thickness of the face mask, nose mask and mouth guard of the ventilator mask;

[0028] According to the preset material thicknesses of the face mask, nose mask and mouth stopper in the ventilator mask, the target size parameters of the face mask area, nose mask area and mouth stopper area are adjusted to obtain the manufacturing parameters of the ventilator mask.

[0029] In one embodiment, the method for determining the manufacturing parameters of a ventilator mask further includes:

[0030] Based on a plurality of facial images whose total number exceeds a threshold, a mask making parameter determination model is trained;

[0031] The facial image of the target object is input into the mask production parameter determination model to obtain the production parameters output by the mask production parameter determination model; the production parameters are used to produce a ventilator mask matching the face of the target object.

[0032] In a second aspect, the present application also provides a device for determining manufacturing parameters of a ventilator mask, comprising:

[0033] A facial 3D reconstruction module is used to obtain a facial image of a target object, and perform facial 3D reconstruction based on the facial image to obtain a facial 3D model; the facial 3D model includes the 3D coordinates of multiple facial feature points in the facial image;

[0034] A facial region segmentation module is used to segment the facial region of the three-dimensional facial model, and determine the mask region in the three-dimensional facial model, as well as the nose mask region and the mouth guard region within the mask region;

[0035] A size parameter determination module, for determining target size parameters of the mask area, the nasal mask area and the mouth stop area respectively based on the three-dimensional coordinates of facial feature points in the areas covered by the mask area, the nasal mask area and the mouth stop area;

[0036] The manufacturing parameter determination module is used to determine the manufacturing parameters of the ventilator mask according to the target size parameters of the mask area, the nasal mask area and the mouth stop area.

[0037] In a third aspect, the present application further provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following steps are implemented:

[0038] Acquire a facial image of a target object, perform three-dimensional facial reconstruction based on the facial image, and obtain a three-dimensional facial model; the three-dimensional facial model includes three-dimensional coordinates of a plurality of facial feature points in the facial image;

[0039] Performing facial region segmentation on the three-dimensional facial model to determine the mask region in the three-dimensional facial model, as well as the nose mask region and the mouth guard region within the mask region;

[0040] Determine the target size parameters of the mask area, the nasal mask area, and the mouth stop area based on the three-dimensional coordinates of the facial feature points in the areas covered by the mask area, the nasal mask area, and the mouth stop area respectively;

[0041] The manufacturing parameters of the ventilator mask are determined according to the respective target size parameters of the mask area, the nasal mask area and the mouth stop area.

[0042] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the following steps are implemented:

[0043] Acquire a facial image of a target object, perform three-dimensional facial reconstruction based on the facial image, and obtain a three-dimensional facial model; the three-dimensional facial model includes three-dimensional coordinates of a plurality of facial feature points in the facial image;

[0044] Performing facial region segmentation on the three-dimensional facial model to determine the mask region in the three-dimensional facial model, as well as the nose mask region and the mouth guard region within the mask region;

[0045] Determine the target size parameters of the mask area, the nasal mask area, and the mouth stop area based on the three-dimensional coordinates of the facial feature points in the areas covered by the mask area, the nasal mask area, and the mouth stop area respectively;

[0046] The manufacturing parameters of the ventilator mask are determined according to the respective target size parameters of the mask area, the nasal mask area and the mouth stop area.

[0047] In a fifth aspect, the present application further provides a computer program product, including a computer program, which implements the following steps when executed by a processor:

[0048] Acquire a facial image of a target object, perform three-dimensional facial reconstruction based on the facial image, and obtain a three-dimensional facial model; the three-dimensional facial model includes three-dimensional coordinates of a plurality of facial feature points in the facial image;

[0049] Performing facial region segmentation on the three-dimensional facial model to determine the mask region in the three-dimensional facial model, as well as the nose mask region and the mouth guard region within the mask region;

[0050] Determine the target size parameters of the mask area, the nasal mask area, and the mouth stop area based on the three-dimensional coordinates of the facial feature points in the areas covered by the mask area, the nasal mask area, and the mouth stop area respectively;

[0051] The manufacturing parameters of the ventilator mask are determined according to the respective target size parameters of the mask area, the nasal mask area and the mouth stop area.

[0052] The above-mentioned method, device, computer equipment, computer-readable storage medium and computer program product for determining manufacturing parameters of a ventilator mask first obtain a facial image of a target object, and then perform a three-dimensional facial reconstruction based on the facial image to obtain a three-dimensional facial model, wherein the three-dimensional facial model includes the three-dimensional coordinates of a plurality of facial feature points in the facial image. Further, the facial three-dimensional model is segmented into facial regions, and the mask region in the facial three-dimensional model, as well as the nasal mask region and the mouth stop region in the mask region, are determined, so that the target size parameters of the mask region, the nasal mask region and the mouth stop region are determined based on the three-dimensional coordinates of the facial feature points in the regions covered by the mask region, the nasal mask region and the mouth stop region, respectively, so as to determine the manufacturing parameters of the ventilator mask according to the target size parameters of the mask region, the nasal mask region and the mouth stop region. In the whole process, the manufacturing parameters of the ventilator mask can be determined based on the three-dimensional facial model of the target object, so that the ventilator mask finally obtained fits the face of the target object, thereby avoiding air leakage of the ventilator mask during use, thereby improving the use effect of the ventilator. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the drawings required for use in the embodiments of the present application or related technical descriptions will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0054] FIG1 is a diagram showing an application environment of a method for determining manufacturing parameters of a ventilator mask according to an embodiment;

[0055] FIG2 is a schematic flow chart of a method for determining manufacturing parameters of a ventilator mask in one embodiment;

[0056] FIG3 is a schematic diagram of a process of performing facial region segmentation on a facial 3D model in one embodiment;

[0057] FIG4 is a schematic diagram of a process for obtaining a facial region segmentation model in one embodiment;

[0058] FIG5 is a schematic diagram of a process for detecting a mouth barrier area in one embodiment;

[0059] FIG6 is a schematic diagram of a process for determining target size parameters of an oral and nasal mask in one embodiment;

[0060] FIG7 is a schematic diagram of a process for determining target size parameters of a nasal mask in one embodiment;

[0061] FIG8 is a schematic diagram of a process of determining a model and a facial image of a target object based on mask manufacturing parameters and obtaining manufacturing parameters in one embodiment;

[0062] FIG9 is a flow chart of a method for determining manufacturing parameters of a ventilator mask in another embodiment;

[0063] FIG10 is a block diagram of a device for determining manufacturing parameters of a ventilator mask according to an embodiment;

[0064] FIG. 11 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0065] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0066] The method for determining the manufacturing parameters of a ventilator mask provided in an embodiment of the present application can be applied in an application environment as shown in Figure 1. Among them, a terminal 102 equipped with an image acquisition device can communicate with a server 104 through a network, and the image acquisition device can specifically be a camera. The data storage system can store the data that the server 104 needs to process. The data storage system can be integrated on the server 104, or it can be placed on the cloud or other network servers. The terminal 102 can be, but is not limited to, various personal computers, laptops, smart phones, tablet computers, Internet of Things devices, etc. The server 104 can be an independent physical server, or it can be a server cluster or distributed system composed of multiple physical servers, or it can be a cloud server that provides cloud computing services.

[0067] Specifically, the server 104 can communicate with the terminal 102 to obtain the facial image of the target object collected by the terminal 102, thereby performing facial three-dimensional reconstruction based on the facial image to obtain a facial three-dimensional model of the target object, wherein the facial three-dimensional model includes the three-dimensional coordinates of multiple facial feature points in the facial image of the target object. Further, the server 104 can perform facial region segmentation on the facial three-dimensional model to determine the mask area in the facial three-dimensional model, and the nasal mask area and the mouth stop area in the mask area. Further, the server 104 can determine the target size parameters of the mask area, the nasal mask area and the mouth stop area based on the three-dimensional coordinates of the facial feature points in the areas covered by the mask area, the nasal mask area and the mouth stop area, respectively, so as to determine the manufacturing parameters of the ventilator mask according to the target size parameters of the mask area, the nasal mask area and the mouth stop area.

[0068] In an exemplary embodiment, as shown in FIG2 , a method for determining manufacturing parameters of a ventilator mask is provided, and the method is applied to the server 104 in FIG1 as an example for description, including the following steps 202 to 208. Among them:

[0069] Step 202, obtain a facial image of the target object, perform three-dimensional facial reconstruction based on the facial image, and obtain a three-dimensional facial model; the three-dimensional facial model includes three-dimensional coordinates of multiple facial feature points in the facial image.

[0070] The number of facial images of the target object is at least one. If there is only one, the facial image must be an image of the front face of the target object. If there are multiple, the facial images of the target object can be collected from multiple angles, so as to accurately establish a facial 3D model of the target object later.

[0071] Optionally, after acquiring the facial image of the target object through the terminal, the server may pre-process the facial image of the target object, and then input the processed facial image into a pre-built facial 3D reconstruction model to obtain a facial 3D model of the target object output by the facial 3D reconstruction model. The pre-processing may include: image denoising, image denoising, image resizing, image rotation and flipping, etc.

[0072] In a specific application scenario, the facial 3D reconstruction model in this embodiment can automatically perform face area detection on the input facial image, and then use the feature point detection algorithm to locate the key facial feature points in the facial image within the detected face area, and automatically map the key facial feature points to the three-dimensional space to construct a facial 3D model of the target object.

[0073] Among them, the key facial feature points include but are not limited to: the middle point and contour point of the eyebrows; the nose tip point, the contour points of the bridge of the nose and the wings of the nose, and the raised point on the bridge of the nose; the pupil center point, the eye corner point, and the contour points of the upper and lower eyelids of the eyes; the mouth corner points and the contour points of the upper and lower lips of the mouth; the facial edge points; and the preset sampling points in other facial areas. Face area detection can be implemented based on face detection algorithms, and the face detection algorithms that can be used include but are not limited to: Haar feature classifier (a machine learning algorithm for target detection, especially face detection), MTCNN (Multi-task Cascaded Convolutional Networks, a face detection method based on cascaded convolutional neural networks), SSD (Single Shot MultiBox Detector, a method for detecting multiple targets in a single shot), CNN (Convolutional Neural Network). Feature point detection algorithms include but are not limited to: SIFT (Scale Invariant Feature Transform, a feature point detection algorithm based on scale invariance), FAST (Features from Accelerated Segment Test, a fast feature point detection method), Harris corner point detection algorithm. Regarding how to map key facial feature points to three-dimensional space, it can be implemented based on any of 3DMM (3D Morphable Model), deep learning (DL), and epipolar geometry. 3DMM is a generative model for facial appearance and shape, 3D generative adversarial networks and 3D convolutional neural networks in deep learning technology can be used for 3D facial reconstruction, and epipolar geometry can use multiple non-synthetic perspective images of the same subject to generate a single 3D image.

[0074] Step 204, performing facial region segmentation on the three-dimensional facial model to determine the mask region in the three-dimensional facial model, and the nose mask region and the mouth guard region within the mask region.

[0075] Among them, the ventilator mask can include: an oronasal mask that covers the nose and mouth, and a nasal mask that only covers the nose. The oronasal mask is provided with a mouth stop area, and the mouth stop area can be provided with a silicone mouth stop, which can fit the face of the ventilator mask user tightly, especially in the mouth and nose area, to form an effective seal. This sealing performance helps prevent air leakage and ensure the use effect of the ventilator.

[0076] Optionally, the server can perform facial region segmentation on the target object's facial three-dimensional model based on a pre-trained facial region segmentation model and the facial three-dimensional structure formed by facial feature points in the facial three-dimensional model (such as the facial central axis, the shape of facial organs, etc.), and determine the mask area in the facial three-dimensional model, as well as the nasal mask area and the mouth stop area within the mask area, so that an oral and nasal mask (with a silicone mouth stop inside) suitable for the target object can be produced according to the mask area and the mouth stop area in the facial three-dimensional model, and a nasal mask suitable for the target object can be produced according to the nasal mask area in the facial three-dimensional model.

[0077] Step 206, based on the three-dimensional coordinates of the facial feature points in the areas covered by the mask area, the nose mask area and the mouth stop area, respectively determine the target size parameters of the mask area, the nose mask area and the mouth stop area.

[0078] Optionally, the server can perform spatial geometric calculations based on the three-dimensional coordinates of facial feature points in the areas covered by the mask area, nasal mask area and mouth stop area, respectively determine the width, length and depth of the mask area, nasal mask area and mouth stop area in the three-dimensional facial model, and further convert the width, length and depth of the mask area, nasal mask area and mouth stop area in the three-dimensional facial model into actual sizes, thereby determining the target size parameters of the mask area, nasal mask area and mouth stop area.

[0079] Step 208, determining the manufacturing parameters of the ventilator mask according to the target size parameters of the mask area, the nasal mask area and the mouth stop area.

[0080] The target size parameters of the mask area can be used to determine the manufacturing parameters of the entire oronasal mask, and the target size parameters of the mouth stop area can be used to determine the manufacturing parameters of the silicone mouth stop in the oronasal mask, so as to manufacture an oronasal mask that can fit the face of the target object and has good sealing. The target size parameters of the nasal mask area can be used to determine the manufacturing parameters of the nasal mask.

[0081] Optionally, to ensure that the final ventilator mask can fit the face of the target object, for the oronasal mask that can be manufactured, the server can adjust the target size parameters of the mask area and the mouth stop area according to the preset thickness of the required (oronasal) mask and the mouth stop (in the oronasal mask), for example, appropriately increase the size of the mask area and the mouth stop area, so as to determine the final manufacturing parameters of the oronasal mask in the ventilator mask; for the nasal mask that can be manufactured, the server can adjust the target size parameters of the nasal mask area according to the preset thickness of the required nasal mask, for example, appropriately increase the size of the nasal mask area, so as to determine the final manufacturing parameters of the nasal mask in the ventilator mask. Among them, the preset materials of the (oronasal) mask, nasal mask, and mouth stop (in the oronasal mask) in the ventilator mask need to provide sufficient comfort and sealing, and the preset material thickness can be configured according to actual needs and stored in a data storage system connected to the server for the server to obtain.

[0082] Exemplarily, if the server detects that the target object has chosen to collect and upload a facial image and agrees to a personalized ventilator mask, the server can quickly determine the manufacturing parameters of the ventilator mask (oronasal mask and / or nasal mask) based on the target object's facial image based on the above-mentioned ventilator mask manufacturing parameter determination method, and provide real-time feedback of the manufacturing parameters to the ventilator mask manufacturer, so that the manufacturer can promptly and accurately manufacture a ventilator mask suitable for the target object, ensuring that when the target object subsequently uses a ventilator non-invasively, the ventilator mask will not leak and will not affect the wearing comfort due to inappropriate size.

[0083] The above-mentioned method for determining the manufacturing parameters of the ventilator mask first obtains the facial image of the target object, and then performs a three-dimensional facial reconstruction based on the facial image to obtain a three-dimensional facial model, wherein the three-dimensional facial model includes the three-dimensional coordinates of multiple facial feature points in the facial image. Further, the facial three-dimensional model is segmented into facial regions to determine the mask region in the facial three-dimensional model, as well as the nasal mask region and the mouth stop region in the mask region, so as to determine the target size parameters of the mask region, the nasal mask region and the mouth stop region respectively based on the three-dimensional coordinates of the facial feature points in the regions covered by the mask region, the nasal mask region and the mouth stop region, so as to determine the manufacturing parameters of the ventilator mask according to the target size parameters of the mask region, the nasal mask region and the mouth stop region respectively. In the whole process, the manufacturing parameters of the ventilator mask can be determined based on the three-dimensional facial model of the target object, so that the ventilator mask finally obtained fits the face of the target object, thereby avoiding air leakage of the ventilator mask during use, thereby improving the use effect of the ventilator.

[0084] In an exemplary embodiment, performing facial region segmentation on the three-dimensional facial model to determine the mask region in the three-dimensional facial model, and the nose mask region and the mouth guard region in the mask region include:

[0085] According to the facial three-dimensional structure formed by facial feature points in the facial three-dimensional model, the facial three-dimensional model is segmented into facial regions to obtain a facial region segmentation result;

[0086] According to the facial region segmentation result, a mask region and a nose mask region within the mask region are determined in the three-dimensional facial model;

[0087] Based on the three-dimensional facial structure formed by facial feature points in the mask area, mouth stop area detection is performed in the mask area to determine the mouth stop area.

[0088] Optionally, the server can perform facial region segmentation on the facial three-dimensional model according to the facial three-dimensional structure composed of facial feature points in the facial three-dimensional model through the facial region segmentation model obtained by pre-training, and obtain the facial region segmentation result. Among them, the facial region segmentation model can be used to identify the mask area that the oral and nasal mask should cover in the facial three-dimensional model, and the nasal mask area that the nasal mask should cover. The facial three-dimensional model includes a plurality of facial feature points, and the facial region segmentation model can automatically mark the facial feature points in the mask area and the nasal mask area, respectively, and output the facial region segmentation result with the marking. The server can determine the mask area and the nasal mask area in the mask area in the facial three-dimensional model according to the facial region segmentation result output by the facial three-dimensional model. Further, the server can also perform mouth block area detection in the mask area based on the facial three-dimensional structure composed of facial feature points in the mask area to determine the mouth block area in the mask area.

[0089] Exemplarily, as shown in FIG3, a flow chart of facial region segmentation for a facial three-dimensional model is provided. After obtaining at least one facial image of the target object, the server can input the facial image of the target object into the facial three-dimensional reconstruction model to obtain a facial three-dimensional model output by the facial three-dimensional reconstruction model. Further, the server can input the facial three-dimensional model of the target object into the pre-trained facial region segmentation model, thereby obtaining a facial region segmentation result marked with a mask region (as shown in FIG3, covering the nose and mouth) and a nasal mask region (as shown in FIG3, covering the nose). Further, the server can further detect the mask region to determine the mouth stop region suitable for the target object. It should be noted that the facial three-dimensional model, nasal mask region, mask region, and mouth stop region in FIG3 are only examples. When actually making a ventilator mask, the shape of the oral and nasal mask (with a silicone mouth stop inside) and the nasal mask can be flexibly adjusted to ensure that the final ventilator mask fits the face of the target object and meets the use requirements.

[0090] In this embodiment, the mask area, nose mask area and mouth stop area suitable for the target object can be determined so as to personalize the ventilator mask for the target object, ensuring that when the target object uses the manufactured ventilator mask, there will be no leakage of the ventilator mask and the comfort will not be affected by inappropriate size.

[0091] In one of the embodiments, according to the facial three-dimensional structure formed by facial feature points in the facial three-dimensional model, the facial three-dimensional model is segmented into facial regions to obtain a facial region segmentation result, including:

[0092] Inputting the facial 3D model into the facial region segmentation model, so that the facial region segmentation model performs facial region segmentation on the facial 3D model according to the facial 3D structure formed by facial feature points in the facial 3D model, and outputs the facial region segmentation result;

[0093] The facial region segmentation model is generated in the following way:

[0094] Acquire a data set including a plurality of preset three-dimensional facial models; each preset three-dimensional facial model has a mask area and a nose mask area marked;

[0095] Based on the data set, a facial region segmentation model is trained.

[0096] Optionally, the server may first obtain facial images of multiple historical objects, and clean and preprocess the facial images of the multiple historical objects. Then, for each historical object, the server may input at least one preprocessed facial image of the historical object into a pre-constructed facial 3D reconstruction model to obtain a facial 3D model of each historical object. Furthermore, the facial 3D model of each historical object may be manually labeled with a mask area covering the mouth and nose, and a nasal mask area covering the nose. Based on this, the server may obtain a data set of multiple preset facial 3D models, each preset facial 3D model having annotated mask areas and nasal mask areas. Furthermore, the server may divide the data set according to a preset ratio to obtain a training set including multiple preset facial 3D models, a verification set including multiple preset facial 3D models, and a test set including multiple preset facial 3D models. The server can iteratively optimize the initial facial region segmentation model based on the training set until the model iteration optimization stop condition is reached (which can be flexibly configured according to actual training needs), and obtain at least one trained model, thereby performing model evaluation on the at least one trained model based on the validation set to select at least one test model from the at least one trained model, and performing model testing on the at least one test model based on the test set to determine the final facial region segmentation model from the at least one test model. The validation set can be used to filter out at least one best model from the at least one trained model, and the test set is used to perform model performance testing on the filtered best model. After obtaining the facial 3D model of the target object, the server can input the facial 3D model into the facial region segmentation model that has passed the model performance test, so that the facial region segmentation model performs facial region segmentation on the facial 3D model according to the facial 3D structure formed by the facial feature points in the facial 3D model, and outputs the facial region segmentation result.

[0097] Exemplarily, the initial facial region segmentation model can be built based on a deep learning architecture, including but not limited to: U-Net (Convolutional Networks for Biomedical Image Segmentation, a semantic segmentation model widely used in image segmentation tasks in the field of deep learning), FCN (Fully Convolutional Networks, a convolutional neural network mainly used for image semantic segmentation), SegNet (an image semantic segmentation model based on deep learning, which innovatively adopts an encoder-decoder structure and introduces a pooling index transfer mechanism, which can effectively improve the accuracy of the segmentation results and the computational efficiency of the model).

[0098] Exemplarily, as shown in FIG4 , a flow chart of obtaining a facial region segmentation model is provided, which mainly includes the following steps:

[0099] Step 402, collecting facial images of multiple historical objects, and cleaning and preprocessing the facial image of each historical object;

[0100] Step 404, inputting the preprocessed facial image of each historical object into the facial 3D reconstruction model to obtain the facial 3D model of each historical object;

[0101] Step 406, after labeling the face mask area and the nose mask area of ​​each historical object's facial 3D model, a data set is obtained;

[0102] Step 408, dividing the data set into a training set, a validation set, and a test set;

[0103] Step 410, training an initial facial region segmentation model based on a training set to obtain at least one training model;

[0104] Step 412, based on the validation set, select the best model from at least one training model, and based on the test set, perform a model performance test on the selected best model;

[0105] If the model performance test is passed, execute step 414 and use the model that passes the model performance test as the facial region segmentation model; otherwise, return to step 410.

[0106] In this embodiment, the initial facial region segmentation model can be trained, screened, and tested through a training set, a validation set, and a test machine to obtain a final facial region segmentation model, so that the facial region segmentation model can accurately perform facial region segmentation on the facial three-dimensional model, thereby facilitating accurate determination of the mask area and nasal mask area suitable for the target object.

[0107] In one embodiment, based on the three-dimensional facial structure formed by facial feature points in the mask area, performing mouth stop area detection in the mask area to determine the mouth stop area includes:

[0108] Get the basic outline of the mouth block area;

[0109] According to the facial three-dimensional structure formed by the basic contour of the mouth stop area and the facial feature points in the mask area, contour point detection is performed on the facial feature points in the mask area to determine the target feature points in the mask area that belong to the edge of the mouth stop area;

[0110] Based on the target feature points, the mouth blocking area is determined.

[0111] Among them, the mouth block area can be provided with a silicone mouth, which can improve the dryness and pain in the mouth and nose caused by the target object breathing with the mouth open when using the ventilator mask, thereby improving both the operation effect of the ventilator and the user experience of the target object.

[0112] Optionally, the server may first obtain the basic contour of the mouth guard area, and then perform contour point detection on the facial feature points in the mask area based on the three-dimensional facial structure composed of the basic contour of the mouth guard area and the facial feature points in the mask area, based on an edge detection algorithm or a shape-based algorithm, to determine the target feature points in the mask area that belong to the edge of the mouth guard area, and then determine the mouth guard area in the mask area based on the target feature points. Among them, the edge detection algorithm can be specifically a Canny edge detection algorithm, while the shape-based algorithm relies more on the overall shape information of the object and extracts the contour by fitting the shape model. In the shape-based algorithm, Hough Transform can be used to detect simple shapes such as lines and circles, while more complex shapes may require the use of active contour models (such as Snake models) or level set methods.

[0113] Exemplarily, as shown in FIG5 , a flow chart of detecting the mouth stop area is provided. The server can perform contour point detection on the mask area based on the basic contour of the mouth stop area shown in FIG5 , thereby determining the mouth stop area within the mask area, so that the production parameters of the silicone mouth stop can be determined later based on the facial feature points within the mouth stop area. It should be noted that the contour of the mask area and the basic contour of the mouth stop area in FIG5 are only examples. When actually making a ventilator mask, the mask area and the mouth stop area can be flexibly adjusted according to the shape of the ventilator mask, which can ensure that the final mouth and nose mask fits the face of the target object and meets the use requirements.

[0114] In this embodiment, the mouth stop area suitable for the target object can be determined based on the mask area suitable for the target object, which is conducive to the subsequent production of a silicone mouth stop customized for the target object based on the production parameters of the mask area, on the basis of improving the overall coverage effect of the mouth and nose mask, thereby obtaining an mouth and nose mask suitable for the target object.

[0115] In one embodiment, based on the three-dimensional coordinates of facial feature points in the areas covered by the mask area, the nasal mask area and the mouth stop area, target size parameters of the mask area, the nasal mask area and the mouth stop area are determined respectively, including:

[0116] Based on the three-dimensional coordinates of the facial feature points in the areas covered by the mask area, the nose mask area and the mouth stop area, respectively, the initial size parameters of the mask area, the nose mask area and the mouth stop area are determined;

[0117] Determining a size conversion ratio between the three-dimensional facial model and the actual face of the target subject;

[0118] According to the size conversion ratio, the initial size parameters of the mask area, the nasal mask area and the mouth stop area are converted to obtain the target size parameters of the mask area, the nasal mask area and the mouth stop area.

[0119] In the process of pre-building the facial 3D reconstruction model, the size of the facial 3D model of each historical object output by the facial 3D reconstruction model can be recorded, and the size of the facial 3D model of each historical object can be compared with the actual facial size to calculate a unified size conversion ratio, and the size conversion ratio can be stored in the data storage system.

[0120] Optionally, the server can determine the width (which can be calculated by subtracting the x-axis coordinates), length (which can be calculated by subtracting the y-axis coordinates), and depth (which can be calculated by subtracting the z-axis coordinates) of the mask area, nasal mask area, and mouth stop area in the facial three-dimensional model based on the three-dimensional coordinates of the facial feature points in the areas covered by the mask area, nasal mask area, and mouth stop area, respectively, with reference to the central axis of the facial three-dimensional model, thereby respectively determining the initial size parameters of the mask area, nasal mask area, and mouth stop area in the facial three-dimensional model. Further, the server can obtain the size conversion ratio between the pre-calculated facial three-dimensional model and the actual face of the target object, thereby performing size conversion on the initial size parameters of the mask area, nasal mask area, and mouth stop area according to the size conversion ratio, and obtaining the target size parameters of the mask area, nasal mask area, and mouth stop area in the actual usage scenario.

[0121] Exemplarily, as shown in FIG6, a flow chart for determining the target size parameters of the oronasal mask is provided. Taking the production of an oronasal mask in a ventilator mask as an example, the initial size parameters of the mask area may include: multiple widths, multiple lengths, and multiple depths in the mask area, and the initial size parameters of the mouth stop area may include: multiple widths, multiple lengths, and multiple depths in the mouth stop area. FIG6 shows the longest length, the widest width, the width of a certain place, and the depth of a certain place in the mask area (i.e., the height of a certain point on the bridge of the nose at the central axis of the three-dimensional facial model), as well as the longest length, the width of a certain place, and the depth of a certain place in the mouth stop area, but the size parameters obtainable in this embodiment are not limited to the several places shown in FIG6. Further, the server can summarize all the length data, width data, and depth data of the mask area and the mouth stop area to determine the initial size parameters of the mask area and the mouth stop area, and after size conversion, determine the target size parameters of the mask area and the mouth stop area, i.e., the target size parameters of the oronasal mask, so as to subsequently produce an oronasal mask (with a mouth stop silicone inside) that can fit the face of the target object.

[0122] Exemplarily, as shown in FIG7 , a flow chart for determining the target size parameters of a nasal mask is provided. Taking the manufacture of a nasal mask in a ventilator mask as an example, the initial size parameters of the nasal mask area may include: multiple widths, multiple lengths, and multiple depths in the nasal mask area. FIG7 shows the longest length of the nasal mask area, the width at a certain point, and the depth at a certain point (i.e., the height of a certain point on the bridge of the nose at the central axis of the three-dimensional facial model), but the size parameters obtainable in this embodiment are not limited to the several points shown in FIG7 . Furthermore, the server can aggregate all length data, width data, and depth data of the nasal mask area to determine the initial size parameters of the nasal mask area, and after size conversion, determine the target size parameters of the nasal mask area, i.e., the target size parameters of the nasal mask, so as to subsequently manufacture a nasal mask that can fit the face of the target object.

[0123] In this embodiment, the size conversion can be performed using the initial size parameters in the target object's facial three-dimensional model to accurately determine the target size parameters for making the ventilator mask for the target object, thereby facilitating subsequent customization of the ventilator mask (nasal mask and / or oral and nasal mask with a silicone mouth stopper) for the target object.

[0124] In one embodiment, the manufacturing parameters of the ventilator mask are determined according to the target size parameters of the mask area, the nasal mask area and the mouth stop area, including:

[0125] Get the preset material thickness of the face mask, nose mask and mouth guard of the ventilator mask;

[0126] According to the preset material thicknesses of the face mask, nose mask and mouth stopper in the ventilator mask, the target size parameters of the face mask area, nose mask area and mouth stopper area are adjusted to obtain the manufacturing parameters of the ventilator mask.

[0127] Among them, the preset materials of the face mask, nose mask and mouth guard in the ventilator mask must provide sufficient comfort and sealing, and the preset material thickness can be configured according to actual needs. The specific preset material thickness may vary depending on the manufacturer and material selection.

[0128] Optionally, for the oronasal mask that can be manufactured, the server can adjust the target size parameters of the mask area and the mouth stop area according to the preset thickness of the required mask (oronasal mask) and the mouth stop (silicone mouth stop in the oronasal mask). For example, the sizes of the mask area and the mouth stop area are appropriately increased to determine the final manufacturing parameters of the oronasal mask in the ventilator mask. For the nasal mask that can be manufactured, the server can adjust the target size parameters of the nasal mask area according to the preset thickness of the required nasal mask, for example, the size of the nasal mask area is appropriately increased to determine the final manufacturing parameters of the nasal mask in the ventilator mask.

[0129] For example, when adjusting the manufacturing parameters of a ventilator mask, it is necessary to consider the preset material thickness of the mask, nose mask and mouth stopper, and adjust the target size parameters accordingly to ensure the comfort and sealing of the manufactured ventilator mask. Taking the silicone mouth stopper in the oronasal mask as an example, if the material thickness of the silicone mouth stopper is about 2mm, the width, length and depth of the target size parameters of the mouth stopper area can be evenly increased by 2mm.

[0130] In this embodiment, taking into account the effect of material thickness on the size of the ventilator mask, the comfort of the final ventilator mask can be improved by adjusting the target size parameters of the mask area, the nasal mask area and the mouth stop area. It also ensures that the final ventilator mask can fit the face of the target object and avoid air leakage during use.

[0131] In a possible implementation, as shown in FIG8 , a flow chart of determining a model and a facial image of a target object based on mask making parameters and obtaining making parameters is provided, which mainly includes the following steps:

[0132] Step 802: Based on a plurality of facial images whose total number exceeds a threshold, a mask making parameter determination model is trained.

[0133] Among them, the threshold can be flexibly configured according to the actual application scenario.

[0134] In this embodiment, when enough facial images are collected, a regression model that can directly output ventilator mask manufacturing parameters based on facial images, i.e., a mask manufacturing parameter determination model, can also be trained based on a large number of facial images. Among them, the regression model can predict the dependent variable based on the independent variable, and the types of regression models include but are not limited to: convolutional neural network regression, visual transformer regression, restricted regression, polynomial regression, ridge regression, decision tree regression, random forest regression, and support vector machine regression.

[0135] Step 804, inputting the facial image of the target object into the mask manufacturing parameter determination model to obtain the manufacturing parameters output by the mask manufacturing parameter determination model; the manufacturing parameters are used to manufacture a ventilator mask that matches the face of the target object.

[0136] Optionally, after obtaining the facial image of the target object, the server may input the facial image of the target object into a mask manufacturing parameter determination model to obtain manufacturing parameters of a ventilator mask suitable for the target object output by the mask manufacturing parameter determination model, thereby improving the efficiency of obtaining the manufacturing parameters.

[0137] In this embodiment, based on a large number of facial image samples, a model that can directly and accurately output production parameters based on facial images can be trained to improve the efficiency of obtaining production parameters.

[0138] In a detailed embodiment, as shown in FIG9 , a flow chart of a method for determining manufacturing parameters of a ventilator mask is provided, which mainly includes the following steps:

[0139] Step 902, obtaining a facial image of the target object, and performing facial 3D reconstruction based on the facial image to obtain a facial 3D model;

[0140] Step 904, inputting the facial 3D model into the facial region segmentation model to obtain a facial region segmentation result output by the facial 3D model;

[0141] Step 906, determining the mask area and the nose mask area within the mask area in the three-dimensional facial model according to the facial area segmentation result;

[0142] Step 908, based on the facial three-dimensional structure formed by the basic contour of the mouth stop area and the facial feature points in the mask area, contour point detection is performed on the facial feature points in the mask area to determine the target feature points in the mask area that belong to the edge of the mouth stop area;

[0143] Step 910, determining the mouth blocking area based on the target feature points;

[0144] Step 912, based on the three-dimensional coordinates of the facial feature points in the areas covered by the mask area, the nose mask area and the mouth stop area, respectively determine the initial size parameters of the mask area, the nose mask area and the mouth stop area;

[0145] Step 914, determining a size conversion ratio between the three-dimensional facial model and the actual face of the target object;

[0146] Step 916, converting the initial size parameters of the mask area, the nasal mask area, and the mouth stop area according to the size conversion ratio to obtain target size parameters of the mask area, the nasal mask area, and the mouth stop area;

[0147] Step 918, obtaining the preset material thickness of the face mask, nose mask and mouth guard of the ventilator mask;

[0148] Step 920, according to the preset material thickness of the face mask, nose mask and mouth stop in the ventilator mask, the target size parameters of the face mask area, nose mask area and mouth stop area are adjusted to obtain the manufacturing parameters of the ventilator mask.

[0149] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0150] Based on the same inventive concept, the embodiment of the present application also provides a ventilator mask manufacturing parameter determination device for implementing the above-mentioned ventilator mask manufacturing parameter determination method. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above-mentioned method, so the specific limitations in one or more ventilator mask manufacturing parameter determination device embodiments provided below can refer to the limitations of the ventilator mask manufacturing parameter determination method above, and will not be repeated here.

[0151] In an exemplary embodiment, as shown in FIG10 , a device for determining manufacturing parameters of a ventilator mask is provided, comprising: a facial 3D reconstruction module 1002, a facial region segmentation module 1004, a size parameter determination module 1006, and a manufacturing parameter determination module 1008, wherein:

[0152] A facial 3D reconstruction module is used to obtain a facial image of a target object, and perform facial 3D reconstruction based on the facial image to obtain a facial 3D model; the facial 3D model includes the 3D coordinates of multiple facial feature points in the facial image;

[0153] A facial region segmentation module is used to segment the facial region of the three-dimensional facial model, and determine the mask region in the three-dimensional facial model, as well as the nose mask region and the mouth guard region within the mask region;

[0154] A size parameter determination module, for determining target size parameters of the mask area, the nasal mask area and the mouth stop area respectively based on the three-dimensional coordinates of facial feature points in the areas covered by the mask area, the nasal mask area and the mouth stop area;

[0155] The manufacturing parameter determination module is used to determine the manufacturing parameters of the ventilator mask according to the target size parameters of the mask area, the nasal mask area and the mouth stop area.

[0156] The above-mentioned ventilator mask manufacturing parameter determination device first obtains the facial image of the target object, and then performs facial three-dimensional reconstruction based on the facial image to obtain a facial three-dimensional model, wherein the facial three-dimensional model includes the three-dimensional coordinates of multiple facial feature points in the facial image. Further, the facial three-dimensional model is segmented into facial regions, and the mask region in the facial three-dimensional model, as well as the nasal mask region and the mouth stop region in the mask region are determined, so that the target size parameters of the mask region, the nasal mask region and the mouth stop region are determined based on the three-dimensional coordinates of the facial feature points in the regions covered by the mask region, the nasal mask region and the mouth stop region, respectively, so as to determine the manufacturing parameters of the ventilator mask according to the target size parameters of the mask region, the nasal mask region and the mouth stop region. In the whole process, the manufacturing parameters of the ventilator mask can be determined based on the facial three-dimensional model of the target object, so that the ventilator mask finally obtained fits the face of the target object, thereby avoiding air leakage of the ventilator mask during use, thereby improving the use effect of the ventilator.

[0157] In one of the embodiments, the facial region segmentation module is also used to: perform facial region segmentation on the facial three-dimensional model according to the facial three-dimensional structure formed by facial feature points in the facial three-dimensional model to obtain a facial region segmentation result; determine the mask area and the nose mask area within the mask area in the facial three-dimensional model according to the facial region segmentation result; perform mouth block area detection within the mask area based on the facial three-dimensional structure formed by facial feature points in the mask area to determine the mouth block area.

[0158] In one of the embodiments, the facial region segmentation module is further used to: input the facial three-dimensional model into the facial region segmentation model, so that the facial region segmentation model performs facial region segmentation on the facial three-dimensional model according to the facial three-dimensional structure formed by facial feature points in the facial three-dimensional model, and output the facial region segmentation result.

[0159] In one of the embodiments, the ventilator mask manufacturing parameter determination device also includes a facial region segmentation model generation module, which is used to: obtain a data set including multiple preset facial three-dimensional models; each preset facial three-dimensional model has annotated mask area and nasal mask area; based on the data set, train a facial region segmentation model.

[0160] In one of the embodiments, the facial area segmentation module is also used to: obtain the basic contour of the mouth guard area; perform contour point detection on the facial feature points in the mask area based on the basic contour of the mouth guard area and the three-dimensional facial structure composed of the facial feature points in the mask area, and determine the target feature points in the mask area that belong to the edge of the mouth guard area; and determine the mouth guard area based on the target feature points.

[0161] In one of the embodiments, the size parameter determination module is also used to: determine the initial size parameters of the mask area, the nasal mask area and the mouth stop area respectively based on the three-dimensional coordinates of the facial feature points in the areas covered by the mask area, the nasal mask area and the mouth stop area; determine the size conversion ratio between the facial three-dimensional model and the actual face of the target object; and convert the initial size parameters of the mask area, the nasal mask area and the mouth stop area according to the size conversion ratio to obtain the target size parameters of the mask area, the nasal mask area and the mouth stop area.

[0162] In one of the embodiments, the manufacturing parameter determination module is also used to: obtain the preset material thickness of the face mask, nose mask and mouth stop in the ventilator mask; adjust the target size parameters of the face mask area, nose mask area and mouth stop area according to the preset material thickness of the face mask, nose mask and mouth stop in the ventilator mask to obtain the manufacturing parameters of the ventilator mask.

[0163] In one of the embodiments, the ventilator mask manufacturing parameter determination device also includes a mask manufacturing parameter determination model training module, which is used to: train a mask manufacturing parameter determination model based on multiple facial images whose total number exceeds a threshold; input the facial image of the target object into the mask manufacturing parameter determination model to obtain manufacturing parameters output by the mask manufacturing parameter determination model; and the manufacturing parameters are used to manufacture a ventilator mask that matches the face of the target object.

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

[0165] In an exemplary embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be shown in FIG11. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. Among them, the processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store ventilator mask manufacturing parameter determination data. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a method for determining the parameters of a ventilator mask manufacturing is implemented.

[0166] Those skilled in the art will understand that the structure shown in FIG. 11 is merely a block diagram of a partial structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different arrangement of components.

[0167] In an exemplary embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above-mentioned method embodiments when executing the computer program.

[0168] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.

[0169] In one embodiment, a computer program product is provided, including a computer program, which implements the steps in the above method embodiments when executed by a processor.

[0170] It should be noted that the user information (including but not limited to user facial information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0171] A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, an artificial intelligence (AI) processor, etc., but are not limited to this.

[0172] The technical features of the above embodiments can be combined arbitrarily. In order to make the description concise, all possible combinations of the technical features in the above embodiments are not described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of the present application. The above-mentioned embodiments only express several implementation methods of the present application, and their descriptions are relatively specific and detailed, but they cannot be understood as limiting the scope of the patent of the present application. It should be pointed out that for ordinary technicians in this field, without departing from the concept of the present application, several deformations and improvements can be made, which all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be based on the attached claims.

Claims

1. A method for determining manufacturing parameters of a ventilator mask, characterized in that: The method comprises: Acquire a facial image of a target object, and perform three-dimensional facial reconstruction based on the facial image to obtain a three-dimensional facial model; the three-dimensional facial model includes three-dimensional coordinates of a plurality of facial feature points in the facial image; Performing facial region segmentation on the three-dimensional facial model to determine a mask region in the three-dimensional facial model, and a nose mask region and a mouth guard region within the mask region; Determine the target size parameters of the mask area, the nasal mask area and the mouth stop area respectively based on the three-dimensional coordinates of the facial feature points in the areas covered by the mask area, the nasal mask area and the mouth stop area; The manufacturing parameters of the ventilator mask are determined according to the respective target size parameters of the mask area, the nasal mask area and the mouth stop area.

2. The method according to claim 1, characterized in that: The step of segmenting the facial three-dimensional model to determine a mask area in the facial three-dimensional model, and a nose mask area and a mouth guard area in the mask area includes: According to the facial three-dimensional structure formed by the facial feature points in the facial three-dimensional model, performing facial region segmentation on the facial three-dimensional model to obtain a facial region segmentation result; Determine, in the three-dimensional facial model, a mask area and a nose mask area within the mask area according to the facial area segmentation result; Based on the three-dimensional facial structure formed by the facial feature points in the mask area, a mouth stop area detection is performed in the mask area to determine the mouth stop area.

3. The method according to claim 2, characterized in that The step of performing facial region segmentation on the three-dimensional facial model according to the three-dimensional facial structure formed by facial feature points in the three-dimensional facial model to obtain a facial region segmentation result includes: Inputting the facial 3D model into a facial region segmentation model, so that the facial region segmentation model performs facial region segmentation on the facial 3D model according to the facial 3D structure formed by facial feature points in the facial 3D model, and outputs a facial region segmentation result; The facial region segmentation model is generated in the following way: Acquire a data set including a plurality of preset three-dimensional facial models; each of the preset three-dimensional facial models has a mask area and a nasal mask area marked; Based on the data set, the facial region segmentation model is trained.

4. The method according to claim 2, characterized in that: The method of performing mouth stop area detection in the mask area based on the facial three-dimensional structure formed by facial feature points in the mask area to determine the mouth stop area includes: Get the basic outline of the mouth block area; According to the three-dimensional facial structure formed by the basic contour of the mouth stop area and the facial feature points in the mask area, contour point detection is performed on the facial feature points in the mask area to determine the target feature points in the mask area that belong to the edge of the mouth stop area; Based on the target feature points, a mouth blocking area is determined.

5. The method according to claim 1, characterized in that The method of determining target size parameters of the mask area, the nose mask area, and the mouth stop area based on the three-dimensional coordinates of facial feature points in the areas covered by the mask area, the nose mask area, and the mouth stop area respectively comprises: Based on the three-dimensional coordinates of the facial feature points in the areas covered by the mask area, the nose mask area and the mouth stop area, respectively, determining the initial size parameters of the mask area, the nose mask area and the mouth stop area; Determining a size conversion ratio between the three-dimensional facial model and the actual face of the target object; According to the size conversion ratio, the initial size parameters of the mask area, the nasal mask area and the mouth stop area are converted to obtain the target size parameters of the mask area, the nasal mask area and the mouth stop area.

6. The method according to claim 1, characterized in that Determining the manufacturing parameters of the ventilator mask according to the target size parameters of the mask area, the nasal mask area and the mouth stop area respectively includes: Get the preset material thickness of the face mask, nose mask and mouth guard of the ventilator mask; According to the preset material thicknesses of the face mask, nose mask and mouth stop in the ventilator mask, the target size parameters of the face mask area, the nose mask area and the mouth stop area are adjusted to obtain the manufacturing parameters of the ventilator mask.

7. The method according to claim 1, characterized in that The method further comprises: Based on a plurality of facial images whose total number exceeds a threshold, a mask making parameter determination model is trained; The facial image of the target object is input into the mask manufacturing parameter determination model to obtain the manufacturing parameters output by the mask manufacturing parameter determination model; the manufacturing parameters are used to manufacture a ventilator mask that matches the face of the target object.

8. A device for determining manufacturing parameters of a ventilator mask, characterized in that: The device comprises: A facial 3D reconstruction module is used to obtain a facial image of a target object, and perform facial 3D reconstruction based on the facial image to obtain a facial 3D model; the facial 3D model includes 3D coordinates of a plurality of facial feature points in the facial image; A facial region segmentation module is used to perform facial region segmentation on the three-dimensional facial model to determine a mask region in the three-dimensional facial model, and a nose mask region and a mouth guard region within the mask region; A size parameter determination module, for determining target size parameters of the mask area, the nasal mask area and the mouth stop area respectively based on the three-dimensional coordinates of facial feature points in the areas covered by the mask area, the nasal mask area and the mouth stop area respectively; The manufacturing parameter determination module is used to determine the manufacturing parameters of the ventilator mask according to the target size parameters of the mask area, the nasal mask area and the mouth stop area.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images