Human eye positioning device and human eye positioning method for confocal laser ophthalmoscope

Abstract

The present invention discloses a human eye positioning device for a confocal laser ophthalmoscope and a human eye positioning method thereof. The human eye positioning device for a confocal laser ophthalmoscope comprises: a three-dimensional motion platform and a forehead support arranged on a base, an eyepiece and a refractive component corresponding to the eyepiece are arranged on the three-dimensional motion platform; the eyepiece is located between the refractive component and the forehead support, and the eyepiece is installed on one side of the refractive component through a lens barrel; the eyepiece is used to refract light into the human eye; the three-dimensional motion platform is used to drive the relative displacement of the refractive component and the eyepiece relative to the object to be detected supported by the forehead support; a refractive motor is arranged in the refractive component, the refractive motor is connected to an optical path component, and the optical path component is used in conjunction with the eyepiece. The present invention discloses a human eye positioning device for a confocal laser ophthalmoscope and a human eye positioning method thereof. The confocal device that can be used for fundus imaging can also image the cornea and pupil of the human eye.

Description

Technical Field

[0001] The invention relates to the technical field of optical equipment, and in particular to a human eye positioning device and a human eye positioning method for a confocal laser ophthalmoscope. Background Art

[0002] As the shooting range of the laser confocal fundus camera continues to expand, the aperture of the eyepiece and the distance from the eyepiece to the human eye will become very close. At the same time, the current manual shooting method requires the operator to manually adjust the height, left and right, front and back position and focus of the equipment to capture the fundus image of the person being collected. During the entire process, the equipment needs to be manually adjusted first, and the operator manually adjusts the eyepiece of the equipment to the height of the person being photographed, roughly overlapping the left and right, and then adjusts the front and back position of the equipment. The operator visually observes the size of the fundus image, and then manually adjusts the refraction until the image reaches a certain clarity, and then shoots. The entire shooting process requires manual intervention by the operator. The entire shooting process takes a long time and has high requirements on the proficiency of the operator. At the same time, because the shooting time is long, the person being photographed needs to have a high degree of cooperation. At the same time, keeping the eyes open for a long time is not friendly to eye detection. Summary of the invention

[0003] The present invention overcomes the shortcomings of the prior art and provides a human eye positioning device and a human eye positioning method for a confocal laser ophthalmoscope; the confocal device that can be used for fundus imaging can also image the human cornea and pupil.

[0004] To achieve the above-mentioned purpose, the technical solution adopted by the present invention is: a human eye positioning device for a confocal laser ophthalmoscope, comprising a three-dimensional motion platform and a forehead rest arranged on a base, an eyepiece and a refractive assembly corresponding to the eyepiece are arranged on the three-dimensional motion platform; the eyepiece is located between the refractive assembly and the forehead rest, and the eyepiece is installed on one side of the refractive assembly through a lens barrel; the eyepiece is used to refract light into the human eye; the three-dimensional motion platform is used to drive the relative displacement of the refractive assembly and the eyepiece relative to the object to be detected supported by the forehead rest; a refractive motor is arranged in the refractive assembly, the refractive motor is connected to an optical path assembly, the optical path assembly is used in conjunction with the eyepiece, the refractive assembly is used for refractive compensation when photographing the fundus image of the human eye, and realizes fundus image focusing; the three-dimensional motion platform and the refractive assembly are both interconnected with the confocal laser system of the confocal laser ophthalmoscope, the confocal laser system is used for fundus and / or ocular surface image acquisition, and the refractive assembly realizes refractive compensation through a refractive compensation mechanism.

[0005] In a preferred embodiment of the present invention, infrared cameras are arranged on the left and right sides of the eyepiece, and an illuminating lamp is arranged on the outside of the infrared camera; the infrared camera is used to collect images of the cornea and pupil of the human eye; the illuminating lamp is used to fill in the light of the object to be detected during image collection; and the lens barrel is used to assemble and fix the infrared camera and the illuminating lamp to the eyepiece.

[0006] In a preferred embodiment of the present invention, an infrared camera is provided on the upper side of the eyepiece and / or on the lower side of the eyepiece; and an illumination lamp is provided on one side of the infrared camera; the infrared camera is used to collect images of the cornea and pupil of the human eye; and the illumination lamp is used to fill in the light of the object to be detected during image collection.

[0007] In a preferred embodiment of the present invention, a method for locating human eyes for a human eye locating device of a confocal laser ophthalmoscope comprises the following steps:

[0008] Step 1, initial calibration of the test position, including calibrating and adjusting the relative position relationship between the refractive component and the human eye and the eyepiece in height, left-right, front-back direction; the refractive component and the human eye are coaxial with the eyepiece in height, left-right, front-back direction, and the fundus retinal image acquired by the confocal laser ophthalmoscope fills the preview window of the confocal laser ophthalmoscope;

[0009] Step 2, after completing the initial calibration of the test position in step 1, obtaining a grayscale image containing human pupil information;

[0010] Step 3, performing image processing on the grayscale image containing the pupil information of the human eye obtained in step 2 to obtain the central pixel coordinates of the pupil of the human eye;

[0011] Step 4, compare the central pixel coordinates of the human eye pupil obtained in step 3 with the central position of the template image, and control the refractive motor to automatically locate the human eye pupil after unit conversion; make the fundus retinal image in the preview window of the confocal laser ophthalmoscope the clearest, and complete the focusing process of the retina of the human eye.

[0012] In a preferred embodiment of the present invention, in step 1, the initial calibration of the test position includes the following steps:

[0013] The positioning of the refractive component, the human eye and the eyepiece in height and left-right directions comprises the following steps: positioning the refractive component, the human eye and the eyepiece in height and left-right directions by using the positions of the human cornea or the front surface image of the human eye and the pupil in the image acquired by the confocal laser system of the laser confocal ophthalmoscope or the image acquired by the infrared camera; making the refractive component and the human eye coaxial with the eyepiece in height and left-right directions;

[0014] The positioning of the front-to-back distance between the refractive component and the human eye and the eyepiece includes the following steps: adjusting the distance position of the refractive component and the human eye from the eyepiece in the front-to-back direction, observing the size of the human eye fundus image in the preview window of the fundus confocal ophthalmoscope until the fundus retinal image completely fills the preview window, at which time the human eye and the eyepiece are coaxial in three dimensions: height, left and right, and front and back.

[0015] In a preferred embodiment of the present invention, the positioning of the refractive assembly, the human eye and the eyepiece in height and left and right directions includes the following steps:

[0016] Based on the positions of the human cornea and pupil in the image field obtained by the confocal laser system, the three-dimensional motion stage is driven to adjust the position of the refractive component and the eyepiece so that the height and left-right direction of the human eye and the eyepiece are in the coaxial position;

[0017] Or, in the coaxial adjustment method, the cornea and pupil of the human eye are positioned in height and left-right directions by using an infrared camera, comprising the following steps: when a pair of infrared cameras and a pair of lighting lamps are arranged on the left and right sides of the eyepiece, the lighting lamps on the left and right sides illuminate the front surface of the human eye so that the infrared cameras on the left and right sides of the eyepiece can capture the image of the front surface of the human eye of the human eye; and the two left and right infrared cameras simultaneously take pictures of the human eye, observe the positions of the human eye surface and the pupil in the field of view of the left and right infrared cameras, determine the relative positions of the human eye and the device up and down, left and right, front and back, and then drive the three-dimensional motion stage to adjust the relative positions between the refractive component and the eyepiece and the human eye, until the coordinate position of the image of the human eye captured by the infrared cameras on the left and right sides coincides with the preset position, and the preset position is located at the center of the human eye in the template image;

[0018] Alternatively, the coaxial adjustment method uses an infrared camera to position the cornea and pupil of the human eye in height and left and right directions, including the following steps: when the infrared camera and the lighting lamp are arranged at the upper end of the eyepiece, the lighting lamp provides lighting for the infrared camera at the top of the eyepiece to increase the brightness of the image of the human eye taken by the infrared camera, and the infrared camera is adjusted by a pre-three-dimensional motion stage so that the reference point of the image of the infrared camera coincides with the optical axis of the eyepiece; when photographing and collecting the human eye, by judging the position relationship between the position of the cornea and pupil of the human eye in the infrared camera and the position of the reference point, the three-dimensional motion stage is driven to adjust the eyepiece and the refractive assembly so that the cornea and pupil of the human eye coincide with the reference point of the image of the infrared camera, and at this time, the refractive assembly and the human eye are coaxial with the eyepiece in height and left and right directions.

[0019] In a preferred embodiment of the present invention, adjusting the distance between the refractive assembly and the human eye and the eyepiece in the front-to-back direction comprises the following steps:

[0020] After the refractive assembly and the eyepiece are adjusted to be coaxial with the height and left-right direction of the human eye, the three-dimensional motion stage is driven to move the refractive assembly and the eyepiece to a position close to the human eye. At this time, the size of the image of the human eye retinal image in the preview window of the fundus confocal ophthalmoscope is observed until the fundus retinal image completely fills the preview window. At this time, the front-to-back distance adjustment between the eyepiece and the human eye is completed. At this time, the human eye and the eyepiece are coaxial in three dimensions: height, left-right, and front-to-back.

[0021] Alternatively, after adjusting the refractive assembly and the eyepiece to be coaxial with the height and left-right direction of the human eye, adjust the front-to-back position of the eyepiece according to the position of the human eye in the two infrared cameras, and judge the distance between the device and the human eye based on the proportion of the preview window occupied by the image by observing the size of the fundus retinal image in the preview window of the fundus confocal ophthalmoscope interface, until the fundus retinal image completely fills the preview window, so that the human eye and the eyepiece completely overlap in three dimensions of height, left-right, and front-to-back.

[0022] In a preferred embodiment of the present invention, in step 2, the template image is obtained by obtaining an ocular surface image of a human eye through a confocal laser system of a confocal laser ophthalmoscope, and the obtained ocular surface image includes a grayscale image of pupil information of the human eye;

[0023] The method for obtaining the coordinates of the center of the pupil of the human eye in step 3 includes the following steps:

[0024] Step L3-A1, filtering and denoising, adaptive binarization and boundary removal processing are performed on the grayscale image to obtain a binary image containing human pupil information;

[0025] Step L3-A2, screening and contour extraction of the largest pixel block in the binary image of step L3-A1, to obtain a binary image containing only the pupil boundary of the human eye;

[0026] Step L3-A3, performing least square fitting on the pixel points at the boundary of the pupil of the human eye extracted from the binary image in step L3-A2 to obtain the central pixel coordinates of the pupil of the human eye;

[0027] Step L3-A1 includes the following steps:

[0028] Step La1: Filter the grayscale image obtained in step 2 through an improved mean filter operator, where the improved operator kernel is for:

[0029] ;

[0030] Where I is the processed image, w and h are the width and height of the operator kernel respectively; after filtering, the denoised human pupil grayscale image is obtained;

[0031] Step La2: Binarize the grayscale image obtained in step La1 using an improved binarization algorithm. The calculation formula is:

[0032] ;

[0033] Among them, s is the grayscale value of the image, which is [0,255]. Represents the inverse function of the total correlation of the image grayscale probability distribution, where The calculation method is:

[0034] ;

[0035] in, yes A probability distribution function of is the total probability that the gray value in an image reaches s; m is an integer, ; After binarization, a binary image containing human pupil information is obtained;

[0036] Step La3: The binary image obtained in step La2 is subjected to a boundary removal operator to delete the signal blocks at the image boundary, thereby obtaining a binary image of the pupil of the human eye after the boundary is removed;

[0037] Step L3-A2 includes the following steps:

[0038] Step Lb1: extract the pixel block with the largest number of pixels from the binary image obtained in step La3 to obtain a binary image containing only pupil information of the human eye;

[0039] Step Lb2: extracting the boundary pixel points of the pupil of the human eye from the binary image obtained in step Lb1 through the CHAIN_APPROX_SIMPLE operator to obtain a binary image containing only the boundary of the pupil of the human eye;

[0040] Step L3-A3 includes the following steps:

[0041] Step Lc1: The pixel points obtained in step Lb2 are converted into T X) -1 X T Y is fitted into a circle, and the pupil center coordinates are the center of the fitted circle , where P is the parameter information of the fitted circle, X and Y are the row vectors constructed by the horizontal and vertical coordinates of the pixel points, respectively. T is the transpose of the row vector X;

[0042] Step Lc2: Draw a fitting line according to the parameters of the fitting circle obtained in step Lc1 and map it to the original image;

[0043] The automatic positioning method in the method for automatically positioning the pupil of the human eye in step 4 comprises the following steps:

[0044] Step L4-A1, calculating the horizontal and vertical pixel differences between the pixel coordinates of the pupil center and the image center, and obtaining the horizontal and vertical pixel offsets of the pupil center;

[0045] Step L4-A2, converting the horizontal and vertical pixel offsets of step L4-A1 into the movement of the motor, and controlling the motor to complete the automatic positioning of the pupil of the human eye;

[0046] Automatic positioning of the pupil of the human eye includes the following steps:

[0047] Step Ld1: The horizontal coordinate of the pupil center pixel obtained in step Lc1 is and pixel ordinate The horizontal coordinate of the image center and the vertical coordinate Subtract and get the horizontal pixel offset and vertical pixel offset ,in , ;

[0048] Step Ld2: The horizontal pixel offset obtained in step Ld1 and vertical pixel offset By formula The actual movement of the refractive motor is calculated, so as to control the movement of the refractive motor to complete the automatic positioning of the pupil of the human eye. is the pixel size of the image, The horizontal or vertical pixel offset.

[0049] In a preferred embodiment of the present invention, when an infrared camera and an illumination lamp used in conjunction with the infrared camera are provided on the left and right sides of the eyepiece of the confocal laser ophthalmoscope;

[0050] In step 2, two grayscale images containing pupil information of human eyes are obtained;

[0051] In step 3, the two grayscale images obtained in step 2 are processed simultaneously to obtain the central pixel coordinates of the pupils of the two human eye images respectively;

[0052] In step 4, the two pixel coordinates obtained in step 3 are compared with the center position of the template image, and after unit conversion, the refractive motor is controlled to complete the automatic positioning of the pupil of the human eye.

[0053] In a preferred embodiment of the present invention, the method for obtaining the template image in step 4 includes the following steps:

[0054] The two grayscale images acquired by the infrared cameras set on the left and right sides are stitched up and down to construct a template image;

[0055] The method for obtaining the coordinates of the center of the pupil of the human eye in step 3 includes the following steps:

[0056] Step S3-A1, filtering and denoising the grayscale image, and adaptively binarizing the image to obtain a binary image containing pupil information of the human eye; comprising the following steps:

[0057] Step Sa1: Filter the grayscale image obtained in step 3 through a Gaussian filter operator, where the Gaussian filter formula is:

[0058] ;

[0059] in, is the standard deviation of the Gaussian filter, (x0, y0) is the center coordinate of the Gaussian filter, and (x, y) is the coordinate of the rest of the Gaussian filter except the center coordinate;

[0060] Step Sa2: binarizing the image obtained in step Sa1 using a binarization algorithm to obtain a binary image containing pupil information of the human eye;

[0061] Step S3-A2, filtering pixel blocks containing pupil information in the binary image, and extracting and sampling the edges of the filtered pixel blocks to obtain a pixel point group to be fitted; comprising the following steps:

[0062] Step Sb1: fill all pixel blocks in the binary image Sa2 to obtain a filled binary image;

[0063] Step Sb2: Delete the pixel blocks whose pixel number is less than the set threshold in the binary image obtained by Sb1, and remove the binary image of interference information;

[0064] Step Sb3: Delete the pixel blocks at the image boundary in the binary image obtained by Sb2 to obtain a binary image containing only the pupil information of the human eye;

[0065] Step Sb4: extracting the edge of the pupil of the human eye by using the Canny edge detection operator for the pupil information of the human eye in the binary image obtained in Sb3, and obtaining a binary image of the edge of the pupil of the human eye;

[0066] Step Sb5: performing forward gradient sampling in a clockwise direction on the edge of the pupil of the human eye obtained in Sb4 to remove pixel points that do not belong to the edge of the pupil of the human eye, and obtaining a filtered pixel point group of the edge of the pupil of the human eye;

[0067] Step S3-A3, performing least square fitting on the pixel point group in step S3-A2 to obtain the central pixel coordinates of the pupil of the human eye; comprising the following steps:

[0068] Step Sc1: Fit the pixel point group obtained in step Sb5 into an ellipse, and obtain the pupil center coordinate as the center of the fitted ellipse;

[0069] Step Sc2: Draw a fitting line according to the parameters of the fitting circle obtained in step Sc1 and map it to the original image;

[0070] The method for automatically locating the pupil of the human eye in step 4 includes the following steps:

[0071] Step S4-A1, respectively calculating the horizontal and vertical pixel differences between the pixel coordinates of the two pupil centers and the center of the template image to obtain the horizontal and vertical pixel offsets of the pupil centers;

[0072] Step S4-A2, converting the horizontal and vertical pixel offsets in step S4-A1 into the movement of the motor, and controlling the motor to complete the automatic positioning of the pupil of the human eye;

[0073] The method for automatically locating the pupil of a human eye, step S3-A1 comprises the following steps:

[0074] Step Sd1: Control the diopter motor to move to a rough position so that the left and right iris cameras can capture images containing pupil information of the human eye;

[0075] Step Sd2: The coordinates of the pupil center of the human eye are obtained by real-time processing of the pupil images collected by the left and right iris cameras: and ;

[0076] Step Sd3: Control the refraction motor to make the template image center Respectively and Overlap, and we get: and , after the overlap, the refractive motor is controlled to stop moving, thereby completing the automatic positioning of the human eye pupil.

[0077] Compared with the prior art, the present invention has the following beneficial effects:

[0078] The invention discloses a human eye positioning device for a confocal laser ophthalmoscope and a human eye positioning method thereof; the confocal device that can be used for fundus imaging can also image the human cornea and pupil.

[0079] 1. By introducing the fully automatic human eye positioning process, the entire acquisition time is greatly reduced, the complexity of acquisition is reduced, and the operator's proficiency requirements are greatly reduced.

[0080] 2. Reduce errors and misoperations caused by subjective differences between operators. BRIEF DESCRIPTION OF THE DRAWINGS

[0081] The present invention is further described below in conjunction with the accompanying drawings and embodiments.

[0082] Figure 1 It is a schematic diagram of the overall layout of the fully automatic laser confocal fundus camera of the present invention;

[0083] Figure 2 is a refractive layout diagram of the present invention;

[0084] Figure 3 Schematic diagram of the human eye cornea pupil image and human eye positioning workflow of Embodiment 1 and Embodiment 4 of the present invention;

[0085] Figure 4 is a flow chart of an image processing algorithm of Embodiment 1 and Embodiment 4 of the present invention;

[0086] Figure 5 It is a front view schematic diagram of the layout of the left and right infrared cameras and lighting lamps of the second and fifth embodiments of the present invention. Figure 1 ;

[0087] Figure 6 Schematic diagram of the top view of the layout of the infrared cameras and lighting lamps arranged on the left and right of the second and fifth embodiments of the present invention Figure 1 ;

[0088] Figure 7 Schematic diagram of image adjustment when infrared cameras and lighting lamps are arranged on the left and right sides of the second and fifth embodiments of the present invention Figure 1 ;

[0089] Figure 8 It is a front view schematic diagram of the layout of the left and right infrared cameras and lighting lamps of the second and fifth embodiments of the present invention. Figure 2 ;

[0090] Fig. 9 Schematic diagram of the top view of the layout of the infrared cameras and lighting lamps arranged on the left and right of the second and fifth embodiments of the present invention Figure 2 ;

[0091] Fig.10 Schematic diagram of image adjustment when infrared cameras and lighting lamps are arranged on the left and right in the second and fifth embodiments of the present invention Figure 2 ;

[0092] Fig.11 is a flow chart of an image processing algorithm of Embodiment 2 and Embodiment 5 of the present invention;

[0093] Fig.12 It is a front view schematic diagram of the layout of the infrared camera and the lighting lamp arranged on the top of the third embodiment and the sixth embodiment of the present invention;

[0094] Fig.13 It is a top view schematic diagram of the layout of the infrared camera and the lighting lamp arranged on the top of the third embodiment and the sixth embodiment of the present invention;

[0095] Fig.14 It is a schematic diagram of image adjustment when an infrared camera and lighting are arranged on the top of the third embodiment and the sixth embodiment of the present invention;

[0096] Fig.15 It is a schematic diagram of the workflow of the present invention;

[0097] Among them, 1-base, 2-three-dimensional motion table, 3-refractive component, 4-housing, 5-eyepiece, 6-eyepiece barrel, 7-human eye, 8-forehead rest, 91-infrared camera, 92-illumination lamp. DETAILED DESCRIPTION

[0098] The technical solution of the present invention is described in detail below through the accompanying drawings and specific embodiments. It should be understood that the embodiments of the present invention and the specific features in the embodiments are detailed descriptions of the technical solution of the present invention, rather than limitations on the technical solution of the present invention. The embodiments of the present invention and the technical features in the embodiments may be combined with each other unless there is a conflict.

[0099] The term "and / or" is only a description of the association relationship between related objects, indicating that there can be three relationships. For example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " generally indicates that the related objects are in an "or" relationship. Embodiment 1

[0100] like Figure 1 , Figure 2 As shown, the present invention discloses a human eye positioning device for a confocal laser ophthalmoscope, comprising a three-dimensional motion platform 2 and a forehead support 8 arranged on a base 1, an eyepiece 5 and a refractive component 3 corresponding to the eyepiece 5 are arranged on the three-dimensional motion platform 2; the eyepiece 5 is located between the refractive component 3 and the forehead support 8, and the eyepiece 5 is installed on one side of the refractive component 3 through a lens barrel; the eyepiece 5 is used to refract light into the human eye 7; the three-dimensional motion platform 2 is used to drive the relative displacement of the refractive component 3 and the eyepiece 5 relative to the object to be detected supported by the forehead support 8; a refractive motor is arranged in the refractive component 3, the refractive motor is connected to an optical path component, the optical path component is used in conjunction with the eyepiece 5, and the refractive component 3 is driven and arranged on the three-dimensional motion platform 2; the refractive component 3 is used for refractive compensation when shooting the fundus image of the human eye 7 to achieve fundus image focusing; the three-dimensional motion platform 2 and the refractive component 3 are interconnected with a confocal laser system, the confocal laser system is used for fundus and / or ocular surface image acquisition, and the refractive component 3 achieves refractive compensation through a refractive compensation mechanism. Specifically, the confocal laser system in this embodiment adopts the confocal laser system in the prior art, as long as it can realize the confocal laser system for photographing the fundus, or the confocal laser system for photographing the fundus or the ocular surface. The refractive component 3 in this embodiment adopts the refractive component for collecting fundus and / or ocular surface images in the prior art, and the specific structure and model are not described in detail.

[0101] Furthermore, the overall layout of the system is as follows Figure 1:The base 1 is used to fix the device; the three-dimensional motion platform is used for the forward, backward, up and down, left and right movement of the device; the refractive component 3 is used to compensate for the fundus images of different human eyes 7 when shooting, and is mainly used for focusing the fundus images; the outer shell 4 is used to protect the device; the eyepiece 5 is used to refract light into the human eye 7; the eyepiece barrel 6 is used to fix the eyepiece 5 and the infrared camera 91 and the lighting component; the forehead rest 8 is used to fix the human head to reduce the impact of shaking during shooting on the collected image. The layout of the refractive system is as follows Figure 2 As shown: focusing is achieved by adjusting the distance from the optical path to the human eye 7 through an electric adjustment device.

[0102] like Figure 3 The center of the middle image is the center of the optical system image acquisition, and also represents the optical axis position of the optical system, wherein the bright spot is the light spot reflected by the pupil position of the human eye 7. When working: when the center of the human eye pupil coincides with the center of the image, it can be considered that the human eye pupil coincides with the optical axis of the device. Embodiment 2

[0103] like Figure 1 , Figure 2 As shown, the present invention discloses a human eye 7 positioning device for a confocal laser ophthalmoscope, comprising a three-dimensional motion platform 2 and a forehead support 8 arranged on a base 1, an eyepiece 5 and a refractive component 3 corresponding to the eyepiece 5 are arranged on the three-dimensional motion platform 2; the eyepiece 5 is located between the refractive component 3 and the forehead support 8, and the eyepiece 5 is installed on one side of the refractive component 3 through a lens barrel; the eyepiece 5 is used to refract light into the human eye 7; the three-dimensional motion platform 2 is used to drive the relative displacement of the refractive component 3 and the eyepiece 5 relative to the object to be detected supported by the forehead support 8; a refractive motor is arranged in the refractive component 3, the refractive motor is connected to an optical path component, the optical path component is used in conjunction with the eyepiece 5, and the refractive component 3 is driven and arranged on the three-dimensional motion platform 2; the refractive component 3 is used for refractive compensation when shooting the fundus image of the human eye 7 to achieve fundus image focusing; the three-dimensional motion platform 2 and the refractive component 3 are interconnected with a confocal laser system, the confocal laser system is used for fundus image acquisition, and the refractive component 3 realizes refractive compensation through a refractive compensation mechanism.

[0104] Specifically, an infrared camera 91 and a lighting lamp 92 used in conjunction with the infrared camera 91 are arranged on the left and right sides of the eyepiece 5, and a lighting lamp 92 is arranged on the outer side of the infrared camera 91; the infrared camera 91 is used to collect images of the cornea and pupil of the human eye 7; the lighting lamp 92 is used to fill in the light of the detected object during image collection; and the lens barrel is used to assemble and fix the infrared camera 91 and the lighting lamp 92 to the eyepiece 5.

[0105] like Figure 1 , Figure 2 , Figure 5-Figure 7As shown, two infrared cameras 91 are symmetrically placed on the outside of the eyepiece 5, as well as a lighting component for the auxiliary camera when taking pictures. In this embodiment, the lighting lamp 92 adopts LED for direct lighting, and LED and optical elements can also be used, although not limited to lenses, homogenizers, binary optical elements such as DOE, etc. The optical axis of the infrared camera 91 is not parallel to the optical axis of the eyepiece 5. The two infrared cameras 91 are pre-adjusted with a fixture to adjust the angle between the two infrared cameras 91 and the optical axis of the eyepiece 5, so that the pupil position of the human eye 7 when working normally is at a known position in the picture of the infrared camera 91, and this position can be used as a reference position of the left and right infrared cameras 91, and this reference position is saved in the device as a known parameter.

[0106] like Figure 5-Figure 7 As shown: for the convenience of adjustment, the angle of the infrared camera 91 will be adjusted to the point where the center of the pupil of the human eye 7 is at the center of the screen of the infrared camera 91 when the human eye 7 is working normally. At this time, the position of the infrared camera 91 is symmetrical. Of course, the infrared camera 91 can also be adjusted to a position other than the center of the infrared camera 91 when the human eye 7 is working normally. The reference point will not be at the center of the screen of the infrared camera 91, but at another position other than the center of the screen. When the normal human eye 7 is collected, the device can be adjusted to the point where the infrared camera 91 captures the center of the human eye pupil at the corresponding reference position of the infrared camera 91. Figure 7 In the two pictures on the left, the upper one shows the human eye leaning to the right relative to the preset position, and the lower one shows the state during alignment; the same is true for the two pictures on the right, the upper one shows the human eye leaning to the left relative to the preset position, and the lower one shows the state during alignment.

[0107] This embodiment can also adopt Figure 8-Figure 10 In the working mode shown, the infrared camera 91 is parallel to the optical axis of the eyepiece 5. At this time, the pupil center of the human eye 7 when working normally will not be in the image center of the infrared camera 91. By using a fixture that is coaxial with the eyepiece 5 and has the same working distance as the human eye 7, the device records the pixel positions of the fixture in the image of the infrared camera 91. When the human eye 7 is photographed normally, the height, front and back, and left and right positions of the device are adjusted so that the pupil center of the human eye coincides with the pixel position of the reference point recorded by the infrared camera 91, and the positioning of the human eye 7 is completed. Fig.10 In the two pictures on the left, the upper picture shows the human eye leaning to the right relative to the preset position, and the lower picture shows the state when it is aligned; the same is true for the two pictures on the right, the upper picture shows the human eye leaning to the left relative to the preset position, and the lower picture shows the state when it is aligned. Embodiment 3

[0108] Based on the first embodiment, Fig.12As shown, the present invention discloses a human eye positioning device for a confocal laser ophthalmoscope, comprising a three-dimensional motion platform 2 and a forehead support 8 arranged on a base 1, an eyepiece 5 and a refractive component 3 corresponding to the eyepiece 5 are arranged on the three-dimensional motion platform 2; the eyepiece 5 is located between the refractive component 3 and the forehead support 8, and the eyepiece 5 is installed on one side of the refractive component 3 through a lens barrel; the eyepiece 5 is used to refract light into the human eye 7; the three-dimensional motion platform 2 is used to drive the relative displacement of the refractive component 3 and the eyepiece 5 relative to the object to be detected supported by the forehead support 8; a refractive motor is arranged in the refractive component 3, the refractive motor is connected to an optical path component, the optical path component is used in conjunction with the eyepiece 5, and the refractive component 3 is driven and arranged on the three-dimensional motion platform 2; the refractive component 3 is used for refractive compensation when shooting the fundus image of the human eye 7 to achieve fundus image focusing; the three-dimensional motion platform 2 and the refractive component 3 are interconnected with a confocal laser system, the confocal laser system is used for fundus image acquisition, and the refractive component 3 realizes refractive compensation through a refractive compensation mechanism. Fig.14 In the figure, the two upper figures are the front view and the side view respectively; the two lower figures, the left one is when the human eye is not aligned relative to the preset position, and the right one is when it is aligned.

[0109] Specifically, an infrared camera 91 is provided on the upper side and / or the lower side of the eyepiece 5, and an illumination lamp 92 is provided on one side of the infrared camera 91; the infrared camera 91 is used to collect images of the cornea and pupil of the human eye 7; and the illumination lamp 92 is used to fill in light for the detected object during image collection.

[0110] Further, such as Fig.12 As shown, an infrared camera 91 is placed on the top of the eyepiece 5, and the infrared camera 91 is vertically parallel to the eyepiece 5. The lighting lamp 92 is not limited to the right side of the infrared camera 91, but can also be placed at any angle on the top or left side.

[0111] Specifically, the working method of this embodiment is as follows Fig.13 As shown, through a pre-set device and a preset working distance of most human eyes 7 statistics, it is ensured that the photographed human eye 7 is within the field of view of the infrared camera 91, and then the device is adjusted left and right to the horizontal center position of the human eye pupil and the image in the field of view of the infrared camera 91, so as to distinguish the left and right eyes, and then move forward and backward until the fundus retinal image of the human eye 7 appears, and then determine the movement direction of the device through the effective field of view of the fundus image, and finally complete the positioning function of the human eye 7. Embodiment 4

[0112] Based on the first embodiment, Figure 1 , Figure 2 , Figure 3 As shown, a human eye positioning method for a human eye positioning device of a confocal laser ophthalmoscope comprises the following steps:

[0113] Step 1, initial calibration of the test position, including calibrating and adjusting the relative position relationship between the refractive component 3 and the human eye 7 and the eyepiece 5 in height, left and right, and front and back directions; and the refractive component 3 and the human eye 7 are coaxial with the eyepiece 5 in height, left and right, and front and back directions, and the fundus retinal image completely fills the preview window.

[0114] Specifically, in step 1, the initial calibration of the test position includes the following steps:

[0115] The positioning of the refractive component 3, the human eye 7 and the eyepiece 5 in height and left-right directions includes the following steps: positioning the refractive component 3, the human eye 7 and the eyepiece 5 in height and left-right directions based on the positions of the human cornea or the anterior surface image of the human eye and the pupil in the image obtained by the confocal laser system of the laser confocal ophthalmoscope or the image obtained by the infrared camera 91; making the refractive component 3 and the human eye 7 coaxial with the eyepiece 5 in height and left-right directions.

[0116] The positioning of the refractive assembly 3, the human eye 7 and the eyepiece 5 in height and left and right directions includes the following steps:

[0117] Based on the positions of the human cornea and pupil in the image field acquired by the confocal laser system, the three-dimensional motion stage 2 is driven to adjust the positions of the refractive component 3 and the eyepiece 5 so that the height and left-right direction of the human eye 7 and the eyepiece 5 are in a coaxial position;

[0118] The positioning of the front-to-back distance between the refractive component 3 and the human eye 7 and the eyepiece 5 includes the following steps: adjusting the distance position of the refractive component 3 and the human eye 7 in the front-to-back direction from the eyepiece 5, observing the size of the retinal image of the human eye 7 in the preview window of the fundus confocal ophthalmoscope, until the fundus retinal image completely fills the preview window, at which time the human eye 7 and the eyepiece 5 are coaxial in three dimensions of height, left and right, and front and back.

[0119] Furthermore, after the refractive component 3 and the eyepiece 5 are adjusted to be coaxial with the height and left-right direction of the human eye 7, the three-dimensional motion platform 2 is driven to move the refractive component 3 and the eyepiece 5 to a position close to the human eye 7. At this time, the size of the retinal image of the human eye 7 in the preview window of the fundus confocal ophthalmoscope is observed until the fundus retinal image completely fills the preview window. At this time, the front-to-back distance adjustment between the eyepiece 5 and the human eye 7 is completed. At this time, the human eye 7 and the eyepiece 5 are coaxial in three dimensions of height, left-right, and front-to-back.

[0120] Specifically, when the iris positioning is completed, that is, the image of the human eye surface is acquired, the refractive motor is driven to adjust when the fundus function is photographed in the next step.

[0121] Step 2, after completing the initial calibration of the test position in step 1, obtaining a grayscale image containing human pupil information;

[0122] Step 3, performing image processing on the grayscale image containing the pupil information of the human eye obtained in step 2 to obtain the central pixel coordinates of the pupil of the human eye;

[0123] Step 4, compare the central pixel coordinates of the human eye pupil obtained in step 3 with the central position of the template image, and control the refractive motor to automatically position the human eye pupil after unit conversion; make the fundus retinal image in the preview window the clearest, and the focusing process of the retina of the human eye 7 is completed at this time.

[0124] Specifically, the template image is an image with the correct center position of the human eye taken with a standard fixture (located at the working distance of the system and at the center of the system). The working distance refers to the universal shooting distance of the standard template image, and the distance between the eyepiece and the human eye is 10 mm. In this embodiment, the standard fixture used is a confocal laser system of a confocal laser ophthalmoscope, and the grayscale image containing the pupil information of the human eye is taken with an artificial simulated eyeball as the object to be detected.

[0125] Furthermore, this embodiment Figure 3 The center of the middle image is the center of the optical system image acquisition, and also represents the optical axis position of the optical system. The bright spot is the light spot reflected by the position of the human eye pupil. When working: when the center of the human eye pupil coincides with the center of the image, it can be considered that the human eye pupil coincides with the optical axis of the device.

[0126] Furthermore, Figure 4 As shown, the method for obtaining the coordinates of the center of the pupil of the human eye in step 3 includes the following steps:

[0127] Step L3-A1, filtering and denoising, adaptive binarization and boundary removal processing are performed on the grayscale image to obtain a binary image containing human pupil information;

[0128] Step L3-A2, screening and contour extraction of the largest pixel block in the binary image of step L3-A1, to obtain a binary image containing only the pupil boundary of the human eye;

[0129] Step L3-A3, performing least square fitting on the pixel points at the boundary of the pupil of the human eye extracted from the binary image in step L3-A2 to obtain the central pixel coordinates of the pupil of the human eye;

[0130] Step L3-A1 includes the following steps:

[0131] Step La1: Filter the grayscale image obtained in step 2 through an improved mean filter operator, where the improved operator kernel is for:

[0132] ;

[0133] Where I is the processed image, w and h are the width and height of the operator kernel respectively; after filtering, the denoised human pupil grayscale image is obtained;

[0134] Step La2: Binarize the grayscale image obtained in step La1 using an improved binarization algorithm. The calculation formula is:

[0135] ;

[0136] Among them, s is the grayscale value of the image, which is [0,255]. Represents the inverse function of the total correlation of the image grayscale probability distribution, where The calculation method is:

[0137] ;

[0138] in, yes A probability distribution function of is the total probability that the gray value in an image reaches s; m is an integer, ; After binarization, a binary image containing human pupil information is obtained;

[0139] Step La3: The binary image obtained in step La2 is subjected to a boundary removal operator to delete the signal blocks at the image boundary, thereby obtaining a binary image of the pupil of the human eye after the boundary is removed;

[0140] Step L3-A2 includes the following steps:

[0141] Step Lb1: extract the pixel block with the largest number of pixels from the binary image obtained in step La3 to obtain a binary image containing only pupil information of the human eye;

[0142] Step Lb2: extracting the boundary pixel points of the pupil of the human eye 7 from the binary image obtained in step Lb1 through the CHAIN_APPROX_SIMPLE operator to obtain a binary image containing only the pupil boundary of the human eye 7;

[0143] Step L3-A3 includes the following steps:

[0144] Step Lc1: The pixel points obtained in step Lb2 are converted into T X) -1 X T Y is fitted into a circle, and the pupil center coordinates are the center of the fitted circle , where P is the parameter information of the fitted circle, X and Y are the row vectors constructed by the horizontal and vertical coordinates of the pixel points, respectively. T is the transpose of the row vector X;

[0145] Step Lc2: Draw a fitting line according to the parameters of the fitting circle obtained in step Lc1 and map it to the original image;

[0146] The automatic positioning method in the method for automatically positioning the pupil of the human eye 7 in step 4 comprises the following steps:

[0147] Step L4-A1, calculating the horizontal and vertical pixel differences between the pixel coordinates of the pupil center and the image center, and obtaining the horizontal and vertical pixel offsets of the pupil center;

[0148] Step L4-A2, converting the horizontal and vertical pixel offsets of step L4-A1 into the movement of the motor, and controlling the motor to complete the automatic positioning of the pupil of the human eye 7;

[0149] The automatic positioning of the pupil of the human eye 7 comprises the following steps:

[0150] Step Ld1: The horizontal coordinate of the pupil center pixel obtained in step Lc1 is and pixel ordinate The horizontal coordinate of the image center and the vertical coordinate Subtract and get the horizontal pixel offset and vertical pixel offset ,in , ;

[0151] Step Ld2: The horizontal pixel offset obtained in step Ld1 and vertical pixel offset By formula The actual movement amount of the motor is calculated, so as to control the movement of the motor to complete the automatic positioning of the pupil of the human eye 7, wherein: is the pixel size of the image, The horizontal or vertical pixel offset. Embodiment 5

[0152] On the basis of the second embodiment, when the left and right sides of the eyepiece 5 of the confocal laser ophthalmoscope are provided with an infrared camera 91 and an illumination lamp 92 used in conjunction with the infrared camera 91: Fig.11 As shown, the following steps are included:

[0153] A human eye positioning method for a human eye positioning device of a confocal laser ophthalmoscope comprises the following steps:

[0154] Step 1, initial calibration of the test position, including calibrating and adjusting the relative position relationship between the refractive component 3 and the human eye 7 and the eyepiece 5 in height, left-right, front-back direction; and the refractive component 3 and the human eye 7 are coaxial with the eyepiece 5 in height, left-right, front-back direction, and the fundus retinal image completely fills the preview window;

[0155] Specifically, in step 1, the initial calibration of the test position includes the following steps:

[0156] The positioning of the refractive component 3, the human eye 7 and the eyepiece 5 in height and left-right directions includes the following steps: positioning the refractive component 3, the human eye 7 and the eyepiece 5 in height and left-right directions by using the image of the human cornea or the front surface of the human eye and the position of the pupil in the image obtained by the confocal laser system of the laser confocal ophthalmoscope or the image obtained by the infrared camera 91; making the refractive component 3 and the human eye 7 coaxial with the eyepiece 5 in height and left-right directions;

[0157] The positioning of the refractive assembly 3, the human eye 7 and the eyepiece 5 in height and left and right directions includes the following steps:

[0158] In the coaxial adjustment method, the cornea and pupil of the human eye 7 are positioned in height and left and right directions by using an infrared camera 91, and the following steps are included: when a pair of infrared cameras 91 and a pair of lighting lamps 92 are arranged on the left and right sides of the eyepiece 5, the lighting lamps 92 on the left and right sides illuminate the front surface of the human eye 7 so that the infrared cameras 91 on the left and right sides of the eyepiece 5 can capture the image of the front surface of the human eye 7; and the two left and right infrared cameras 91 take pictures of the human eye 7 at the same time, observe the position of the eye surface and pupil of the human eye 7 in the field of view of the left and right infrared cameras 91, determine the relative position of the human eye 7 and the equipment in terms of up and down, left and right, and front and back, and then drive the three-dimensional motion platform 2 to adjust the relative position between the refractive component 3 and the eyepiece 5 and the human eye 7, until the human eye 7 is at the preset position of the image captured by the left and right infrared cameras 91.

[0159] The positioning of the front-to-back distance between the refractive component 3 and the human eye 7 and the eyepiece 5 includes the following steps: adjusting the distance position of the refractive component 3 and the human eye 7 in the front-to-back direction from the eyepiece 5, observing the size of the image of the retinal image of the human eye 7 in the preview window of the fundus confocal ophthalmoscope, until the fundus retinal image completely fills the preview window, at which time the human eye 7 and the eyepiece 5 are coaxial in three dimensions of height, left and right, and front and back.

[0160] Adjusting the distance between the refractive assembly 3 and the human eye 7 and the eyepiece 5 in the front-to-back direction comprises the following steps:

[0161] After the refractive assembly 3 and the eyepiece 5 are adjusted to be coaxial with the height and left-right direction of the human eye 7, the front-back position of the eyepiece 5 is adjusted according to the position of the human eye 7 in the two infrared cameras 91, and the distance between the device and the human eye 7 is judged based on the proportion of the preview window occupied by the image by observing the size of the fundus retinal image in the preview window of the fundus confocal ophthalmoscope interface, until the fundus retinal image completely fills the preview window, so that the human eye 7 and the eyepiece 5 are completely overlapped in three dimensions of height, left-right, and front-back.

[0162] Step 2, after completing the initial calibration of the test position in step 1, obtaining a grayscale image containing human eye pupil information; specifically, obtaining two grayscale images containing human eye pupil information;

[0163] Step 3, performing image processing on the grayscale image containing human eye pupil information obtained in step 2 to obtain the central pixel coordinates of the human eye pupil; specifically, performing image processing on the two grayscale images obtained in step 2 at the same time to obtain the central pixel coordinates of the human eye pupil in the two human eye images respectively;

[0164] Step 4, compare the central pixel coordinates of the pupil of the human eye obtained in step 3 with the central position of the template image, and control the refractive motor to automatically locate the pupil of the human eye after unit conversion; make the fundus retinal image in the preview window the clearest, and complete the focusing process of the retina of the human eye 7. Specifically, compare the two pixel coordinates obtained in step 3 with the central position of the template image, and control the motor to automatically locate the pupil of the human eye after unit conversion.

[0165] Specifically, the template image is an image with the correct center position of the human eye taken with a standard fixture (located at the working distance of the system and at the center of the system). The working distance refers to the universal shooting distance of the standard template image, and the distance between the eyepiece and the human eye is 10 mm. In this embodiment, the standard fixture used is two grayscale images containing human eye pupil information taken by infrared cameras set on the left and right sides of the eyepiece, and the two grayscale images are merged to obtain the template image.

[0166] Furthermore, for the method of obtaining the template image in step 2, the method includes the following steps:

[0167] Step S2-A1, stitching the two grayscale images up and down to construct a template image;

[0168] The method for obtaining the coordinates of the center of the pupil of the human eye in step 3 includes the following steps:

[0169] Step S3-A1, filtering and denoising the grayscale image and adaptively binarizing it to obtain a binary image containing pupil information of the human eye;

[0170] Step S3-A2, filtering pixel blocks containing pupil information in the binary image, and extracting and sampling the edges of the filtered pixel blocks to obtain a pixel point group to be fitted;

[0171] Step S3-A3, performing least square fitting on the pixel point group in step S3-A2 to obtain the central pixel coordinates of the pupil of the human eye;

[0172] Step S3-A1 includes the following steps:

[0173] Step Sa1: Filter the grayscale image obtained in step 3 through a Gaussian filter operator, where the Gaussian filter formula is:

[0174] ;

[0175] In the above formula, is the standard deviation of the Gaussian filter; (x0, y0) is the center coordinate of the Gaussian filter, and (x, y) is the coordinate of the rest of the Gaussian filter except the center coordinate;

[0176] Step Sa2: binarizing the image obtained in step Sa1 using a binarization algorithm to obtain a binary image containing pupil information of the human eye;

[0177] Step S3-A2 includes the following steps:

[0178] Step Sb1: fill all pixel blocks in the binary image in Sa2 to obtain a filled binary image;

[0179] Step Sb2: Delete the pixel blocks whose pixel number is less than a certain threshold in the binary image obtained by Sb1, and remove the binary image of interference information;

[0180] Step Sb3: Delete the pixel blocks at the image boundary in the binary image obtained by Sb2 to obtain a binary image containing only the pupil information of the human eye;

[0181] Step Sb4: extracting the edge of the pupil of the human eye by using the Canny edge detection operator for the pupil information of the human eye in the binary image obtained in Sb3, and obtaining a binary image of the edge of the pupil of the human eye;

[0182] Step Sb5: performing forward gradient sampling in a clockwise direction on the edge of the pupil of the human eye obtained in Sb4 to remove pixel points that do not belong to the edge of the pupil of the human eye, and obtaining a filtered pixel point group of the edge of the pupil of the human eye;

[0183] Step S3-A3 includes the following steps:

[0184] Step Sc1: Fit the pixel point group obtained in step Sb5 into an ellipse, and obtain the pupil center coordinate as the center of the fitted ellipse;

[0185] Step Sc2: Draw a fitting line according to the parameters of the fitting circle obtained in step Sc1 and map it to the original image;

[0186] The method for automatically locating the pupil of the human eye in step 4 includes the following steps:

[0187] Step S4-A1, respectively calculating the horizontal and vertical pixel differences between the pixel coordinates of the two pupil centers and the center of the template image to obtain the horizontal and vertical pixel offsets of the pupil centers;

[0188] Step S4-A2, converting the horizontal and vertical pixel offsets in step S4-A1 into the movement of the motor, and controlling the motor to complete the automatic positioning of the pupil of the human eye;

[0189] The method for automatically locating the pupil of a human eye, step S3-A1 comprises the following steps:

[0190] Step Sd1: Control the motor to move to a rough position so that the left and right iris cameras can capture images containing human pupil information;

[0191] Step Sd2: The coordinates of the pupil center of the human eye are obtained by real-time processing of the pupil images collected by the left and right iris cameras: and ;

[0192] Step Sd3: Control the motor to make the template image center Respectively and Overlap, that is: and , after the overlap, the motor is controlled to stop moving, thereby completing the automatic positioning of the human eye pupil. Embodiment 6

[0193] Based on the third embodiment, Fig.12 As shown, an infrared camera 91 is placed on the top of the eyepiece 5, and the infrared camera 91 is vertically parallel to the eyepiece 5. The lighting lamp 92 is not limited to be placed on the right side of the infrared camera 91, and can also be placed at any angle on the top or left side.

[0194] Specifically, the working method of this embodiment is as follows Fig.13 , through a pre-set device and a preset working distance counted by most human eyes 7, it is ensured that the photographed human eye 7 is within the field of view of the infrared camera 91, and then the device is adjusted left and right until the pupil and image of one of the human eyes 7 are at the horizontal center position of the field of view of the infrared camera 91, so as to distinguish the left and right eyes, and then move forward and backward until the fundus retinal image of the human eye 7 appears, and then determine the movement direction of the device through the effective field of view of the fundus image, and finally complete the positioning function of the human eye 7.

[0195] A method for locating a human eye 7 for a human eye 7 locating device of a confocal laser ophthalmoscope comprises the following steps:

[0196] Step 1, initial calibration of the test position, including calibrating and adjusting the relative position relationship between the refractive component 3 and the human eye 7 and the eyepiece 5 in height, left and right, and front and back directions; and the refractive component 3 and the human eye 7 are coaxial with the eyepiece 5 in height, left and right, and front and back directions, and the fundus retinal image completely fills the preview window.

[0197] In step 1, the initial calibration of the test location includes the following steps:

[0198] The positioning of the refractive component 3, the human eye 7 and the eyepiece 5 in height and left-right directions includes the following steps: positioning the refractive component 3, the human eye 7 and the eyepiece 5 in height and left-right directions based on the positions of the human cornea or the anterior surface image of the human eye and the pupil in the image obtained by the confocal laser system of the laser confocal ophthalmoscope or the image obtained by the infrared camera 91; making the refractive component 3 and the human eye 7 coaxial with the eyepiece 5 in height and left-right directions.

[0199] The positioning of the refractive assembly 3, the human eye 7 and the eyepiece 5 in height and left and right directions includes the following steps:

[0200] In the coaxial adjustment method, the cornea and pupil of the human eye are positioned in height and left and right directions by using an infrared camera 91, and the following steps are included: when the infrared camera 91 and the illuminator 92 are arranged at the upper end of the eyepiece 5, the illuminator 92 provides lighting for the infrared camera 91 on the top of the eyepiece 5 to increase the brightness of the image of the human eye taken by the infrared camera 91, and the infrared camera 91 is adjusted by the pre-three-dimensional motion stage 2 so that the reference point of the image of the infrared camera 91 coincides with the optical axis of the eyepiece 5; when the human eye 7 is photographed and collected, by judging the position of the cornea and pupil of the human eye 7 in the infrared camera 91 and the position relationship of the reference point, the three-dimensional motion stage 2 is driven to adjust the eyepiece 5 and the refractive component 3 so that the cornea and pupil of the human eye 7 coincide with the image reference point of the infrared camera 91, and at this time, the refractive component 3 and the human eye 7 are coaxial with the eyepiece 5 in height and left and right directions.

[0201] The positioning of the front-to-back distance between the refractive component 3 and the human eye 7 and the eyepiece 5 includes the following steps: adjusting the distance position of the refractive component 3 and the human eye 7 in the front-to-back direction from the eyepiece 5, observing the size of the image of the retinal image of the human eye 7 in the preview window of the fundus confocal ophthalmoscope, until the fundus retinal image completely fills the preview window, at which time the human eye 7 and the eyepiece 5 are coaxial in three dimensions of height, left and right, and front and back.

[0202] Adjusting the distance between the refractive assembly 3 and the human eye 7 and the eyepiece 5 in the front-to-back direction comprises the following steps:

[0203] After adjusting the refractive component 3 and the eyepiece 5 to be coaxial with the height and left-right direction of the human eye 7, the three-dimensional motion platform 2 is driven to move the refractive component 3 and the eyepiece 5 to a position close to the human eye 7. At this time, the size of the retinal image of the human eye 7 in the preview window of the fundus confocal ophthalmoscope is observed until the fundus retinal image completely fills the preview window. At this time, the front-to-back distance adjustment between the eyepiece 5 and the human eye 7 is completed. At this time, the human eye 7 and the eyepiece 5 are coaxial in three dimensions of height, left-right, and front-to-back.

[0204] Step 2, after completing the initial calibration of the test position in step 1, obtaining a grayscale image containing human pupil information;

[0205] Step 3, performing image processing on the grayscale image containing the pupil information of the human eye obtained in step 2 to obtain the central pixel coordinates of the pupil of the human eye;

[0206] Step 4, compare the central pixel coordinates of the human eye pupil obtained in step 3 with the central position of the template image, and control the refractive motor to automatically position the human eye pupil after unit conversion; make the fundus retinal image in the preview window the clearest, and the focusing process of the retina of the human eye 7 is completed at this time.

[0207] Specifically, the template image is an image with the correct center position of the human eye taken with a standard fixture (located at the working distance of the system and at the center of the system). The working distance refers to the universal shooting distance of the standard template image, and the distance between the eyepiece and the human eye is 10 mm. In this embodiment, the standard fixture used is two grayscale images containing human eye pupil information taken by an infrared camera set at the top of the eyepiece, and the two grayscale images are merged to obtain the template image.

[0208] Furthermore, the image processing steps and the conversion method of the motor moving distance and the image pixels of this embodiment are the same as those of the first embodiment and the fourth embodiment.

[0209] Based on the ideal embodiment of the present invention, through the above description, relevant personnel can make various changes and modifications without departing from the technical concept of the present invention. The technical scope of the present invention is not limited to the content in the specification, and the technical scope must be determined according to the scope of the claims.

Claims

1. A human eye positioning method for a human eye positioning device of a confocal laser ophthalmoscope, characterized in that: The following steps are involved: Step 1, initial calibration of the test position, including calibrating and adjusting the relative position relationship between the refractive component and the human eye and the eyepiece in height, left-right, front-back direction; the refractive component and the human eye are coaxial with the eyepiece in height, left-right, front-back direction, and the fundus retinal image acquired by the confocal laser ophthalmoscope fills the preview window of the confocal laser ophthalmoscope; Step 2, after completing the initial calibration of the test position in step 1, obtaining a grayscale image containing human pupil information; Step 3, performing image processing on the grayscale image containing the pupil information of the human eye obtained in step 2 to obtain the central pixel coordinates of the pupil of the human eye; The method for obtaining the coordinates of the center of the pupil of the human eye described in step 3 includes the following steps: Step L3-A1, performing filtering denoising, adaptive binarization and boundary removal processing on the grayscale image to obtain a binary image containing human pupil information; Step L3-A2, screening and contour extraction of the largest pixel block in the binary image in step L3-A1 to obtain a binary image containing only the pupil boundary of the human eye; Step L3-A3, performing least square fitting on the pixel points at the boundary of the pupil of the human eye extracted from the binary image in step L3-A2 to obtain the central pixel coordinates of the pupil of the human eye; Step L3-A1 includes the following steps: Step La1: Filter the grayscale image obtained in step 2 through an improved mean filter operator, where the improved operator kernel is for: ； Where I is the set of all grayscale values ​​in the processed image, w and h are the width and height of the operator kernel respectively; after filtering, the denoised human pupil grayscale image is obtained; Step La2: Binarize the grayscale image obtained in step La1 using an improved binarization algorithm. The calculation formula is: ； Among them, s is the grayscale value of the image, which is [0,255]. Represents the inverse function of the total correlation of the image grayscale probability distribution, where The calculation method is: ； in, yes A probability distribution function of is the total probability that the gray value in an image reaches s; m is an integer, ; After binarization, a binary image containing human pupil information is obtained; Step La3: The binary image obtained in step La2 is subjected to a boundary removal operator to delete the signal blocks at the image boundary, thereby obtaining a binary image of the pupil of the human eye after the boundary is removed; The step L3-A2 comprises the following steps: Step Lb1: extract the pixel block with the largest number of pixels from the binary image obtained in step La3 to obtain a binary image containing only pupil information of the human eye; Step Lb2: extracting the boundary pixel points of the pupil of the human eye from the binary image obtained in step Lb1 through the CHAIN_APPROX_SIMPLE operator to obtain a binary image containing only the boundary of the pupil of the human eye; The step L3-A3 comprises the following steps: Step Lc1: The pixel points obtained in step Lb2 are converted into T X) -1 X T Y is fitted into a circle, and the pupil center coordinates are the center of the fitted circle , where P is the parameter information of the fitted circle, X and Y are the row vectors constructed by the horizontal and vertical coordinates of the pixel points, respectively. T is the transpose of the row vector X; Step Lc2: Draw a fitting line according to the parameters of the fitting circle obtained in step Lc1 and map it to the original image; Step 4, compare the central pixel coordinates of the human eye pupil obtained in step 3 with the central position of the template image, and control the refractive motor to automatically locate the human eye pupil after unit conversion; make the fundus retinal image in the preview window of the confocal laser ophthalmoscope the clearest, and complete the focusing process of the retina of the human eye.

2. The human eye positioning method for the human eye positioning device of the confocal laser ophthalmoscope according to claim 1, characterized in that: In step 1, the initial calibration of the test location includes the following steps: The positioning of the refractive component, the human eye and the eyepiece in height and left-right directions comprises the following steps: positioning the refractive component, the human eye and the eyepiece in height and left-right directions by using the positions of the human cornea or the front surface image of the human eye and the pupil in the image acquired by the confocal laser system of the laser confocal ophthalmoscope or the image acquired by the infrared camera; making the refractive component and the human eye coaxial with the eyepiece in height and left-right directions; The positioning of the front-to-back distance between the refractive component and the human eye and the eyepiece includes the following steps: adjusting the distance position of the refractive component and the human eye from the eyepiece in the front-to-back direction, observing the size of the human eye fundus image in the preview window of the fundus confocal ophthalmoscope until the fundus retinal image completely fills the preview window, at which time the human eye and the eyepiece are coaxial in three dimensions: height, left and right, and front and back.

3. The human eye positioning method for the human eye positioning device of the confocal laser ophthalmoscope according to claim 2, characterized in that: The positioning of the refractive assembly, the human eye and the eyepiece in height and left and right directions includes the following steps: Based on the positions of the human cornea and pupil in the image field obtained by the confocal laser system, the three-dimensional motion stage is driven to adjust the position of the refractive component and the eyepiece so that the height and left-right direction of the human eye and the eyepiece are in the coaxial position; Or, the coaxial adjustment method uses an infrared camera to locate the cornea and pupil of a human eye in height and left-right directions, comprising the following steps: when a pair of infrared cameras and a pair of lighting lamps are arranged on the left and right sides of an eyepiece, the lighting lamps on the left and right sides illuminate the front surface of the human eye so that the infrared cameras on the left and right sides of the eyepiece can capture the image of the front surface of the human eye of the human eye; and the two left and right infrared cameras simultaneously take pictures of the human eye, observe the positions of the human eye surface and the pupil in the field of view of the left and right infrared cameras, determine the relative positions of the human eye and the device up and down, left and right, and front and back, and then drive the three-dimensional motion stage to adjust the relative positions between the refractive component and the eyepiece and the human eye, until the coordinate position of the image of the human eye captured by the infrared cameras on the left and right sides coincides with the preset position, and the preset position is located at the center of the human eye in the template image; Alternatively, the coaxial adjustment method uses an infrared camera to position the cornea and pupil of the human eye in height and left and right directions, including the following steps: when the infrared camera and the lighting lamp are arranged at the upper end of the eyepiece, the lighting lamp provides lighting for the infrared camera at the top of the eyepiece to increase the brightness of the image of the human eye taken by the infrared camera, and the infrared camera is adjusted by a pre-three-dimensional motion stage so that the reference point of the image of the infrared camera coincides with the optical axis of the eyepiece; when photographing and collecting the human eye, by judging the position relationship between the position of the cornea and pupil of the human eye in the infrared camera and the position of the reference point, the three-dimensional motion stage is driven to adjust the eyepiece and the refractive component so that the cornea and pupil of the human eye coincide with the reference point of the image of the infrared camera, and at this time, the refractive component and the human eye are coaxial with the eyepiece in height and left and right directions.

4. The human eye positioning method for the human eye positioning device of the confocal laser ophthalmoscope according to claim 3, characterized in that: Adjusting the distance between the refractive assembly and the human eye and the eyepiece in the front-to-back direction includes the following steps: After the refractive assembly and the eyepiece are adjusted to be coaxial with the height and left-right direction of the human eye, the three-dimensional motion stage is driven to move the refractive assembly and the eyepiece to a position close to the human eye. At this time, the size of the image of the human eye retinal image in the preview window of the fundus confocal ophthalmoscope is observed until the fundus retinal image completely fills the preview window. At this time, the front-to-back distance adjustment between the eyepiece and the human eye is completed. At this time, the human eye and the eyepiece are coaxial in three dimensions: height, left-right, and front-to-back. Alternatively, after adjusting the refractive assembly and the eyepiece to be coaxial with the height and left-right direction of the human eye, adjust the front-to-back position of the eyepiece according to the position of the human eye in the two infrared cameras, and judge the distance between the device and the human eye based on the proportion of the preview window occupied by the image by observing the size of the fundus retinal image in the preview window of the fundus confocal ophthalmoscope interface, until the fundus retinal image completely fills the preview window, so that the human eye and the eyepiece completely overlap in three dimensions of height, left-right, and front-to-back.

5. The human eye positioning method for the human eye positioning device of the confocal laser ophthalmoscope according to claim 4, characterized in that: In step 2, the template image is obtained by obtaining an ocular surface image of a human eye through a confocal laser system of a confocal laser ophthalmoscope, and the obtained ocular surface image includes a grayscale image of human pupil information; The automatic positioning method in the method for automatically positioning the pupil of the human eye described in step 4 comprises the following steps: Step L4-A1, calculating the horizontal and vertical pixel differences between the pixel coordinates of the pupil center and the image center, and obtaining the horizontal and vertical pixel offsets of the pupil center; Step L4-A2, converting the horizontal and vertical pixel offsets described in step L4-A1 into motor movement, and controlling the motor to automatically locate the pupil of the human eye; Automatic positioning of the pupil of the human eye includes the following steps: Step Ld1: The horizontal coordinate of the pupil center pixel obtained in step Lc1 is and pixel ordinate The horizontal coordinate of the image center and the vertical coordinate Subtract and get the horizontal pixel offset and vertical pixel offset ,in , ; Step Ld2: The horizontal pixel offset obtained in step Ld1 and vertical pixel offset By formula The actual movement of the refractive motor is calculated, so as to control the movement of the refractive motor to complete the automatic positioning of the pupil of the human eye. is the pixel size of the image, The horizontal or vertical pixel offset.

6. A human eye positioning device for a confocal laser ophthalmoscope, which realizes the human eye positioning method for a human eye positioning device for a confocal laser ophthalmoscope according to any one of claims 1 to 5, characterized in that: It comprises a three-dimensional motion platform and a forehead support arranged on a base, wherein an eyepiece and a refractive component corresponding to the eyepiece are arranged on the three-dimensional motion platform; The eyepiece is located between the refractive assembly and the forehead support, and the eyepiece is installed on one side of the refractive assembly through a lens barrel; the eyepiece is used to refract light into the human eye; The three-dimensional motion stage is used to drive the relative displacement of the refractive assembly and the eyepiece relative to the object to be detected supported by the forehead support; A refractive motor is arranged in the refractive assembly, and the refractive motor is connected to an optical path assembly, and the optical path assembly is used in conjunction with the eyepiece, and the refractive assembly is used for refractive compensation when photographing a fundus image of a human eye, so as to achieve fundus image focusing; The three-dimensional motion platform and the refractive component are both interconnected with the confocal laser system of the confocal laser ophthalmoscope. The confocal laser system is used for collecting fundus and / or ocular surface images. The refractive component realizes refractive compensation through a refractive compensation mechanism.

7. The human eye positioning device for confocal laser ophthalmoscope according to claim 6, characterized in that: Infrared cameras are arranged on the left and right sides of the eyepiece, and lighting lamps are arranged on the outside of the infrared cameras; the infrared cameras are used to collect images of the cornea and pupil of the human eye; the lighting lamps are used to fill in the light of the object to be detected during image collection; The lens barrel is used to assemble and fix the infrared camera and the illuminating lamp with the eyepiece.

8. The human eye positioning device for confocal laser ophthalmoscope according to claim 6, characterized in that: An infrared camera is arranged on the upper side of the eyepiece and / or on the lower side of the eyepiece; and an illumination lamp is arranged on one side of the infrared camera; the infrared camera is used to collect images of the cornea and pupil of the human eye; and the illumination lamp is used to supplement light for the object to be detected during image collection.

Images