Meibomian gland absence detection method, system and apparatus, and medium

Abstract

The present invention relates to the technical field of image processing, and relates to a meibomian gland absence detection method, system and apparatus, and a medium. The meibomian gland absence detection method comprises the steps of image acquisition, image processing, first grayscale value segmentation point acquisition, first image segmentation, n-th grayscale value segmentation point acquisition, n-th image segmentation, grayscale value determination, area calculation, absence rate calculation, output, etc. The meibomian gland absence detection system comprises an acquisition module, a processing module, a calculation module I, a segmentation module I, a calculation module II, a segmentation module II, a determination module, a calculation module III, and an output module. The present invention can reduce the intervention of human factors in a meibomian gland absence detection process, thereby reducing the error of meibomian gland absence detection, thus improving the accuracy of meibomian gland absence detection.

Description

A meibomian gland absence detection method, system, device and medium TECHNICAL FIELD

[0001] The present application relates to the technical field of image processing, and in particular to a meibomian gland absence detection method, system, device and medium. BACKGROUND

[0002] The meibomian gland is an enlarged modified sebaceous gland, which is composed of two independent functional structural units of acinar and duct, and has the functions of synthesizing, storing and secreting lipids. The lipids secreted by the meibomian gland are called meibum, which participates in the formation of the lipid layer of the tear film. The lipid layer of the tear film plays an important role in maintaining the stability of the tear film, reducing the evaporation rate of the aqueous layer of the tear film, preventing the outflow of tear fluid, and resisting microorganisms.

[0003] Currently, by trying to use image processing algorithms or software, doctors can more accurately and conveniently obtain information of the meibomian gland, increase the contrast of the meibomian gland image, highlight the gland, and facilitate the observation of the meibomian gland by doctors. The image is analyzed, and the meibomian gland presence, absence area is manually drawn, and then the presence area and absence area of the meibomian gland are calculated.

[0004] However, when obtaining the absence area of the meibomian gland, the process of drawing the meibomian gland is relatively cumbersome and time-consuming, and the accuracy of manual drawing is low, and human errors are easily introduced in the detection process. Therefore, there is an urgent need for a meibomian gland absence detection method with high precision to detect the absence of the meibomian gland. TECHNICAL PROBLEM

[0005] In order to improve the accuracy of meibomian gland detection, the present application provides a meibomian gland absence detection method, system, device and medium. TECHNICAL SOLUTION

[0006] In a first aspect, the present application provides a meibomian gland absence detection method, which adopts the following technical scheme:

[0007] A meibomian gland absence detection method, comprising:

[0008] Image acquisition: open the upper and lower eyelids, take a photo of the eye, and obtain an eye photo;

[0009] Image processing: performing gray scale processing on the eye photo, so that each pixel in the eye photo corresponds to a gray scale value K, and obtaining the total number of gray scale values m, the maximum gray scale value Nmax and the minimum gray scale value Nmin;

[0010] First gray scale value segmentation point acquisition: obtaining a first gray scale value segmentation point M1, and the acquisition model of the first gray scale value segmentation point M1 is as follows: ;

[0011] First image segmentation: taking the first gray value segmentation point M1 as the reference, the m gray values in the eye photo are distinguished, and set A1 and set B1 are formed, the elements in set A1 are the gray values K less than M1, and the elements in set B1 are the gray values K greater than M1;

[0012] n-th gray value segmentation point acquisition: acquiring the n-th gray value segmentation point M n n-th gray value segmentation point M n The acquisition model is as follows:

[0013] In the formula, i is the number of elements in set A n-1 K i is the gray value less than the (n-1)-th gray value segmentation point M n-1 , j is the number of elements in set B n-1 K j is the gray value greater than the (n-1)-th gray value segmentation point M n-1 ;

[0014] n-th image segmentation: taking the n-th gray value segmentation point M n as the reference, and replacing M n with Mn-1, the m gray values in the eye photo are distinguished, and set A n and set B n are formed, the elements in set A n are the gray values K less than M n , the elements in set B n are the gray values K greater than M n , and set An replaces set A n-1 , and set B n replaces Bn-1;

[0015] Gray value judgment: setting a gray value threshold , if M n -M n-1 ≥ , the n-th gray value segmentation point acquisition step is performed, and the n-th image segmentation step is performed, if M n -M n-1 < , the area calculation step is performed;

[0016] Area calculation: calculating the area S of the meibomian gland, and the calculation model of the meibomian gland area S is as follows:

[0017] In the formula, X is the width of the eye photo, Y is the length of the eye photo, m is the total number of gray values, and Z is the number of gray values in the meibomian gland region.​​

[0018] The calculation model of the number of gray values Z in the meibomian gland area is as follows:

[0019] :

[0020] ; :

[0021] ;

[0022] Output: output the processed eye photo and the numerical value of the area S of the meibomian gland.

[0023] In the detection of the meibomian gland, the eyelid of the person to be detected is opened, the eye is photographed, and the eye photo is subjected to gray scale processing, so that each pixel in the eye photo corresponds to a gray value K, and the total number of gray values m, the maximum gray value Nmax and the minimum gray value Nmin are obtained, the first degree value division point M1 is calculated from the maximum gray value Nmax and the minimum gray value Nmin, the m gray values in the eye photo are distinguished based on M1, and sets A1 and B1 are formed, the elements in set A1 are gray values with a gray value K less than M1, and the elements in set B1 are gray values with a gray value K greater than M1, after n times of calculation, the value of M n -M n-1 ≥ when M n n is greater than Mn, it is proved that the area of the meibomian gland has been separated from the eye photo, and by obtaining the proportion of the number of gray values greater than Mn in the total number of gray values, the area of the meibomian gland in the eye image is obtained. Advantages

[0024] The technical solutions described above in the embodiments of the present application have at least the following technical effects:

[0025] By processing the eye photo, the meibomian gland is highlighted, and the meibomian gland is observed more clearly, when the meibomian gland is highlighted, the area of the meibomian gland is obtained by calculating the ratio of the number of gray values in the meibomian gland area to the number of gray values in the whole eye photo; in the process of detecting the absence of the meibomian gland, the intervention of human factors is reduced, and the error of detecting the absence of the meibomian gland is reduced, and the accuracy of detecting the absence of the meibomian gland is improved.

[0026] Optionally, the area calculation step is further provided with a missing rate calculation step;

[0027] Missing rate calculation: calculate the missing rate V of the meibomian gland, and the calculation model of the meibomian gland missing rate V is as follows: ; In the formula, S1 is the area of the standard meibomian gland.

[0028] Optionally, the output step also outputs the absence rate V of the meibomian glands.

[0029] In a second aspect, the present application provides a meibomian gland absence detection system, which adopts the following technical scheme:

[0030] A meibomian gland absence detection system comprises an acquisition module, a processing module, a calculation module I, a segmentation module I, a calculation module II, a segmentation module II, a judgment module, a calculation module III, and an output module.

[0031] The acquisition module is electrically connected to the input end of the processing module, and is configured to take a photo of an eye and acquire an eye photo.

[0032] The processing module is electrically connected to the output end of the acquisition module, and is electrically connected to the input end of the calculation module I, and is configured to perform grayscale processing on the eye photo acquired by the acquisition module, so that each pixel in the eye photo corresponds to a grayscale value K, and the total number of grayscale values m, the maximum grayscale value Nmax, and the minimum grayscale value Nmin are acquired.

[0033] The calculation module I is electrically connected to the input end of the segmentation module I, and is configured to calculate a first grayscale value segmentation point M1, and the acquisition model of the first grayscale value segmentation point M1 is as follows: ;

[0034] The segmentation module I is electrically connected to the input end of the calculation module II, and is configured to distinguish the m grayscale values in the eye photo based on the first grayscale value segmentation point M1, and form a set A1 and a set B1, the elements in the set A1 are grayscale values smaller than M1, and the elements in the set B1 are grayscale values larger than M1.

[0035] The calculation module II is electrically connected to the input end of the segmentation module, and is configured to calculate an nth grayscale value segmentation point M n , and the acquisition model of the nth grayscale value segmentation point M n is as follows: ;

[0036] In the formula, i is the number of elements in the set A n-1 , K i is a grayscale value smaller than the (n-1)th grayscale value segmentation point M n-1 , j is the number of elements in the set B n-1 , and K j is a grayscale value larger than the (n-1)th grayscale value segmentation point M n-1 .

[0037] Segmentation module II: the output end is electrically connected with the input end of the judgment module, and the nth gray value segmentation point M n is taken as a reference, and M n is replaced by m gray values in the eye photo, and a set A n is formed n , and a set B n is formed n , the elements in the set A n are gray values K less than M n , the elements in the set B n-1 are gray values K greater than M n , and the set An replaces the set A n , and the set B n-1 replaces Bn-1

[0038] Judgment module: the output end is electrically connected with the input end of the calculation module III and the input end of the calculation module II, and a gray value threshold is set in the judgment module , if M n -M n-1 ≥ , the calculation module II is executed, and the segmentation module II is executed, if M n -M n-1 < , the calculation module III is executed

[0039] Calculation module III: the output end is electrically connected with the input end of the output module, and is used for calculating the area S of the meibomian gland, and the calculation model of the area S of the meibomian gland is as follows:

[0040] In the formula, X is the width of the eye photo, Y is the length of the eye photo, m is the total number of gray values, and Z is the number of gray values of the meibomian gland region

[0041] The calculation model of the number Z of gray values of the meibomian gland region is as follows:

[0042]

[0043]

[0044] Output module: output the processed eye photo and the numerical value of the area S of the meibomian gland.

[0045] Optionally, the calculation module III is further provided with a calculation module IV

[0046] Calculation module IV: the input end is electrically connected with the output end of the calculation module III, the output end is electrically connected with the input end of the output module, and is used for calculating the missing rate V of the meibomian gland, and the calculation model of the missing rate V of the meibomian gland is as follows: ; wherein S1 is the area of a standard meibomian gland.

[0047] Optionally, the output module also outputs a meibomian gland absence rate V.

[0048] In a third aspect, the present application provides a meibomian gland absence detection device, which adopts the following technical scheme:

[0049] An apparatus comprising a processor and a memory storing a computer program, the processor configured to execute the computer program stored in the memory to cause the apparatus to perform the method of the first aspect.

[0050] In a fourth aspect, the present application provides a computer storage medium for meibomian gland absence detection, which adopts the following technical scheme:

[0051] A medium having a computer program stored thereon, the computer program being executable by a processor to implement the method of the first aspect. BRIEF DESCRIPTION OF DRAWINGS

[0052] Fig. 1 is a flowchart of the embodiment 1 of the present application;

[0053] Fig. 2 is a schematic diagram of the system structure of the embodiment 2 of the present application;

[0054] Fig. 3 is a device diagram of the embodiment 3 of the present application. BEST MODE FOR CARRYING OUT THE INVENTION

[0055] In order to make the technical problems, technical schemes and beneficial effects of the present application clearer, the present application will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and not intended to limit the present application.

[0056] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs; the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the present application. The description and the drawings of the specification are to be considered as illustrative and not restrictive in character.

[0057] It should be noted that when an element is referred to as being "fixed" or "set" on another element, it can be directly on the other element or indirectly on the other element. When an element is referred to as being "connected" to another element, it can be directly connected to the other element or indirectly connected to the other element.

[0058] It should be understood that the terms “length”, “width”, “upper”, “lower”, “front”, “back”, “left”, “right”, “vertical”, “horizontal”, “top”, “bottom”, “inner”, “outer” and the like indicate directions or positions based on the directions or positions shown in the drawings, and are used only for the purpose of facilitating the description of the present application and simplifying the description, and do not indicate or imply that the device or element referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as limiting the present application.

[0059] In addition, the terms “first”, “second” are only for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the technical features indicated. Therefore, the features defined with “first”, “second” can explicitly or implicitly include one or more of the features. In the description of the present application, the meaning of “multiple” is two or more, unless otherwise explicitly and specifically limited.

[0060] In the present application, “and / or” is only a description of the association relationship of the associated objects, which means that there can be three relationships; for example, A and / or B can represent the three cases of A alone, A and B together, and B alone. In addition, the character “ / ” in this paper generally represents that the front and rear associated objects are a “or” relationship.

[0061] It should be noted that in the present application, the words “in some embodiments”, “exemplarily”, “for example” and the like are used to represent as an example, illustration or description. Any embodiment or design scheme described as “in some embodiments”, “exemplarily”, “for example” in the present application should not be interpreted as more preferred or more advantageous than other embodiments or design schemes. Rather, the use of “in some embodiments”, “exemplarily”, “for example” and the like is intended to present the relevant concept in a specific way, meaning that the specific features, structures or properties described in conjunction with the embodiments can be included in at least one embodiment of the present application. The appearance of the above words at various places in the specification does not necessarily mean the same embodiment, nor is it an independent or alternative embodiment that is not mutually exclusive with other embodiments. Those skilled in the art explicitly and implicitly understand that the embodiments described herein can be combined with other embodiments.

[0062] The present application will be further described in detail below in conjunction with FIGS. 1-2.

[0063] Embodiment 1: The present embodiment discloses a meibomian gland loss detection method, referring to FIG. 1, a meibomian gland loss detection method comprises the following steps:

[0064] S1: image acquisition: open the upper and lower eyelids, take a photo of the eye, and obtain an eye photo.

[0065] S2: image processing: the eye photo is processed in gray scale, so that each pixel in the eye photo corresponds to a gray scale value K, and the total number of gray scale values m, the maximum gray scale value Nmax and the minimum gray scale value Nmin are obtained. (For example: the number of gray scale values obtained in the eye photo is 10, which are 200, 300, 800, 700, 450, 650, 750, 430, 550, 610, the maximum gray scale value is 800, and the minimum gray scale value is 200).

[0066] S3: first gray scale value segmentation point acquisition: the first gray scale value segmentation point M1 is obtained, and the acquisition model of the first gray scale value segmentation point M1 is as follows: (For example: by the maximum gray scale value of 800 and the minimum gray scale value of 200, M1 is calculated as 500).

[0067] S4: first image segmentation: taking the first gray scale value segmentation point M1 as the reference, the m gray scale values in the eye photo are divided, and set A1 and set B1 are formed, the elements in set A1 are gray scale values K less than M1, and the elements in set B1 are gray scale values K greater than M1. (For example: taking the gray scale value M1=500 as the segmentation point, the points less than 500 gray scale value form the A1 set, and the points greater than 500 gray scale value form the B1 set).

[0068] S5: nth gray scale value segmentation point acquisition: the nth gray scale value segmentation point M n is obtained, and the acquisition model of the nth gray scale value segmentation point M n is as follows: ;

[0069] In the formula, i is the number of elements in set A n-1 , K i is the gray scale value less than the (n-1)th gray scale value segmentation point M n-1 , j is the number of elements in set B n-1 , and K j is the gray scale value greater than the (n-1)th gray scale value segmentation point M n-1 . (For example: the elements in set A2 are 200, 300, 450, 430, the elements in set B2 are 800, 700, 650, 750, 550, 610, and M2 is 511).

[0070] S6: nth image segmentation: taking the nth gray scale value segmentation point M n as the reference and replacing Mn-1 with M n , the m gray scale values in the eye photo are divided, and set A n and set B n are formed, the elements in set A n are gray scale values K less than Mn gray values, set B n gray values K greater than M n gray values, and set An replaces set A n-1 gray values, set B n replaces Bn-1. (For example: after M2 is calculated, the M gray values K are divided by M2 as the division point, and then M3 is calculated through step S5; after M3 is calculated, the M gray values K are divided by M3 as the division point, and then M4 is calculated through step S5; and so on, until M n-1 and M n are calculated).

[0071] S7: Gray value judgment: set a gray value threshold , if M n -M n-1 ≥ , the nth gray value division point acquisition step is executed, and the nth image division step is executed, if M n -M n-1 < , the area calculation step is executed. (For example: the gray value threshold is 10, through the loop step S5, M4=491, M5=498, M5-M4=7<10, then the gray values greater than M5 are all the gray values of the Meibomian glands after the eye photo is processed, and the area calculation step is performed, if M4=491, M5=520, M5-M4=29>10, then the loop step S5 and step S6 are continued).

[0072] S8: Area calculation: calculate the area S of the Meibomian glands, and the calculation model of the Meibomian gland area S is as follows: ;

[0073] In the formula, X is the width of the eye photo, Y is the length of the eye photo, m is the total number of gray values, and Z is the number of gray values of the Meibomian gland region;

[0074] The calculation model of the number of gray values Z of the Meibomian gland region is as follows: :

[0075] .

[0076] S9: Loss rate calculation: calculate the loss rate V of the Meibomian glands, and the calculation model of the Meibomian gland loss rate V is as follows: ; In the formula, S1 is the area of the standard Meibomian glands.

[0077] S10: Output: output the processed eye photo, the numerical value of the area S of the Meibomian glands, and the loss rate V of the Meibomian glands.

[0078] The implementation principle of the meibomian gland absence detection method of the embodiment is as follows: when the meibomian glands are detected, the eyelids of the person to be detected are opened, the eyes are photographed, and the eye photo is subjected to gray scale processing, so that each pixel in the eye photo corresponds to a gray scale value K, and the total number of gray scale values m, the maximum gray scale value Nmax, and the minimum gray scale value Nmin are obtained, the first degree value division point M1 is calculated according to the maximum gray scale value Nmax and the minimum gray scale value Nmin, the m gray scale values in the eye photo are divided according to the M1 as the reference, and the set A1 and the set B1 are formed, the elements in the set A1 are the gray scale values smaller than M1, the elements in the set B1 are the gray scale values greater than M1, the value of M n - n-1 ≥ when M n n is greater than Mn, it is proved that the area of the meibomian glands has been separated from the eye photo, and the area of the meibomian glands in the eye image is obtained by obtaining the proportion of the number of gray scale values greater than Mn in the total number of gray scale values.

[0079] Embodiment 2: The embodiment discloses a meibomian gland absence detection system, referring to FIG. 2, a meibomian gland absence detection system comprises: an acquisition module, a processing module, a calculation module I, a division module I, a calculation module II, a division module II, a judgment module, a calculation module III, a calculation module IV, and an output module.

[0080] The acquisition module: the output end is electrically connected to the input end of the processing module, and is used for photographing the eyes and obtaining the eye photo.

[0081] The processing module: the input end is electrically connected to the output end of the acquisition module, and the output end is electrically connected to the input end of the calculation module I, and is used for performing gray scale processing on the eye photo obtained by the acquisition module, so that each pixel in the eye photo corresponds to a gray scale value K, and the total number of gray scale values m, the maximum gray scale value Nmax, and the minimum gray scale value Nmin are obtained.

[0082] The calculation module I: the output end is electrically connected to the input end of the division module I, and is used for calculating the first gray scale value division point M1, and the acquisition model of the first gray scale value division point M1 is as follows: ;

[0083] The division module I: the output end is electrically connected to the input end of the calculation module II, and the first gray scale value division point M1 is used as the reference to divide the m gray scale values in the eye photo, and the set A1 and the set B1 are formed, the elements in the set A1 are the gray scale values smaller than M1, and the elements in the set B1 are the gray scale values greater than M1.

[0084] The output of the calculation module II is connected with the input of the segmentation module, and the calculation module II is used to calculate the nth gray value segmentation point M n The acquisition model of the nth gray value segmentation point M n is as follows: ;

[0085] In the formula, i is the number of elements in the set A n-1 , K i is the gray value less than the (n-1)th gray value segmentation point M n-1 , j is the number of elements in the set B n-1 , and K j is the gray value greater than the (n-1)th gray value segmentation point M n-1 .

[0086] The output of the segmentation module II is connected with the input of the judgment module, and the nth gray value segmentation point M n is used as the reference, M n is replaced by Mn-1, the m gray values in the eye photo are distinguished, and the set A n and the set B n are formed, the elements in the set A n are the gray values less than M n , the elements in the set B n are the gray values greater than M n , and the set An is used to replace the set A n-1 , and the set B n is replaced by Bn-1.

[0087] The output of the judgment module is connected with the input of the calculation module III and the input of the calculation module II, and the gray value threshold is set in the judgment module, if M n -M n-1 ≥ , the calculation module II is calculated, and the segmentation module II is executed, if M n -M n-1 < , the calculation module III is executed.

[0088] The output of the calculation module III is connected with the input of the output module, and the calculation module III is used to calculate the meibomian gland area S, and the calculation model of the meibomian gland area S is as follows: ;

[0089] In the formula, X is the width of the eye photo, Y is the length of the eye photo, m is the total number of gray values, and Z is the number of gray values in the meibomian gland area.

[0090] The calculation model of the number of gray values Z in the meibomian gland area is as follows: :

[0091] 。

[0092] Computing module IV: the input end is electrically connected with the output end of the computing module III, and the output end is electrically connected with the input end of the output module, and is used for calculating the loss rate V of the meibomian glands. The calculation model of the loss rate V of the meibomian glands is as follows: ;In the formula, S1 is the area of the standard meibomian glands.

[0093] Output module: output the processed eye photo and the numerical value of the area S of the meibomian glands.

[0094] The implementation principle of the meibomian gland loss detection system in the embodiment is as follows: the acquisition module acquires an eye photo, the processing module performs gray scale processing on the eye photo acquired by the acquisition module, so that each pixel in the eye photo corresponds to a gray scale value K, and the total number m of gray scale values, the maximum gray scale value Nmax and the minimum gray scale value Nmin are acquired; the computing module I calculates the first gray scale value division point M1, the division module I divides the m gray scale values in the eye photo based on the first gray scale value division point M1, and forms a set A1 and a set B1. The elements in the set A1 are the gray scale values whose gray scale values K are less than M1, and the elements in the set B1 are the gray scale values whose gray scale values K are greater than M1. The computing module II calculates the nth gray scale value division point M n , the division module II divides the m gray scale values in the eye photo based on the nth gray scale value division point M n , and replaces M n with Mn-1. The m gray scale values in the eye photo are divided, and a set A n and a set B n are formed. The elements in the set A n are the gray scale values whose gray scale values K are less than M n , the elements in the set B n are the gray scale values whose gray scale values K are greater than M n , the set An replaces the set A n-1 , and the set B n replaces Bn-1. The gray scale value threshold is set in the judgment module. If M n -M n-1 ≥ , the computing module II is calculated, and the division module II is executed. If M n -M n-1 < , the computing module III is executed. The computing module III calculates the area S of the meibomian glands, the computing module IV calculates the loss rate V of the meibomian glands, and the output module outputs the processed eye photo, the numerical value of the area S of the meibomian glands and the loss rate V of the meibomian glands.

[0095] Embodiment 3: This embodiment discloses a Meibomian gland loss detection device, referring to FIG. 3, a Meibomian gland loss detection device comprises:

[0096] a storage for storing a computer program;

[0097] a processor for executing the computer program stored in the storage, thereby realizing the method described in Embodiment 1.

[0098] The storage can include a mass storage for storing data or instructions. By way of example and not limitation, the storage can include a hard disk, a floppy disk, a flash memory, an optical disk, a magneto-optical disk, a magnetic tape, or a combination of two or more of these. The storage can be removable or non-removable (or fixed) as appropriate. The storage can be internal or external to the data processing device as appropriate. In a particular embodiment, the storage is a non-volatile solid-state storage. In a particular embodiment, the storage includes a read-only memory (ROM). The ROM can be a mask-programmed ROM, a programmable ROM (PROM), an erasable PROM (EPROM), an electrically erasable PROM (EEPROM), an electrically alterable ROM (EBROM), or a combination of two or more of these as appropriate.

[0099] Embodiment 4: This embodiment discloses a computer storage medium for Meibomian gland loss detection, wherein the computer storage medium stores a program, and the program can realize part or all steps of the method described in Embodiment 1 when executed.

[0100] The above are preferred embodiments of the present application, and do not limit the protection scope of the present application, so: any equivalent changes made according to the structure, shape, principle of the present application should be covered within the protection scope of the present application.

Claims

1. A method for detecting meibomian gland absence, characterized in that: Image acquisition: open the upper and lower eyelids, take a photo of the eye, and obtain the eye photo; Image processing: perform grayscale processing on the eye photo, so that each pixel in the eye photo corresponds to a grayscale value K, and obtain the total number of grayscale values m, the maximum grayscale value Nmax, and the minimum grayscale value Nmin; First grayscale value segmentation point acquisition: obtain the first grayscale value segmentation point M1, and the acquisition model of the first grayscale value segmentation point M1 is as follows: ； First image segmentation: taking the first grayscale value segmentation point M1 as the reference, the m grayscale values in the eye photo are distinguished, and sets A1 and B1 are formed, the elements in set A1 are grayscale values less than M1, and the elements in set B1 are grayscale values greater than M1; n-th gray value division point acquisition: acquire an n-th gray value division point M n , the n-th gray value division point M n is acquired according to the following model: ； where i is the number of elements in set A n-1 K is the number of elements in set B i M is the (n-1)th gray value division point n-1 M is the (n-1)th gray value division point n-1 K is the number of elements in set B j M is the (n-1)th gray value division point n-1 M is the (n-1)th gray value division point n-th image segmentation: segmenting point M with the n-th gray value n as a reference and making M n Replace Mn-1, distinguish m gray values in the eye photo, and form set A n and set B n , the elements in set A n are gray values K less than M n , the elements in set B n are gray values K greater than M n , and make set An replace set A n-1 , set B n replace B n-1 ; Gray value determination: Set a gray value threshold , if M n - M n-1 ≥ then the nth gray value division point obtaining step is executed, and the nth image division step is executed, if M n -M n-1 < Then execute the area calculation step; Area calculation: calculate the area S of the meibomian gland, and the calculation model of the meibomian gland area S is as follows: ； In the formula, X is the width of the eye photo, Y is the length of the eye photo, m is the total number of grayscale values, and Z is the number of grayscale values in the meibomian gland area. The calculation model of the number of grayscale values Z in the meibomian gland area is as follows: ： ； Output: output the processed eye photo and the numerical value of the area S of the meibomian gland.

2. The method of Meibomian gland loss detection according to claim 1, wherein: The area calculation step is further provided with a missing rate calculation step; Missing rate calculation: calculate the missing rate V of the meibomian gland, and the calculation model of the meibomian gland missing rate V is as follows: In the formula, S1 is the area of the standard meibomian gland.

3. The method of Meibomian gland loss detection according to claim 2, wherein: The output step also outputs the missing rate V of the meibomian gland.

4. A meibomian gland loss detection system applying the method of any one of claims 1-3, characterized in that: It comprises an acquisition module, a processing module, a calculation module I, a segmentation module I, a calculation module II, a segmentation module II, a judgment module, a calculation module III, and an output module. The acquisition module: the output end is electrically connected with the input end of the processing module, and is used for taking a photo of the eye and obtaining an eye photo; The processing module: the input end is electrically connected with the output end of the acquisition module, and the output end is electrically connected with the input end of the calculation module I, which is used for performing grayscale processing on the eye photo obtained by the acquisition module, so that each pixel in the eye photo corresponds to a grayscale value K, and the total number of grayscale values m, the maximum grayscale value Nmax and the minimum grayscale value Nmin are obtained; The calculation module I: the output end is electrically connected with the input end of the segmentation module I, and is used for calculating the first grayscale value segmentation point M1, and the acquisition model of the first grayscale value segmentation point M1 is as follows: ； The segmentation module I: the output end is electrically connected with the input end of the calculation module II, and taking the first grayscale value segmentation point M1 as the reference, the m grayscale values in the eye photo are distinguished, and sets A1 and B1 are formed, the elements in set A1 are grayscale values less than M1, and the elements in set B1 are grayscale values greater than M1; The computing module II is electrically connected with the input end of the dividing module, and is used for calculating the nth gray value dividing point M. n The acquisition model of the nth gray value dividing point M n is as follows: ； where i is the number of elements in set A n-1 K is the number of elements in set B i M is the (n-1)th gray value division point n-1 M is the (n-1)th gray value division point n-1 K is the number of elements in set B j M is the (n-1)th gray value division point n-1 M is the (n-1)th gray value division point Segmentation module II: output connected to input of the judging module with the n-th gray value segmentation point M n For reference, and M n Instead of Mn-1, m gray values in the eye photo are distinguished and form the set A n And the set B n The elements in the set A n The gray value K is less than M n The elements in the set B n The gray value K is greater than M n And the set An replaces the set A n-1 The set B n Replaces Bn-1; a judging module: the output end is electrically connected with the input end of the calculating module III and the input end of the calculating module II, and a gray value threshold is set in the judging module , if M n - M n-1 ≥ then compute module II and perform segmentation module II if M n -M n-1 < Then execute the calculation module III; The calculation module III: the output end is electrically connected with the input end of the output module, and is used for calculating the area S of the meibomian gland, and the calculation model of the meibomian gland area S is as follows: ； In the formula, X is the width of the eye photo, Y is the length of the eye photo, m is the total number of grayscale values, and Z is the number of grayscale values in the meibomian gland area. The calculation model of the number of grayscale values Z in the meibomian gland area is as follows: ： ； The output module outputs the processed eye photo and the numerical value of the meibomian gland area S.

5. The Meibomian Gland Loss Detection System of claim 4, wherein: The calculation module III is further connected with a calculation module IV; The calculation module IV is electrically connected with the output end of the calculation module III and the input end of the output module, and is used for calculating the meibomian gland loss rate V, and the calculation model of the meibomian gland loss rate V is as follows: In the formula, S1 is the area of a standard meibomian gland.

6. A meibomian gland loss detection system according to claim 5, wherein: The output module further outputs the meibomian gland loss rate V.

7. A meibomian gland loss detection device characterized by: The device comprises a processor and a storage, the storage is used for storing a computer program, and the processor is used for executing the computer program stored in the storage, so that the device executes the method in any one of claims 1-3.

8. A computer storage medium of meibomian gland loss detection, having stored thereon a computer program, characterized in that: The computer program is executed by the processor to implement the method in any one of claims 1-3.

Images