An airway assessment method based on depth camera image recognition

Abstract

The embodiment of this specification provides an airway assessment method based on depth camera image recognition, including: collecting a depth image of a person to be identified, including a front image, a side image, and a front maximum mouth opening image; obtaining the thyromental distance, the degree of small mandible, the degree of dental abnormality, the degree of upper lip occlusion, and the neck mobility according to the front image and the side image; obtaining the mouth opening and Mahalanobis grade according to the front maximum mouth opening image; grading and scoring the thyromental distance, the degree of small mandible, the degree of dental abnormality, the degree of upper lip occlusion, the neck mobility, the mouth opening, and Mahalanobis grade of the person to be identified respectively to obtain multiple scores; performing weighted calculation on the multiple scores to obtain an evaluation value; comparing the evaluation value with the set value to obtain the airway condition of the person to be identified. The embodiment of this specification can improve the efficiency and accuracy of airway assessment.

Description

Technical Field

[0001] The embodiments of this specification relate to the field of image recognition, and in particular, to an airway assessment method based on depth camera image recognition. Background Art

[0002] The incidence of difficult endotracheal intubation during general anesthesia is about 6%. Difficult endotracheal intubation is life-threatening. An analysis pointed out that difficult endotracheal intubation accounts for 17% of abnormal respiratory events, of which 85% caused brain damage and death. The higher the difficulty, the greater the risk of brain damage or death. Therefore, it is very important to do a good preoperative airway assessment before anesthesia.

[0003] Preoperative airway assessment can identify difficult airway early and reduce the incidence of unexpected difficult airway. It is also a prerequisite for correctly handling difficult airway and making adequate preparations. Some risk factors for difficult airway can be obtained by observing the patient's appearance, such as congenital craniofacial malformations, oral and maxillofacial deformities or defects caused by trauma, infection, and tumors, microstomia and chin-chest adhesions caused by scar adhesions after burns, abnormal anatomical structures near the airway after surgery or radiotherapy, ankylosis of the temporomandibular joint, obesity, short neck, small mandible, high larynx, macroglossia, etc. However, determining whether a patient has a difficult airway only by observation is inefficient and inaccurate.

[0004] Therefore, it is necessary to propose an airway assessment method based on depth camera image recognition to improve the efficiency and accuracy of airway assessment. Summary of the invention

[0005] The purpose of the embodiments of this specification is to provide an airway assessment method based on depth camera image recognition to improve the efficiency and accuracy of airway assessment.

[0006] To achieve the above objectives, on the one hand, the embodiments of this specification provide an airway assessment method based on depth camera image recognition, including:

[0007] Collecting depth images of the person to be identified, including front image, side image and front image with maximum mouth opening;

[0008] According to the frontal image and the side image, the thyromental distance, the degree of small jaw, the degree of dental abnormality, the degree of upper lip occlusion and the range of motion of the neck are obtained;

[0009] According to the frontal maximum mouth opening image, the mouth opening degree and Mahalanobis grade were obtained;

[0010] The thyromental distance, small mandible, dental abnormality, upper lip occlusion, neck mobility, mouth opening and Mahalanobis classification of the person to be identified are graded and scored respectively to obtain multiple scores;

[0011] Performing weighted calculation on the multiple scores to obtain an evaluation value;

[0012] The evaluation value is compared with the set value to obtain the airway condition of the person to be identified.

[0013] Preferably, obtaining the thyromental distance, the degree of small mandible, the degree of dental abnormality, the degree of upper lip occlusion and the range of motion of the neck based on the front image and the side image further comprises:

[0014] Obtaining a 3D model using 3D reconstruction technology according to the front image and the side image;

[0015] By detecting the key points of the maxilla and mandible in the 3D model, the thyromental distance is obtained;

[0016] By analyzing the length and angle of the mandible in the 3D model and combining it with standard facial proportions, the degree of small mandible is obtained;

[0017] Analyzing the 3D model using a trained deep learning model to obtain the degree of tooth abnormality;

[0018] The degree of upper lip occlusion is obtained according to the key points of the lower incisor and the upper lip in the side image corresponding to the lower incisor occluding the upper lip;

[0019] The neck mobility is obtained according to the key points of the neck in the side images at multiple different neck angles.

[0020] Preferably, obtaining the mouth opening degree and Mahalanobis grade according to the frontal maximum mouth opening image further comprises:

[0021] Using two cameras at fixed positions, respectively collect the maximum mouth opening images of the person to be identified from the front, and obtain the maximum mouth opening images corresponding to the two cameras;

[0022] The maximum mouth opening images corresponding to the two cameras are processed respectively to obtain composite images corresponding to the two cameras;

[0023] Determine the disparity between the same facial feature points in two composite images;

[0024] According to the parallax and the parameters of the camera, combined with the principle of similar triangles, the actual distance between the person to be identified and the camera is calculated;

[0025] The actual distance between the two selected points is calculated based on the actual distance, the focal length of the camera and the image distance between the two selected points on the two composite images, in combination with the principle of similar triangles;

[0026] The mouth opening degree and the Mahalanobis grade are obtained according to the actual distance between the two selected points.

[0027] Preferably, the processing of the maximum mouth opening images corresponding to the two cameras to obtain the composite images corresponding to the two cameras further includes:

[0028] For each camera, obtain the grayscale image and depth image corresponding to the maximum mouth opening image;

[0029] Adjusting the resolution of the grayscale image and the depth image to be the same resolution;

[0030] After adjusting the bit depth of the grayscale image to match the bit depth of the depth image, the grayscale image is stacked with the depth image to obtain a composite image for each camera.

[0031] Preferably, the two selected points are two points corresponding to the upper teeth and the lower teeth, or two points corresponding to the upper lip and the lower lip;

[0032] The obtaining of the mouth opening degree and the Mahalanobis classification according to the actual distance between the two selected points further comprises:

[0033] Obtaining the mouth opening according to the actual distance between the two selected points;

[0034] The Mahalanobis classification is obtained according to the positions and visibility of the oral cavity feature points in the maximum mouth opening image.

[0035] Preferably, it also includes:

[0036] The face position is fixed by the chin-pull limiter.

[0037] Using a sensor to measure the distance between the person to be identified and the camera;

[0038] The actual distance between the person to be identified and the camera is compensated for an error according to the measured distance.

[0039] Preferably, the thyromental distance, small mandible, dental abnormality, upper lip occlusion, neck mobility, mouth opening and Mahalanobis classification of the person to be identified are graded and scored respectively to obtain multiple scores, further comprising:

[0040] Determine the correlation between any two of the factors of the thyromental distance, micrognathia, dental abnormality, upper lip occlusion, neck mobility, mouth opening and Mahalanobis classification;

[0041] According to the basic grading rules of each factor, the basic scores corresponding to the thyromental distance, small mandible, dental abnormality, upper lip occlusion, neck mobility, mouth opening and Mahalanobis classification of the person to be identified are obtained respectively;

[0042] Based on the association relationship, a correction rule for each factor is obtained;

[0043] According to the correction rules of the various factors, each basic score is corrected respectively to obtain multiple scores.

[0044] Preferably, the determining of the correlation between any two of the factors of the thyromental distance, the degree of small mandible, the degree of dental abnormality, the degree of upper lip occlusion, the range of motion of the neck, the degree of mouth opening and the Mahalanobis classification further comprises:

[0045] Obtain the thyromental distance, small mandible, dental abnormality, upper lip occlusion, neck mobility, mouth opening and Mahalanobis classification corresponding to multiple known patients, and obtain multiple parameters corresponding to each factor;

[0046] Analyze the correlation between each two factors one by one and obtain the correlation coefficient between each two factors;

[0047] Two factors whose correlation coefficients are within a set coefficient range are determined to have an associated relationship, where the associated relationship includes a positive correlation and a negative correlation.

[0048] Preferably, the basic scores corresponding to the thyromental distance, small mandible, dental abnormality, upper lip occlusion, neck mobility, mouth opening and Mahalanobis classification of the person to be identified are obtained according to the basic classification rules of each factor, and further include:

[0049] According to the multiple parameters corresponding to each factor, the parameter range corresponding to each factor is obtained;

[0050] The parameter range is equally divided to obtain a plurality of intervals corresponding to each factor, wherein the number of intervals corresponding to each factor is equal;

[0051] Arrange the multiple intervals corresponding to each factor according to the risk of airway obstruction from small to large, and obtain multiple intervals after arrangement;

[0052] Assign values ​​to the multiple intervals after arrangement from small to large in the order of arrangement to obtain the basic score corresponding to each factor.

[0053] Preferably, the correction rules for obtaining each factor based on the association relationship further include:

[0054] When the correlation is positively correlated, the two factors with the correlation are determined as correlation factors;

[0055] All factors with correlation factors are determined as positive correlation factors;

[0056] When the difference between the basic score of at least one factor among the positively associated factors of the person to be identified and the basic scores of other factors is greater than a first set difference, the basic score of the at least one factor is corrected or the basic scores of other factors are corrected according to the ratio between the number of the at least one factor and the number of other factors;

[0057] When the correlation is negatively correlated, the two factors with the correlation are determined as correlation factors;

[0058] All factors with correlation factors were determined as negative correlation factors;

[0059] When the difference between the basic score of at least one factor among the negatively associated factors of the person to be identified and the basic scores of other factors is less than the second set difference, the basic score of the at least one factor is corrected or the basic scores of other factors are corrected according to the ratio between the number of the at least one factor and the number of other factors.

[0060] The process of assessing airway status through image acquisition and analysis includes the acquisition of frontal images, lateral images, and frontal images of maximum mouth opening. Through these images, key airway parameters such as thyromental distance, small mandible, dental abnormality, neck mobility, mouth opening, and Mahalanobis grade can be extracted. These parameters are graded and scored to reflect the potential problems and risks of the airway. Subsequently, each parameter is weighted to obtain a comprehensive assessment value. By comparing the assessment value with the set value, the severity of the airway can be determined. This process can accurately and standardizedly assess the airway status, help detect problems early, and provide reliable data support for clinical decision-making. In addition, as a non-invasive detection method, it is simple and easy to use, and can be widely used in clinical scenarios such as pre-anesthesia assessment and sleep apnea syndrome screening to improve the accuracy and objectivity of the assessment.

[0061] In order to make the above and other purposes, features and advantages of the present specification more obvious and easy to understand, the following specifically cites preferred embodiments and describes them in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] In order to more clearly illustrate the embodiments of this specification or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this specification. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0063] Figure 1 A schematic diagram of a flow chart of an airway assessment method based on depth camera image recognition provided in an embodiment of this specification is shown;

[0064] Figure 2 A schematic diagram of a process for processing a front image and a side image provided in an embodiment of this specification is shown;

[0065] Figure 3 A schematic diagram of a process for processing a frontal maximum mouth opening image provided in an embodiment of this specification is shown;

[0066] Figure 4 A schematic diagram of a process of synthesizing an image provided by an embodiment of this specification is shown;

[0067] Figure 5 A schematic diagram of a method for mouth opening and Marsh grading provided in an embodiment of this specification is shown;

[0068] Figure 6 Another schematic diagram of the process of an airway assessment method based on depth camera image recognition provided by an embodiment of this specification is shown;

[0069] Figure 7 A schematic diagram of a scoring method provided in an embodiment of this specification is shown;

[0070] Figure 8 A schematic diagram of a method for determining an association relationship provided in an embodiment of this specification is shown;

[0071] Fig. 9 A schematic diagram of a method for determining a basic score provided in an embodiment of this specification is shown;

[0072] Fig.10 A schematic diagram showing a portion of the flow chart of the correction rules provided in the embodiments of this specification;

[0073] Fig.11 Another partial flow chart of the correction rules provided in the embodiments of this specification is shown;

[0074] Fig.12 A schematic diagram of the module structure of an airway assessment device based on depth camera image recognition provided in an embodiment of this specification is shown;

[0075] Fig.13 A schematic diagram of the structure of a computer device provided in an embodiment of this specification is shown.

[0076] Explanation of the symbols in the accompanying drawings: 100, acquisition module; 200, first determination module; 300, second determination module; 400, scoring module; 500, weighting module; 600, comparison module. DETAILED DESCRIPTION

[0077] The following will be combined with the drawings in the embodiments of this article to clearly and completely describe the technical solutions in the embodiments of this article. Obviously, the described embodiments are only part of the embodiments of this article, not all of the embodiments. Based on the embodiments of this article, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this article.

[0078] It should be noted that the terms "first", "second", etc. in the specification and claims of this document and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of this document described here can be implemented in an order other than those illustrated or described here. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, device, product or equipment that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or equipment. It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data 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 acquisition, storage, use, processing, etc. of data in the technical solutions described in the embodiments of this application are in compliance with the provisions of relevant laws and regulations.

[0079] like Figure 1 , an airway assessment method based on depth camera image recognition, comprising:

[0080] S101: collecting a depth image of the person to be identified, including a front image, a side image, and a front image with the largest mouth opening;

[0081] S102: obtaining the thyromental distance, the degree of small mandible, the degree of dental abnormality, the degree of upper lip occlusion and the range of motion of the neck according to the front image and the side image;

[0082] S103: obtaining the mouth opening degree and Mahalanobis grade according to the frontal maximum mouth opening image;

[0083] S104: grading and scoring the thyromental distance, small mandible, dental abnormality, upper lip occlusion, neck mobility, mouth opening and Mahalanobis classification of the person to be identified, respectively, to obtain multiple scores;

[0084] S105: performing weighted calculation on the multiple scores to obtain an evaluation value;

[0085] S106: Compare the evaluation value with the set value to obtain the airway condition of the person to be identified.

[0086] During the acquisition, a depth camera may be used to acquire depth images, wherein the front image is an image of the front face, the side image is an image of the side face, and the front maximum mouth opening image is an image of the front face with the mouth opened to the maximum.

[0087] like Figure 2 , wherein obtaining the thyromental distance, the degree of small mandible, the degree of dental abnormality, the degree of upper lip occlusion and the range of motion of the neck according to the front image and the side image further comprises:

[0088] S201: Obtaining a 3D model using 3D reconstruction technology according to the front image and the side image;

[0089] S202: Obtaining the thyromental distance by detecting key points of the maxilla and mandible in the 3D model;

[0090] S203: deriving the degree of small mandible by analyzing the length and angle of the mandible in the 3D model and combining with standard facial proportions;

[0091] S204: Analyze the 3D model using the trained deep learning model to obtain the degree of tooth abnormality;

[0092] S205: Obtaining the upper lip occlusion degree according to the lower incisor key points and the upper lip key points in the side image corresponding to the lower incisor occluding the upper lip;

[0093] S206: Obtain neck mobility according to key points of the neck in the side images at multiple different neck angles.

[0094] Specifically, the input data is the collected frontal image and side image, and there can be multiple frontal images and side images. Structured light scanning, stereo vision, or multi-view stereo (MVS) algorithm based on deep learning are used to generate point cloud data. Point cloud data is composed of a large number of points, each point represents a specific position on the surface of the object, and these points together describe the three-dimensional shape of the object. Then, a 3D reconstruction algorithm (such as Poisson Surface Reconstruction) is used to process the point cloud data to generate a 3D model of the face, including the structure of the upper and lower jaws.

[0095] Use a key point detection algorithm (such as SIFT, SURF, or a key point detection network trained by deep learning) to identify the key points of the maxilla and mandible in the 3D model. Calculate the Euclidean distance between these key points to obtain the thyromental distance, which is usually expressed as a numerical value reflecting the relative distance between the maxilla and mandible.

[0096] The length of the mandible and the angle between the mandible and the maxilla are calculated, and the standard mandible length and angle are defined in combination with standard facial proportions (such as the golden ratio or medical standards). The golden ratio (1:1.618) is a commonly used ratio in aesthetics. For example, the ratio of facial height to width, maxillary to mandibular length, etc. can be compared with the golden ratio. Medical standards include common facial proportions, such as the ratio of mandibular length to total facial length.

[0097] To further evaluate the degree of mandibular development, on the one hand, the mandibular length is evaluated by comparing the mandibular length in the 3D model with the standard mandibular length, and calculating the difference percentage. For example, if the standard mandibular length is 80mm, and the mandibular length in the 3D model is 75mm, the difference percentage is: (80-75) / 80=6.25%. On the other hand, the mandibular angle is evaluated by comparing the mandibular angle in the 3D model with the standard mandibular angle, and calculating the deviation. For example, if the standard mandibular angle is 120 degrees, and the mandibular angle in the 3D model is 115 degrees, the deviation is: 120-115=5 degrees.

[0098] Finally, the degree of micrognathia is quantified by comparing with the standard mandibular length and angle. The length difference quantification standard is as follows: no micrognathia: length difference within a%, mild micrognathia: length difference between a% and b%, moderate micrognathia: length difference between b% and c%, severe micrognathia: length difference exceeds c%, where a<b<c. ​​The angle deviation quantification standard is as follows: no micrognathia: angle deviation within d degrees, mild micrognathia: angle deviation between d degrees and e degrees, moderate micrognathia: angle deviation between e degrees and f degrees, severe micrognathia: angle deviation exceeds f degrees, where d<e<f.

[0099] Comprehensive evaluation is performed by combining the length difference percentage and the angle deviation. If the degree of small mandible obtained by the above two quantitative standards is the same, the degree of small mandible is the corresponding degree of small mandible. If the degree of small mandible obtained by the above two quantitative standards is different, the degree of small mandible is the more severe one.

[0100] The 3D model is analyzed using a trained deep learning model (such as a 3D convolutional neural network or a PointNet-based model). The deep learning model may identify morphological abnormalities of the teeth, such as tooth misalignment, defects, deformities, etc., through feature extraction and classification layers. The output is the degree of tooth abnormality, which may be a numerical score or classification (such as no abnormality, mild abnormality, moderate abnormality, severe abnormality).

[0101] For the side image corresponding to the lower incisor biting the upper lip, key point detection is performed to extract the key points of the lower incisor and the upper lip. The key points of the upper lip can be selected at the lowest edge of the upper lip (the part close to the lower incisor), while the key points of the lower incisor can be selected at the incisal edge of the lower incisor (the front end of the tooth). The positional relationship between the key points of the lower incisor and the key points of the upper lip is determined to obtain the degree of upper lip biting.

[0102] If the key point of the lower incisors is above or on the same horizontal line as the key point of the upper lip, it means that the lower incisors have moved up to the same position as or higher than the upper lip, which means that the lower incisors partially or completely cover the upper lip, which corresponds to the first level of upper lip occlusion. If the key point of the lower incisors partially overlaps with the key point of the upper lip, or the key point of the lower incisors only covers a part of the key point of the upper lip, it is judged as partial coverage, which corresponds to the second level of upper lip occlusion. If the key point of the lower incisors is still below the key point of the upper lip, it means that the lower incisors fail to cover the upper lip, which corresponds to the third level of upper lip occlusion. The above only provides a method for determining the degree of upper lip occlusion. The degree of upper lip occlusion can include levels one to three, levels one to four, levels one to five, etc., and this manual does not limit this.

[0103] Key point detection is performed on multiple side images. Multiple side images of the person to be identified are collected during the neck movement. The neck angles of the multiple side images are different. Key points related to the neck are extracted, including: mandibular point, front end point of the neck (such as the spinous process of the C7 vertebra) and shoulder point.

[0104] According to the relative motion trajectory of the key points of the neck at different angles, the range of neck movement (such as the angles of flexion, backward movement, and left and right rotation) is obtained, and the maximum angle of each activity (such as the maximum angle of flexion, the maximum angle of backward movement, and the maximum angle of left and right rotation) is calculated.

[0105] The neck mobility of the person to be identified is evaluated, wherein each activity corresponds to a weight coefficient, and the maximum angle of each activity of the person to be identified and the corresponding weight coefficient are weighted and summed to obtain the corresponding neck mobility of the person to be identified.

[0106] like Figure 3 , wherein obtaining the mouth opening degree and Mahalanobis grading according to the frontal maximum mouth opening image further comprises:

[0107] S301: using two cameras at fixed positions to respectively capture the maximum mouth opening images of the person to be identified from the front, and obtain the maximum mouth opening images corresponding to the two cameras respectively;

[0108] S302: Processing the maximum mouth opening images respectively corresponding to the two cameras to obtain composite images respectively corresponding to the two cameras;

[0109] S303: determining the disparity between the same facial feature points in the two composite images;

[0110] S304: Calculate the actual distance between the person to be identified and the camera according to the parallax and camera parameters and the principle of similar triangles;

[0111] S305: calculating the actual distance between the two selected points according to the actual distance, the focal length of the camera and the image distance between the two selected points on the two synthetic images in combination with the principle of similar triangles;

[0112] S306: Obtaining the mouth opening degree and the Mahalanobis grade according to the actual distance between the two selected points.

[0113] When setting up two fixed cameras, the following aspects need to be noted: Camera spacing (baseline distance): The horizontal or vertical distance between the cameras should be large enough to obtain a good parallax effect, which usually depends on the depth resolution requirements of the application and the distance to the person to be identified. Camera alignment: The cameras should ensure that they are aimed at the same area of ​​the person to be identified. The lens should point to the center of the target, and the viewing angle should be as consistent as possible to reduce errors caused by angle differences. Camera installation height: The camera installation height should be aligned with the face of the person to be identified to obtain the best viewing angle. It usually needs to be adjusted to the eye level or slightly higher than the person to be identified. Camera angle: The angle of the camera should remain stable and relatively fixed to avoid the deviation of the shooting angle from affecting the parallax calculation. The camera should maintain a consistent angle in the horizontal or vertical direction. Camera calibration: Ensure that the two cameras are calibrated, correct the internal parameters (focal length, optical center, etc.) and external parameters (camera position, direction, etc.) to ensure the accuracy of the image data. Lighting conditions: Ensure uniform lighting conditions, avoid the impact of strong light or shadows on image quality, use appropriate light sources or control ambient lighting to obtain clear images.

[0114] like Figure 4 , the processing of the maximum mouth opening images corresponding to the two cameras to obtain the composite images corresponding to the two cameras further includes:

[0115] S401: For each camera, obtain a grayscale image and a depth image corresponding to the maximum mouth opening image;

[0116] S402: Adjusting the resolution of the grayscale image and the depth image to be the same resolution;

[0117] S403: After adjusting the bit depth of the grayscale image to match the bit depth of the depth image, the grayscale image and the depth image are stacked to obtain a composite image of each camera.

[0118] Furthermore, assuming that P is a facial feature point, such as the corner of the mouth, the tip of the nose, the corner of the eye, etc., the optical centers of the two cameras are OR and OT respectively, point P is imaged as P′ and P′′ on the photoreceptors of the two cameras, the focal length of the camera is f, and the distance between the optical centers of the two cameras is the baseline length B. By measuring the distance between points P′ and P′′ on the imaging plane, that is, the parallax dis, the actual distance between the person to be identified and the camera is solved.

[0119] The optical center OR, OT and object P can be regarded as a triangle. The triangle OR-P′-P of one camera OR and the triangle OT-P′′-P of the other camera OT are similar. By the principle of similar triangles, we can use the following formula to calculate the distance Z from point P to the camera:

[0120] ;

[0121] Z is the actual distance between the person to be identified and the camera. The principle of similar triangles is further used to solve the actual distance between the two corresponding selected points on the two composite images.

[0122] Relationship between image distance QQ′ and actual distance: QQ′ is the distance between two points on the image plane. Assuming that Q and Q′ are the imaging points on the image corresponding to two points P1 and P2 in the actual space, using the principle of similar triangles, the distance d between P1 and P2 in the actual space can be calculated by the following formula:

[0123] .

[0124] Wherein, the selected two points are two points corresponding to the upper teeth and the lower teeth respectively, or two points corresponding to the upper lip and the lower lip respectively;

[0125] like Figure 5 , the obtaining of the mouth opening degree and the Mahalanobis classification according to the actual distance between the two selected points further comprises:

[0126] S501: Obtaining the mouth opening according to the actual distance between the two selected points;

[0127] S502: Obtaining Mahalanobis classification according to the positions and visibility of oral feature points in the maximum mouth opening image.

[0128] Specifically, in the maximum mouth opening image, identify and mark the positions of oral feature points, including soft palate, uvula, tonsillar pillars, and hard palate. Use image processing algorithms to determine the positions and visibility of these feature points in the image. Common methods include: edge detection, which is used to segment the outline of the oral structure, feature point detection, which uses feature point detection algorithms (such as SIFT and SURF) to identify and mark these feature points, and image registration, which aligns the feature points in the image with the predefined anatomical structure positions. Finally, the Mahalanobis classification is obtained. Level 1: The soft palate, uvula, and tonsillar pillars are clearly visible. Level 2: The soft palate and uvula are visible, but the tonsillar pillars are partially blocked. Level 3: Only the base of the soft palate and uvula are visible, and the tonsillar pillars are not visible. Level 4: Only the hard palate is visible, and other structures are not visible.

[0129] like Figure 6 , the embodiments of this specification also include:

[0130] S601: Fixing the face position by means of a chin-pushing limit device;

[0131] S602: using a sensor to measure the distance between the person to be identified and the camera;

[0132] S603: Compensating an error of an actual distance between the person to be identified and the camera according to the measured distance.

[0133] Assume that the measured distance obtained by the chin drag limiter and the sensor is z , error compensation can be performed:

[0134] ΔZ = Z − z;

[0135] Adjust the actual distance Z between the person to be identified and the camera by measuring the distance z:

[0136] Zcorrected = Z − ΔZ;

[0137] The adjusted distance Z Substitute corrected into the calculation formula to calculate the distance d between P1 and P2 in real space:

[0138] .

[0139] like Figure 7 , wherein the thyromental distance, small mandible, dental abnormality, upper lip occlusion, neck mobility, mouth opening and Mahalanobis classification of the person to be identified are graded and scored respectively, and the multiple scores obtained further include:

[0140] S701: Determine the correlation between any two of the factors of the thyromental distance, the degree of small mandible, the degree of dental abnormality, the degree of upper lip occlusion, the range of motion of neck, the degree of mouth opening and the Mahalanobis classification;

[0141] S702: according to the basic classification rules of each factor, the basic scores corresponding to the thyromental distance, the degree of small mandible, the degree of dental abnormality, the degree of upper lip occlusion, the neck mobility, the mouth opening and the Mahalanobis classification of the person to be identified are obtained respectively;

[0142] S703: Based on the association relationship, obtain correction rules for each factor;

[0143] S704: According to the correction rules of the various factors, the basic scores are corrected respectively to obtain multiple scores.

[0144] Specifically, Figure 8 The determining of the correlation between any two of the factors of the thyromental distance, the degree of small mandible, the degree of dental abnormality, the degree of upper lip occlusion, the range of motion of the neck, the degree of mouth opening and the Mahalanobis classification further comprises:

[0145] S801: Obtain the thyromental distance, small mandible degree, dental abnormality degree, upper lip occlusion degree, neck mobility, mouth opening degree and Mahalanobis classification corresponding to multiple known patients, and obtain multiple parameters corresponding to each factor;

[0146] S802: Analyze the correlation between every two factors one by one to obtain the correlation coefficient between every two factors;

[0147] S803: Determine that two factors whose correlation coefficients are within a set coefficient range have an associated relationship, where the associated relationship includes a positive correlation and a negative correlation.

[0148] The factors include: thyromental distance, small mandible degree, dental abnormality, upper lip occlusion degree, neck mobility, mouth opening degree and Mahalanobis grade. The known patients are patients with known airway conditions. After collecting the thyromental distance, small mandible degree, dental abnormality, upper lip occlusion degree, neck mobility, mouth opening degree and Mahalanobis grade of each known patient, multiple parameters corresponding to each factor can be obtained. For example, the thyromental distance of all known patients is the multiple parameters corresponding to the thyromental distance.

[0149] Correlation analysis usually uses the Pearson correlation coefficient, Spearman correlation coefficient, etc. Based on the multiple parameters corresponding to each factor, the correlation coefficient between every two factors is calculated. Two factors with correlation coefficients within the set coefficient range are determined to have a correlation relationship. A positive correlation indicates that the two factors have the same changing trends, and a negative correlation indicates that the two factors have opposite changing trends.

[0150] Further, such as Fig. 9The basic scores corresponding to the thyromental distance, small mandible, dental abnormality, upper lip occlusion, neck mobility, mouth opening and Mahalanobis classification of the person to be identified are obtained according to the basic classification rules of each factor, and further include:

[0151] S901: Obtaining a parameter range corresponding to each factor according to multiple parameters corresponding to each factor;

[0152] S902: equally divide the parameter range to obtain multiple intervals corresponding to each factor, wherein the number of intervals corresponding to each factor is equal;

[0153] S903: Arrange the multiple intervals corresponding to each factor according to the risk of airway obstruction from small to large, and obtain multiple intervals after arrangement;

[0154] S904: assign values ​​to the arranged multiple intervals in ascending order according to the arrangement order to obtain a basic score corresponding to each factor.

[0155] According to the lower threshold parameter and upper threshold parameter of the multiple parameters corresponding to each factor, the parameter range corresponding to each factor can be obtained. Taking the thyromental distance as an example, assuming that the upper threshold parameter is 8 cm and the lower threshold parameter is 4 cm, the parameter range is 4-8 cm. Taking the degree of micrognathia as an example, assuming that the upper threshold parameter is severe micrognathia and the lower threshold parameter is no micrognathia, the parameter range is no micrognathia to severe micrognathia.

[0156] The parameter range is divided into multiple intervals, for example, 4 intervals, 4-5cm, 5-6cm, 6-7cm, 7-8cm, no micrognathia, mild micrognathia, moderate micrognathia, severe micrognathia.

[0157] Generally speaking, the shorter the thyromental distance, the higher the risk of airway obstruction. The higher the degree of micrognathia, the higher the risk of airway obstruction. The higher the degree of dental abnormality, the higher the risk of airway obstruction. The higher the degree of upper lip bite, the higher the risk of airway obstruction. The lower the neck mobility, the higher the risk of airway obstruction. The smaller the mouth opening, the higher the risk of airway obstruction. The higher the Mahalanobis grade (grade IV), the higher the risk of airway obstruction.

[0158] Specifically, the shorter the thyromental distance, the higher the risk of airway obstruction. The multiple intervals after arrangement are arranged from small to large according to the risk of airway obstruction, and the multiple intervals after arrangement should be: 7-8cm, 6-7cm, 5-6cm, 4-5cm. The multiple intervals after arrangement are assigned values ​​from small to large according to the arrangement order to obtain the basic score corresponding to each factor. For example, the basic scores from small to large are 1, 2, 3, and 4 points, 7-8cm corresponds to 1 point, 6-7cm corresponds to 2 points, and so on.

[0159] It should be noted that the number of intervals corresponding to each factor is equal, for example, 4 intervals, and when the multiple intervals after arrangement are assigned values ​​in ascending order, the assignment method and specific assignment of each factor are also the same, for example, the basic score assignment is 1, 2, 3, 4 points in ascending order. In this way, assuming that the thyroid-chin distance of the person to be identified is 5.5 cm, the corresponding score of the person to be identified is 3 points.

[0160] Further, such as Fig.10 and Fig.11 , the correction rules for each factor obtained based on the association relationship further include:

[0161] S1001: When the correlation relationship is positively correlated, two factors with the correlation relationship are determined as correlation factors;

[0162] S1002: All factors with associated factors are determined as positive associated factors;

[0163] S1003: when the difference between the basic score of at least one factor among the positively associated factors of the person to be identified and the basic scores of other factors is greater than a first set difference, the basic score of the at least one factor is modified, or the basic scores of other factors are modified, according to the ratio between the number of the at least one factor and the number of other factors;

[0164] S1004: When the correlation relationship is negatively correlated, the two factors with the correlation relationship are determined as correlation factors;

[0165] S1005: All factors with correlation factors are determined as negative correlation factors;

[0166] S1006: When the difference between the basic score of at least one factor among the negatively associated factors of the person to be identified and the basic scores of other factors is less than a second set difference, the basic score of the at least one factor is corrected, or the basic scores of other factors are corrected, according to the ratio between the number of the at least one factor and the number of other factors.

[0167] Taking the positive correlation as an example, assuming that among the above 7 factors, thyromental distance, degree of micrognathia, degree of dental abnormality, degree of upper lip occlusion, neck mobility, mouth opening and Mahalanobis classification, the thyromental distance and degree of micrognathia are positively correlated, and the degree of micrognathia and degree of dental abnormality are positively correlated, then the thyromental distance and degree of micrognathia are associated factors, the degree of micrognathia and degree of dental abnormality are associated factors, and all factors with associated factors are thyromental distance, degree of micrognathia and degree of dental abnormality, and all three are positive associated factors.

[0168] Assume that the basic score of the onychomental distance of the person to be identified is 4 points, the basic score of the small mandible is 1 point, and the score of the degree of dental abnormality is 2 points. The first set gap can be set according to the actual situation, and is generally positively correlated with the number of intervals obtained by dividing the parameter range. The more intervals there are, the finer the division, and the larger the first set gap. Assuming that the first set gap is 1 point, the difference between the basic score of the onychomental distance and the degree of small mandible, and the basic score of the degree of dental abnormality is greater than 1 point, and correction is required. At least one of the factors is the onychomental distance, and the other factors are the degree of small mandible and the degree of dental abnormality. The ratio between the number of at least one factor and the number of other factors is 1:2.

[0169] When the ratio is not 1, the basic score of the smaller number is corrected based on the larger number. For example, the thyromental distance is corrected based on the degree of small mandible and tooth abnormality. When correcting, the average value of the larger number can be calculated and used as the corrected score of the smaller number. Of course, other correction schemes can also be adopted.

[0170] When the ratio is 1, for example, the basic score of the onychomental distance is 3 points, the basic score of the small mandible is 1 point, and the score of the dental abnormality is 2 points. The basic score of the onychomental distance is only greater than 1 point from the basic score of the small mandible. At this time, the average of the basic score of the at least one factor and the basic scores of the other factors is calculated, and the average is used as the corrected score of the at least one factor and as the corrected score of the other factors.

[0171] It should be noted that the other factors are a subset of the remaining factors, and the remaining factors are the factors remaining from the positive correlation factors excluding the at least one factor.

[0172] When the correlation is negative, the logic is the same as that of positive correlation, so I will not elaborate on it.

[0173] Finally, a weighted calculation is performed on the multiple scores (corrected scores) corresponding to the multiple factors. The weight of each factor can be set according to the actual situation. The evaluation value is obtained after the weighted calculation, and the evaluation value is compared with the preset set value. The set value can be set from small to large. For example, the set value includes A, B, and C, where A<B<C. When the evaluation value is less than A, the airway condition is excellent. When the evaluation value is between AB, the airway condition is good. When the evaluation value is between BC, the airway condition is average. When the evaluation value is greater than C, the airway condition is poor.

[0174] The process of assessing airway status through image acquisition and analysis includes the acquisition of frontal images, lateral images, and frontal images of maximum mouth opening. Through these images, key airway parameters can be extracted, such as thyromental distance, small mandible, dental abnormality, upper lip occlusion, neck mobility, mouth opening, and Mahalanobis grade. These parameters are graded and scored to reflect the potential problems and risks of the airway. Subsequently, each parameter is weighted to obtain a comprehensive assessment value. Comparing the assessment value with the set value can determine the severity of the airway. This process can accurately and standardizedly assess airway status, help detect problems early, and provide reliable data support for clinical decision-making. In addition, as a non-invasive detection method, it is simple and easy to use, and can be widely used in clinical scenarios such as pre-anesthesia assessment and sleep apnea syndrome screening to improve the accuracy and objectivity of the assessment.

[0175] The user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this specification are all information and data authorized by the user or fully authorized by all parties. The acquisition, storage, use, and processing of data in the technical solutions described in the embodiments of this application are in compliance with the provisions of relevant laws and regulations.

[0176] Based on the above-mentioned airway assessment method based on depth camera image recognition, the embodiment of this specification also provides an airway assessment device based on depth camera image recognition. The device may include a system (including a distributed system), software (application), module, component, server, client, etc. using the method described in the embodiment of this specification and a device combined with necessary implementation hardware. Based on the same innovative concept, the device in one or more embodiments provided in the embodiment of this specification is as described in the following embodiments. Since the implementation scheme of the device to solve the problem is similar to the method, the implementation of the specific device in the embodiment of this specification can refer to the implementation of the aforementioned method, and the repetition will not be repeated. As used below, the term "unit" or "module" can be a combination of software and / or hardware that implements a predetermined function. Although the device described in the following embodiments is preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceived.

[0177] Specifically, Fig.12 This is a schematic diagram of the module structure of an embodiment of an airway assessment device based on depth camera image recognition provided in the embodiment of this specification, referring to Fig.12 As shown, an airway assessment device based on depth camera image recognition provided in an embodiment of this specification includes: an acquisition module 100, a first determination module 200, a second determination module 300, a scoring module 400, a weighting module 500, and a comparison module 600.

[0178] Acquisition module 100: Acquisition of depth images of the person to be identified, including front image, side image and front maximum mouth opening image;

[0179] The first determination module 200: obtains the thyromental distance, the degree of small mandible, the degree of dental abnormality, the degree of upper lip occlusion and the range of motion of the neck according to the front image and the side image;

[0180] The second determination module 300: obtains the mouth opening degree and Mahalanobis grade according to the frontal maximum mouth opening image;

[0181] Scoring module 400: grading and scoring the thyromental distance, small mandible, dental abnormality, upper lip occlusion, neck mobility, mouth opening and Mahalanobis classification of the person to be identified, respectively, to obtain multiple scores;

[0182] Weighting module 500: performing weighted calculation on the multiple scores to obtain an evaluation value;

[0183] Comparison module 600: compares the evaluation value with the set value to obtain the airway condition of the person to be identified.

[0184] The beneficial effects of the system provided above are consistent with the beneficial effects of the method provided above, and are not described in detail in the embodiments of this specification.

[0185] This embodiment provides a computer device, whose internal structure diagram can be shown as follows: Fig.13 As shown. The computer device includes a processor, a memory and a network interface connected through a system bus. 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 and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with an external terminal through a network connection.

[0186] Those skilled in the art will understand that Fig.13 The structure shown in the figure is only a block diagram of a part of the 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 those shown in the figure, or combine certain components, or have a different arrangement of components.

[0187] In one 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.

[0188] 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.

[0189] 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.

[0190] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods 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 this application can include at least one of non-volatile 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), magnetoresistive 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, etc., but are not limited to this.

[0191] It should also be understood that in the embodiments of this article, the term "and / or" is only a description of the association relationship of the associated objects, indicating that three relationships may exist. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship.

[0192] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in terms of function in the above description. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this article.

[0193] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0194] In the several embodiments provided herein, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of the units is only a logical function division, and there may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, or can be electrical, mechanical or other forms of connection.

[0195] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the embodiments of this article.

[0196] Specific embodiments are used in this article to illustrate the principles and implementation methods of this article. The description of the above embodiments is only used to help understand the methods and core ideas of this article. At the same time, for general technicians in this field, according to the ideas of this article, there will be changes in the specific implementation methods and application scopes. In summary, the content of this specification should not be understood as a limitation on this article.

Claims

1. An airway assessment method based on depth camera image recognition, characterized in that: include: Collecting depth images of the person to be identified, including front image, side image and front image with maximum mouth opening; According to the frontal image and the side image, the thyromental distance, the degree of small jaw, the degree of dental abnormality, the degree of upper lip occlusion and the range of motion of the neck are obtained; According to the frontal maximum mouth opening image, the mouth opening degree and Mahalanobis grade were obtained; The thyromental distance, small mandible, dental abnormality, upper lip occlusion, neck mobility, mouth opening and Mahalanobis classification of the person to be identified are graded and scored respectively to obtain multiple scores; Performing weighted calculation on the multiple scores to obtain an evaluation value; Comparing the evaluation value with the set value to obtain the airway condition of the person to be identified; The thyromental distance, small mandible, dental abnormality, upper lip occlusion, neck mobility, mouth opening and Mahalanobis classification of the person to be identified are graded and scored respectively to obtain multiple scores, further comprising: Determine the correlation between any two of the factors of the thyromental distance, micrognathia, dental abnormality, upper lip occlusion, neck mobility, mouth opening and Mahalanobis classification; According to the basic grading rules of each factor, the basic scores corresponding to the thyromental distance, small mandible, dental abnormality, upper lip occlusion, neck mobility, mouth opening and Mahalanobis classification of the person to be identified are obtained respectively; Based on the association relationship, a correction rule for each factor is obtained; According to the correction rules of the factors, each basic score is corrected respectively to obtain multiple scores; Determining the correlation between any two of the factors of the thyromental distance, the degree of small mandible, the degree of dental abnormality, the degree of upper lip occlusion, the range of motion of the neck, the degree of mouth opening and the Mahalanobis classification further comprises: Obtain the thyromental distance, small mandible, dental abnormality, upper lip occlusion, neck mobility, mouth opening and Mahalanobis classification corresponding to multiple known patients, and obtain multiple parameters corresponding to each factor; Analyze the correlation between each two factors one by one and obtain the correlation coefficient between each two factors; Two factors with correlation coefficients within a set coefficient range are determined to have a correlation relationship, wherein the correlation relationship includes positive correlation and negative correlation; The basic scores corresponding to the thyromental distance, small mandible, dental abnormality, upper lip occlusion, neck mobility, mouth opening and Mahalanobis classification of the person to be identified are obtained according to the basic classification rules of each factor, and further include: According to the multiple parameters corresponding to each factor, the parameter range corresponding to each factor is obtained; The parameter range is equally divided to obtain a plurality of intervals corresponding to each factor, wherein the number of intervals corresponding to each factor is equal; Arrange the multiple intervals corresponding to each factor according to the risk of airway obstruction from small to large, and obtain multiple intervals after arrangement; Assign values ​​to the multiple intervals after arrangement in ascending order according to the arrangement order to obtain the basic score corresponding to each factor; The correction rules for obtaining each factor based on the association relationship further include: When the correlation is positively correlated, the two factors with the correlation are determined as correlation factors; All factors with correlation factors are determined as positive correlation factors; When the difference between the basic score of at least one factor among the positively associated factors of the person to be identified and the basic scores of other factors is greater than a first set difference, the basic score of the at least one factor is corrected or the basic scores of other factors are corrected according to the ratio between the number of the at least one factor and the number of other factors; When the correlation is negatively correlated, the two factors with the correlation are determined as correlation factors; All factors with correlation factors were determined as negative correlation factors; When the difference between the basic score of at least one factor among the negatively associated factors of the person to be identified and the basic scores of other factors is less than the second set difference, the basic score of the at least one factor is corrected or the basic scores of other factors are corrected according to the ratio between the number of the at least one factor and the number of other factors.

2. The method according to claim 1, characterized in that The step of obtaining the thyromental distance, the degree of small mandible, the degree of dental abnormality, the degree of upper lip occlusion and the range of motion of the neck according to the front image and the side image further comprises: Obtaining a 3D model using 3D reconstruction technology according to the front image and the side image; By detecting the key points of the maxilla and mandible in the 3D model, the thyromental distance is obtained; By analyzing the length and angle of the mandible in the 3D model and combining it with standard facial proportions, the degree of small mandible is obtained; Analyzing the 3D model using a trained deep learning model to obtain the degree of tooth abnormality; The degree of upper lip occlusion is obtained according to the key points of the lower incisor and the upper lip in the side image corresponding to the lower incisor occluding the upper lip; The neck mobility is obtained according to the key points of the neck in the side images at multiple different neck angles.

3. The method according to claim 1, characterized in that The obtaining of the mouth opening degree and Mahalanobis classification according to the frontal maximum mouth opening image further comprises: Using two cameras at fixed positions, respectively collect the maximum mouth opening images of the person to be identified from the front, and obtain the maximum mouth opening images corresponding to the two cameras; The maximum mouth opening images corresponding to the two cameras are processed respectively to obtain composite images corresponding to the two cameras; Determine the disparity between the same facial feature points in two composite images; According to the parallax and the parameters of the camera, combined with the principle of similar triangles, the actual distance between the person to be identified and the camera is calculated; The actual distance between the two selected points is calculated based on the actual distance, the focal length of the camera and the image distance between the two selected points on the two composite images, in combination with the principle of similar triangles; The mouth opening degree and the Mahalanobis grade are obtained according to the actual distance between the two selected points.

4. The method according to claim 3, characterized in that The processing of the maximum mouth opening images corresponding to the two cameras to obtain the composite images corresponding to the two cameras further includes: For each camera, obtain the grayscale image and depth image corresponding to the maximum mouth opening image; Adjusting the resolution of the grayscale image and the depth image to be the same resolution; After adjusting the bit depth of the grayscale image to match the bit depth of the depth image, the grayscale image is stacked with the depth image to obtain a composite image for each camera.

5. The method according to claim 3, characterized in that: The two selected points are two points corresponding to the upper teeth and the lower teeth, or two points corresponding to the upper lip and the lower lip, respectively; The obtaining of the mouth opening degree and the Mahalanobis classification according to the actual distance between the two selected points further comprises: Obtaining the mouth opening according to the actual distance between the two selected points; The Mahalanobis classification is obtained according to the positions and visibility of the oral cavity feature points in the maximum mouth opening image.

6. The method according to claim 3, characterized in that Also includes: The face position is fixed by the chin-pull limiter. Using a sensor to measure the distance between the person to be identified and the camera; The actual distance between the person to be identified and the camera is compensated for an error according to the measured distance.

Images