Cervical vertebra area state acquisition method, device and equipment based on infrared image

Abstract

The invention provides a cervical vertebra area state acquisition method, device and equipment based on an infrared image. The method comprises the following steps: acquiring a back infrared image of a user; obtaining a cervical vertebra left side area and a cervical vertebra right side area from the back infrared image; acquiring an average temperature value AvgL corresponding to the left side area of the cervical vertebra and an average temperature value AvgR corresponding to the right side area of the cervical vertebra; and if (AvgL-AvgR) is greater than or equal to a first preset value, or if (AvgL-AvgR) is less than or equal to a second preset value, determining that the state of the cervical vertebra region corresponding to the user is a first state, otherwise, determining that the state of the cervical vertebra region corresponding to the user is a second state. According to the invention, another detection scheme can be provided for state detection of the cervical vertebra area.

Description

Technical Field

[0001] The invention relates to the field of image processing, and in particular to a device and equipment for acquiring the state of a cervical vertebra region based on infrared images. Background Art

[0002] Currently, the detection of the status of the cervical spine area mainly relies on doctors' experience and / or CT images. The detection method that relies solely on doctors' experience will have the problem of inaccurate judgment. The detection method of doctors' experience plus CT images, although it can improve the accuracy to a certain extent, will have the problem of high detection cost and will cause certain harm to the body. Summary of the invention

[0003] In view of the above technical problems, the technical solution adopted by the present invention is:

[0004] According to a first aspect of the present invention, a method for acquiring a cervical vertebra region state based on an infrared image is provided, the method comprising the following steps:

[0005] Get an infrared image of the user's back.

[0006] Based on the correlation features of the cervical vertebra area status correlation, a corresponding correlation area is determined, and based on the determined correlation area, a corresponding area is acquired from the back infrared image, wherein the correlation area includes a left cervical vertebra area and a right cervical vertebra area.

[0007] The average temperature value AvgL corresponding to the left side area of ​​the cervical vertebra and the average temperature value AvgR corresponding to the right side area of ​​the cervical vertebra are obtained.

[0008] If (AvgL-AvgR)≥t1, or if (AvgL-AvgR)≤t2, the state of the cervical vertebra area corresponding to the user is determined to be the first state, otherwise, the state of the cervical vertebra area corresponding to the user is determined to be the second state, t1 is the first preset value, t2 is the second preset value, and t1>t2.

[0009] According to a second aspect of the present invention, a device for acquiring a cervical vertebra region state based on an infrared image is provided, the device comprising:

[0010] The image acquisition module is used to acquire an infrared image of the user's back.

[0011] The first determination module is used to determine the corresponding associated area based on the associated features associated with the cervical vertebra area status, and to obtain the corresponding area from the back infrared image based on the determined associated area, wherein the associated area includes the left cervical vertebra area and the right cervical vertebra area.

[0012] The temperature acquisition module is used to acquire the average temperature value AvgL corresponding to the left side area of ​​the cervical vertebrae and the average temperature value AvgR corresponding to the right side area of ​​the cervical vertebrae.

[0013] The judgment module is used to determine that the state of the cervical vertebra area corresponding to the user is the first state when it is judged that (AvgL-AvgR)≥t1, or when it is judged that (AvgL-AvgR)≤t2; otherwise, determine that the state of the cervical vertebra area corresponding to the user is the second state, t1 is the first preset value, t2 is the second preset value, and t1>t2.

[0014] According to a third aspect of the present invention, there is provided an electronic device comprising a processor and a memory; the processor is used to execute the steps of the method described in the first aspect of the present invention by calling a program or instruction stored in the memory.

[0015] The present invention has at least the following beneficial effects:

[0016] The method, device and apparatus for acquiring the state of the cervical vertebra area based on infrared images provided in the embodiments of the present invention can judge the state of the cervical vertebra area based on infrared images, thereby providing another judgment scheme for judging the state of the cervical vertebra area.

[0017] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present invention, nor are they intended to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0019] Figure 1 A flowchart of a method for acquiring the state of the cervical vertebrae region based on infrared images provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0020] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0021] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art of the present invention. The terms used herein in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. The term "and / or" used herein includes any and all combinations of one or more related listed items.

[0022] It should be noted that some exemplary embodiments are described as processes or methods depicted as flow charts. Although the flow charts describe the steps as sequential processes, many of the steps can be implemented in parallel, concurrently or simultaneously. In addition, the order of the steps can be rearranged. The process can be terminated when its operation is completed, but it can also have additional steps not included in the accompanying drawings. The process can correspond to a method, function, procedure, subroutine, subprogram, etc.

[0023] The embodiment of the present invention provides a method for acquiring the state of the cervical vertebra region based on infrared images, the method comprising: Figure 1 Steps shown:

[0024] S100, obtaining an infrared image of the user's back.

[0025] In an embodiment of the present invention, the back infrared image may be an image that exposes the cervical spine area, and the back infrared image of the user may be obtained by an infrared shooting device. When shooting, the user may shoot according to the specified shooting requirements to obtain a back infrared image that meets the requirements. The specified shooting requirements may be determined according to actual conditions.

[0026] S200, determining a corresponding associated area based on associated features associated with the cervical vertebra area status, and acquiring a corresponding area from the back infrared image based on the determined associated area, wherein the associated area includes a left cervical vertebra area and a right cervical vertebra area.

[0027] In the embodiment of the present invention, the correlation feature of the cervical vertebrae regional state correlation can be obtained by the following steps:

[0028] S10, based on n candidate features associated with the cervical vertebrae region state, obtain Q sample data sets, the i-th sample data set includes (U i , F i1 , F i2 , ..., F ij , ..., F in , S i ), U i is the ID corresponding to the i-th sample data, F ij is the value corresponding to the jth candidate feature corresponding to the i-th sample data, S iis the state of the cervical spine area corresponding to the i-th sample data, the value of i ranges from 1 to Q, and the value of j ranges from 1 to n; the value corresponding to each candidate feature is obtained based on the corresponding infrared image.

[0029] In the embodiment of the present invention, n is a natural number greater than 1. The n candidate features may be features associated with the state of the cervical spine region, which may be obtained based on medical knowledge, expert experience, and analysis of medical records during the cervical spine diagnosis process. The candidate features may include acupuncture points of interest and characteristic temperatures of the region of interest related to the state of the cervical spine region. If it is an acupuncture point of interest, the characteristic temperature is the infrared temperature corresponding to the acupuncture point. If it is a region of interest, the characteristic temperature may be the average value, maximum value, minimum value, etc. of the infrared temperature corresponding to the region.

[0030] In the embodiment of the present invention, according to the distribution of the acupuncture points and regions of interest corresponding to the candidate features on the body, infrared images of different body positions may be taken for each test user to obtain Q sample data sets.

[0031] In the embodiment of the present invention, the state of the cervical vertebrae region may include a first state indicating that the cervical vertebrae region is abnormal, such as cervical spondylosis, and a second state indicating that the cervical vertebrae region is normal.

[0032] S20, obtain f(r) candidate features from the current second intermediate candidate feature set and add them to the current first intermediate candidate feature set as the rth training feature set, and delete the f(r) candidate features from the current second intermediate candidate feature set; the initial value of r is 1, the initial value of the current second intermediate candidate feature set includes n candidate features, and the initial set of the current first intermediate candidate feature set is an empty set.

[0033] In the embodiment of the present invention, f(r) candidate features may be selected based on expert experience.

[0034] S30, based on the rth training feature set, use the data corresponding to the Q sample data sets to train the AI ​​model to obtain the corresponding rth prediction result and the corresponding rth feature weight set. If it is determined based on the rth prediction result that the prediction accuracy of the AI ​​model is greater than the preset accuracy threshold, the candidate features corresponding to the rth training feature set are used as associated features associated with the cervical spine area status. Otherwise, execute S40.

[0035] In an embodiment of the present invention, the AI ​​model may be a deep neural network model. In an exemplary embodiment, it may be a decision tree model.

[0036] During the AI ​​model training process, the importance of each feature to the prediction result, that is, the feature weight, can be known. Therefore, the method for obtaining the feature weight set can be the existing technology.

[0037] In an embodiment of the present invention, the prediction result of each sample data set may be the state of the cervical spine region predicted based on the sample data set. The prediction accuracy of the AI ​​model may be calculated using an existing accuracy calculation method, such as a mean square error calculation method.

[0038] In the embodiment of the present invention, the preset accuracy threshold may be an empirical value, and in an exemplary embodiment, may be 0.8.

[0039] S40, sort the rth feature weight set in descending order to obtain a sorted feature weight set, and obtain the candidate features corresponding to the first K feature weights in the sorted feature weight set as the current first intermediate candidate features to obtain the current first intermediate candidate feature set; K = [g(r)*f], g(r) is the number of feature weights in the rth feature weight set, f is the number of horizontal inches between the Tiantu point and the outermost side of the cervical spine area, 0<f<1.

[0040] In the embodiment of the present invention, f can be set based on actual needs. In an exemplary embodiment, 0.6≤f≤0.8, preferably, f=0.6.

[0041] S50, set r=r+1, and execute S20.

[0042] The technical effect of S10 to S50 is that the most relevant associated feature to the state of the cervical region can be selected from n candidate features associated with the state of the cervical region. In the embodiment of the present invention, the finally selected associated features are the average temperature value of the left cervical region and the average temperature value of the right cervical region.

[0043] In an exemplary embodiment of the present invention, the left cervical region and the right cervical region can be obtained by the following steps:

[0044] S201, identifying the Dazhui point, the Zuofengchi point and the Youfengchi point in the back infrared image, and obtaining the positions L1, L2 and L3 of the Dazhui point, the Zuofengchi point and the Youfengchi point respectively, wherein L1=(x1, y1), L2=(x2, y2), L3=(x3, y3), x1 is the horizontal coordinate of the Dazhui point in the coordinate system corresponding to the back infrared image, y1 is the vertical coordinate of the Dazhui point in the coordinate system corresponding to the back infrared image, x2 is the horizontal coordinate of the Zuofengchi point in the coordinate system corresponding to the back infrared image, y2 is the vertical coordinate of the Zuofengchi point in the coordinate system corresponding to the back infrared image, x3 is the horizontal coordinate of the Youfengchi point in the coordinate system corresponding to the back infrared image, and y3 is the vertical coordinate of the Youfengchi point in the coordinate system corresponding to the back infrared image; the origin of the coordinate system is located at the upper left corner of the back infrared image, the x-axis is the horizontal direction of the human body, and the y-axis is the longitudinal direction of the human body.

[0045] In an embodiment of the present invention, the Dazhui point, the Zuofengchi point, and the Youfengchi point can be identified based on a trained human posture recognition model. The trained human posture recognition model can be obtained by inputting sample images labeled with human acupoints into a neural network model for training, and the specific training method can be the existing technology.

[0046] S202, based on x1, obtain a first endpoint and a second endpoint, wherein the horizontal coordinate x1 of the first endpoint e =x1-△x, the vertical coordinate of the first endpoint y1 e =y1, the horizontal coordinate of the second endpoint x2 e =x2+a*△c, the ordinate of the second endpoint is y1 e =y1, △c is the horizontal inch length, a is the number of horizontal inches between the Tiantu point and the outermost side of the cervical spine area, which can be determined based on traditional Chinese medicine theory. In one example, a=2.5 inches.

[0047] In the embodiment of the present invention, Δc=|x2-x1| / W 21 , W 21 The number of horizontal inches between the Tiantu point and the left shoulder can be determined based on traditional Chinese medicine theory.

[0048] S203, obtaining the intersection point (x0, y0) of a first straight line and a second straight line, wherein the first straight line is a straight line passing through L1 and parallel to the y-axis, and the second straight line is a straight line passing through L2 and L3.

[0049] S204, sequentially connect the first endpoint, L1, the intersection point and L2 to obtain the left side region of the cervical vertebrae, and sequentially connect the second endpoint, L2, the intersection point and L3 to obtain the right side region of the cervical vertebrae.

[0050] In another exemplary embodiment, the left cervical region and the right cervical region may be obtained by the following steps:

[0051] S210, identifying the Dazhui point, Zuofengchi point, Youfengchi point, left shoulder and right shoulder in the back infrared image, and obtaining the positions L1, L2, L3, L4 and L5 of the Dazhui point, Zuofengchi point, Youfengchi point, left shoulder and right shoulder respectively, wherein L1=(x1, y1), L2=(x2, y2), L3=(x3, y3), x1 is the horizontal coordinate of the Dazhui point in the coordinate system corresponding to the back infrared image, y1 is the vertical coordinate of the Dazhui point in the coordinate system corresponding to the back infrared image, x2 is the horizontal coordinate of the Zuofengchi point in the coordinate system corresponding to the back infrared image, y2 is the vertical coordinate of the Zuofengchi point in the coordinate system corresponding to the back infrared image, x3 is the ordinate of the right Fengchi acupoint in the coordinate system corresponding to the back infrared image, x3 is the abscissa of the right Fengchi acupoint in the coordinate system corresponding to the back infrared image, y3 is the ordinate of the right Fengchi acupoint in the coordinate system corresponding to the back infrared image, x4 is the abscissa of the left shoulder in the coordinate system corresponding to the back infrared image, y4 is the ordinate of the left shoulder in the coordinate system corresponding to the back infrared image, x5 is the abscissa of the right shoulder in the coordinate system corresponding to the back infrared image, y5 is the ordinate of the right shoulder in the coordinate system corresponding to the back infrared image; the origin of the coordinate system is located in the upper left corner of the back infrared image.

[0052] S211, if (y4-y5)>△d1, set y2=y2+△d2; if (y5-y4)>△d1, set y1=y1+△d2, where △d1 is the preset distance difference and △d2 is the preset adjustment value.

[0053] In the embodiment of the present invention, if (y4-y5)>Δd1, or (y5-y4)>Δd1, it indicates that the corresponding user has uneven shoulders and needs to repair the cervical spine area.

[0054] In the embodiment of the present invention, Δd1 may be an empirical value, for example, 1 vertical inch. In the embodiment of the present invention, 1 vertical inch may be equal to |y0-y1| / V 01 , where y0 is the ordinate of the Changqiang point in the back infrared image, V 01 It is the number of straight inches between Dazhui point and Changqiang point, which can be determined based on traditional Chinese medicine theory.

[0055] In an exemplary embodiment of the present invention, Δd2 may be a fixed value or an empirical value.

[0056] In another embodiment of the present invention, Δd2 can be obtained by referring to a preset adjustment value reference table, wherein the hth row of the preset adjustment value reference table includes R h and △d h , R h is the absolute value range of the difference between the ordinate of the left shoulder and the ordinate of the right shoulder corresponding to the hth row of data, △d his the adjustment value corresponding to the hth row of data; the value of h is 1 to Z, and Z is the number of row data in the preset adjustment value reference table.

[0057] The preset adjustment value reference table is obtained by the following steps:

[0058] Obtain infrared images of the backs of P1 users.

[0059] Based on P1 back infrared images, the shoulder coordinate set G = {G1, G2, ..., G t1 , ..., G P1}, G t1 is the shoulder coordinate in the t1th back infrared image, G t1 =(y L t1 ,y R t1 ), y L t1 is the ordinate of the left shoulder in the t1th back infrared image, y R t1 is the ordinate of the right shoulder in the t1th back infrared image, and the value of t1 ranges from 1 to P1.

[0060] Traverse G, if |y L t1 -y R t1 ∣>△d1,∣y L t1 -y R t1 ∣Add to the current difference set. The initial value of the current difference set is empty.

[0061] Cluster the differences in the current difference set to obtain Z clusters.

[0062] In the embodiment of the present invention, existing clustering methods may be used for clustering.

[0063] The characteristic value corresponding to the h-th cluster is obtained as the adjustment value corresponding to the h-th cluster.

[0064] In the embodiment of the present invention, the characteristic values ​​corresponding to each cluster may include an average temperature value, a maximum temperature value, a minimum temperature value, a temperature mean square error, and the like.

[0065] Based on the difference range and the corresponding adjustment value corresponding to each cluster, the preset adjustment value reference table is obtained.

[0066] Compared with the method of using a fixed value as Δd2, in this embodiment, different left and right shoulder differences have different Δd2, so the obtained cervical vertebra area can be more accurate.

[0067] S212, based on x1, obtain the first endpoint and the second endpoint, wherein the horizontal coordinate x1 of the first endpoint e =x1-△x, the vertical coordinate of the first endpoint y1 e =y1, the horizontal coordinate of the second endpoint x2 e =x2+a*△c, the ordinate of the second endpoint is y1 e =y1, △c is the horizontal length, a is the number of horizontal inches between Tiantu point and the outermost side of the cervical spine area.

[0068] S213, obtaining the intersection point (x0, y0) of the first straight line and the second straight line, where the first straight line is a straight line passing through L1 and parallel to the y-axis, and the second straight line is a straight line passing through L2 and L3.

[0069] S214, sequentially connect the first endpoint, L1, the intersection point and L2 to obtain the left side region of the cervical vertebrae, and sequentially connect the second endpoint, L2, the intersection point and L3 to obtain the right side region of the cervical vertebrae.

[0070] Compared with the previous embodiments, this embodiment takes into account the influence of high and low shoulders on the cervical spine area, so that the cervical spine area can be obtained more accurately.

[0071] S300, obtaining an average temperature value AvgL corresponding to the left side area of ​​the cervical vertebrae and an average temperature value AvgR corresponding to the right side area of ​​the cervical vertebrae.

[0072] In the embodiment of the present invention, AvgL=(TL1+TL2+...+TL i1 +……+TL w1 ) / w1,TL i1 is the temperature corresponding to the i1th pixel in the left cervical vertebra area, i1 ranges from 1 to w1, w1 is the number of pixels in the left cervical vertebra area. AvgR=(TR1+TR2+……+TR i2 +……+TR w2 ) / w2,TR i2 is the temperature corresponding to the i2th pixel in the right cervical region, i2 ranges from 1 to w2, and w2 is the number of pixels in the right cervical region. It is known to those skilled in the art that any method of obtaining the temperature of each pixel in the cervical region from an infrared image falls within the protection scope of the present invention.

[0073] S400, if (AvgL-AvgR)≥t1, or, if (AvgL-AvgR)≤t2, it indicates that there is an abnormality in the cervical vertebra area, and the state of the cervical vertebra area corresponding to the user is determined to be the first state; otherwise, that is, if t2<(AvgL-AvgR)<t1, it indicates that the cervical vertebra area is normal, and the state of the cervical vertebra area corresponding to the user is determined to be the second state, t1 is the first preset value, t2 is the second preset value, and t1>t2.

[0074] In the embodiment of the present invention, t1 and t2 can be obtained based on the training process of S10 to S50. In a specific embodiment, t1 = 0.345°C, t2 = -0.355°C.

[0075] The infrared image-based cervical vertebra area status acquisition provided by the embodiment of the present invention has at least the following advantages:

[0076] (1) By dividing the cervical spine into left and right partitions and comparing the temperature difference between the two sides, the temperature distribution of the cervical spine area can be analyzed more finely. This partitioning method can capture more subtle temperature changes, thereby improving the accuracy of the status detection results of the cervical spine area.

[0077] (2) Different individuals may have different morphology and structure of the cervical spine, and traditional infrared diagnostic methods may produce errors due to individual differences. The present invention can eliminate the influence of individual differences on the state detection results of the cervical spine area to a certain extent by comparing the temperature difference between the left and right sides.

[0078] (3) By measuring and comparing the temperature difference on both sides of the cervical spine, specific numerical indicators can be obtained. This provides doctors with a more objective basis for diagnosis and helps reduce the subjectivity and uncertainty of subjective judgment.

[0079] (4) The present invention is based on infrared thermal imaging technology and does not require invasive operation or drug injection on the patient, so it has the advantages of being non-invasive and non-invasive. This reduces the pain and discomfort of the patient and improves the comfort of the status detection of the cervical spine area.

[0080] (5) Infrared thermal imaging technology has the characteristics of rapid measurement and can obtain temperature distribution information of the cervical spine area in a short time. Combined with automated data analysis and processing software, a fast and easy detection process can be achieved.

[0081] (6) The manufacturing and maintenance costs of infrared thermal imagers and related equipment are relatively low, so the present invention has advantages in cost-effectiveness. This helps to reduce the operating costs of medical institutions and increase the popularity of detection technology.

[0082] (7) Since infrared thermal imaging technology can be used for repeated measurements and continuous monitoring of the same patient, the present invention has the advantages of repeatability and monitorability, which helps doctors to continuously track and evaluate the patient's condition.

[0083] Based on the same inventive concept, an embodiment of the present invention provides a device for acquiring the state of the cervical vertebra region based on infrared images, and the device may include:

[0084] The image acquisition module is used to acquire an infrared image of the user's back.

[0085] The first determination module is used to determine the corresponding associated area based on the associated features associated with the cervical vertebra area status, and to obtain the corresponding area from the back infrared image based on the determined associated area, wherein the associated area includes the left cervical vertebra area and the right cervical vertebra area.

[0086] The temperature acquisition module is used to acquire the average temperature value AvgL corresponding to the left side area of ​​the cervical vertebrae and the average temperature value AvgR corresponding to the right side area of ​​the cervical vertebrae.

[0087] The judgment module is used to determine that the state of the cervical vertebra area corresponding to the user is the first state when it is judged that (AvgL-AvgR)≥t1, or when it is judged that (AvgL-AvgR)≤t2; otherwise, determine that the state of the cervical vertebra area corresponding to the user is the second state, t1 is the first preset value, t2 is the second preset value, and t1>t2.

[0088] The device can be used to perform Figure 1 Therefore, for the functions that can be realized by each functional module of the device, please refer to Figure 1 The description of the illustrated embodiment is not repeated in detail.

[0089] An embodiment of the present invention also provides an electronic device, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are configured to execute the method described in the embodiment of the present invention.

[0090] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution disclosed in the present invention can be achieved, and this document does not limit this.

[0091] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for acquiring the state of the cervical vertebra area based on infrared images, characterized in that: The method comprises the following steps: Acquire an infrared image of the user's back; Based on the correlation features of the cervical vertebra area status correlation, a corresponding correlation area is determined, and based on the determined correlation area, a corresponding area is acquired from the back infrared image, wherein the correlation area includes a left cervical vertebra area and a right cervical vertebra area; Obtain an average temperature value AvgL corresponding to the left side area of ​​the cervical vertebra and an average temperature value AvgR corresponding to the right side area of ​​the cervical vertebra; If (AvgL-AvgR)≥t1, or if (AvgL-AvgR)≤t2, the state of the cervical vertebra area corresponding to the user is determined to be the first state, otherwise, the state of the cervical vertebra area corresponding to the user is determined to be the second state, t1 is the first preset value, t2 is the second preset value, and t1>t2.

2. The method according to claim 1, characterized in that The left side area of ​​the cervical vertebra and the right side area of ​​the cervical vertebra are obtained by the following steps: Identify the Dazhui point, the Zuofengchi point and the Youfengchi point in the back infrared image, and obtain the positions L1, L2 and L3 of the Dazhui point, the Zuofengchi point and the Youfengchi point respectively, wherein L1=(x1, y1), L2=(x2, y2), L3=(x3, y3), x1 is the horizontal coordinate of the Dazhui point in the coordinate system corresponding to the back infrared image, y1 is the vertical coordinate of the Dazhui point in the coordinate system corresponding to the back infrared image, x2 is the horizontal coordinate of the Zuofengchi point in the coordinate system corresponding to the back infrared image, y2 is the vertical coordinate of the Zuofengchi point in the coordinate system corresponding to the back infrared image, x3 is the horizontal coordinate of the Youfengchi point in the coordinate system corresponding to the back infrared image, y3 is the vertical coordinate of the Youfengchi point in the coordinate system corresponding to the back infrared image; the origin of the coordinate system is located at the upper left corner of the back infrared image; Based on x1, get the first endpoint and the second endpoint, where the horizontal coordinate x1 of the first endpoint e =x1-△x, the vertical coordinate of the first endpoint y1 e =y1, the horizontal coordinate of the second endpoint x2 e =x2+a*△c, the ordinate of the second endpoint is y1 e =y1, △c is the horizontal length, a is the number of horizontal inches between Tiantu point and the outermost side of the cervical vertebrae area; Get the intersection point (x0, y0) of the first straight line and the second straight line, where the first straight line is a straight line passing through L1 and parallel to the y-axis, and the second straight line is a straight line passing through L2 and L3; The first endpoint, L1, the intersection point and L2 are sequentially connected to obtain the left side region of the cervical vertebrae, and the second endpoint, L2, the intersection point and L3 are sequentially connected to obtain the right side region of the cervical vertebrae.

3. The method according to claim 1, characterized in that The left side area of ​​the cervical vertebra and the right side area of ​​the cervical vertebra are obtained by the following steps: The Dazhui point, the Zuo Fengchi point, the You Fengchi point, the left shoulder and the right shoulder in the back infrared image are identified, and the positions L1, L2, L3, L4 and L5 of the Dazhui point, the Zuo Fengchi point, the You Fengchi point, the left shoulder and the right shoulder are obtained respectively, wherein L1=(x1, y1), L2=(x2, y2), L3=(x3, y3), x1 is the horizontal coordinate of the Dazhui point in the coordinate system corresponding to the back infrared image, y1 is the vertical coordinate of the Dazhui point in the coordinate system corresponding to the back infrared image, x2 is the horizontal coordinate of the Zuo Fengchi point in the coordinate system corresponding to the back infrared image, y2 is the vertical coordinate of the Zuo Fengchi point in the coordinate system corresponding to the back infrared image, The ordinate in the coordinate system corresponding to the back infrared image, x3 is the abscissa of the right Fengchi acupoint in the coordinate system corresponding to the back infrared image, y3 is the ordinate of the right Fengchi acupoint in the coordinate system corresponding to the back infrared image, x4 is the abscissa of the left shoulder in the coordinate system corresponding to the back infrared image, y4 is the ordinate of the left shoulder in the coordinate system corresponding to the back infrared image, x5 is the abscissa of the right shoulder in the coordinate system corresponding to the back infrared image, y5 is the ordinate of the right shoulder in the coordinate system corresponding to the back infrared image; the origin of the coordinate system is located at the upper left corner of the back infrared image; If (y4-y5)>△d1, set y2=y2+△d2; if (y5-y4)>△d1, set y1=y1+△d2, where △d1 is the preset distance difference and △d2 is the preset adjustment value; Based on x1, get the first endpoint and the second endpoint, where the horizontal coordinate x1 of the first endpoint e =x1-△x, the vertical coordinate of the first endpoint y1 e =y1, the horizontal coordinate of the second endpoint x2 e =x2+a*△c, the ordinate of the second endpoint is y1 e =y1, △c is the horizontal length, a is the number of horizontal inches between Tiantu point and the outermost side of the cervical vertebrae area; Get the intersection point (x0, y0) of the first straight line and the second straight line, where the first straight line is a straight line passing through L1 and parallel to the y-axis, and the second straight line is a straight line passing through L2 and L3; The first endpoint, L1, the intersection point and L2 are sequentially connected to obtain the left side region of the cervical vertebrae, and the second endpoint, L2, the intersection point and L3 are sequentially connected to obtain the right side region of the cervical vertebrae.

4. The method according to claim 3, characterized in that Δd2 is obtained by referring to a preset adjustment value reference table, wherein the hth row of the preset adjustment value reference table includes R h and △d h , R h is the absolute value range of the difference between the ordinate of the left shoulder and the ordinate of the right shoulder corresponding to the hth row of data, △d h is the adjustment value corresponding to the hth row of data; the value of h ranges from 1 to Z, and Z is the number of row data in the preset adjustment value reference table; The preset adjustment value reference table is obtained by the following steps: Obtain infrared images of the backs of P1 users; Based on P1 back infrared images, the shoulder coordinate set G = {G1, G2, ..., G t1 , ..., G P1 }, G t1 is the shoulder coordinate in the t1th back infrared image, G t1 =(y L t1 ,y R t1 ), y L t1 is the ordinate of the left shoulder in the t1th back infrared image, y R t1 is the ordinate of the right shoulder in the t1th back infrared image, and the value of t1 ranges from 1 to P1; Traverse G, if |y L t1 -y R t1 ∣>△d1,∣y L t1 -y R t1 ∣Add to the current difference set, the initial value of the current difference set is empty; Cluster the differences in the current difference set to obtain Z clusters; Obtain the characteristic value corresponding to the h-th cluster as the adjustment value corresponding to the h-th cluster; Based on the difference range and the corresponding adjustment value corresponding to each cluster, the preset adjustment value reference table is obtained.

5. The method according to claim 1, characterized in that The associated features of the cervical vertebrae region status association are obtained by the following steps: S10, based on n candidate features associated with the cervical vertebrae region state, obtain Q sample data sets, the i-th sample data set includes (U i , F i1 , F i2 , ..., F ij , ..., F in , S i ), U i is the ID corresponding to the i-th sample data, F ij is the value corresponding to the jth candidate feature corresponding to the i-th sample data, S i is the state of the cervical spine area corresponding to the i-th sample data, i ranges from 1 to Q, and j ranges from 1 to n; the value corresponding to each candidate feature is obtained based on the corresponding infrared image; S20, obtaining f(r) candidate features from the current second intermediate candidate feature set and adding them to the current first intermediate candidate feature set as the rth training feature set, and deleting the f(r) candidate features from the current second intermediate candidate feature set; The initial value of r is 1, the initial value of the current second intermediate candidate feature set includes n candidate features, and the initial set of the current first intermediate candidate feature set is an empty set; S30, based on the rth training feature set, the AI ​​model is trained using the data corresponding to the Q sample data sets to obtain the corresponding rth prediction result and the corresponding rth feature weight set. If it is determined based on the rth prediction result that the prediction accuracy of the AI ​​model is greater than a preset accuracy threshold, the candidate feature corresponding to the rth training feature set is used as the associated feature associated with the cervical vertebrae region state. Otherwise, S40 is executed; S40, sorting the rth feature weight set in descending order to obtain a sorted feature weight set, and obtaining candidate features corresponding to the first K feature weights in the sorted feature weight set as the current first intermediate candidate features to obtain the current first intermediate candidate feature set; K = [g(r)*f], g(r) is the number of feature weights in the rth feature weight set, f is the number of horizontal inches between the Tiantu point and the outermost side of the cervical vertebrae region, 0 < f < 1; S50, set r=r+1, and execute S20.

6. The method according to claim 4, characterized in that The AI ​​model is a decision tree model.

7. The method according to claim 1, characterized in that △c=|x2-x1| / W 21 , W 21 It is the number of horizontal inches between Tiantu point and left shoulder.

8. The method according to claim 1, characterized in that t1=0.345℃, t2=-0.355℃.

9. A device for acquiring the state of the cervical vertebra area based on infrared images, characterized in that: The device comprises: An image acquisition module, used to acquire an infrared image of the user's back; A first determination module is used to determine a corresponding associated area based on the associated features associated with the cervical vertebra area state, and to obtain a corresponding area from the back infrared image based on the determined associated area, wherein the associated area includes a left cervical vertebra area and a right cervical vertebra area; A temperature acquisition module, used to acquire an average temperature value AvgL corresponding to the left side area of ​​the cervical vertebrae and an average temperature value AvgR corresponding to the right side area of ​​the cervical vertebrae; The judgment module is used to determine that the state of the cervical vertebra area corresponding to the user is the first state when it is judged that (AvgL-AvgR)≥t1, or when it is judged that (AvgL-AvgR)≤t2; otherwise, determine that the state of the cervical vertebra area corresponding to the user is the second state, t1 is the first preset value, t2 is the second preset value, and t1>t2.

10. An electronic device, characterized in that: including a processor and a memory; The processor is used to execute the steps of the method according to any one of claims 1 to 8 by calling the program or instruction stored in the memory.

Images