JP4751689B2 - Eye surface analysis system - Google Patents

Description

  The present invention relates to a method for acquiring unevenness information existing on the surface of an eyeball, and in particular, an eyeball surface that quantitatively evaluates the state of a streak region by analyzing image data obtained by photographing a fluorescently stained eye to be examined It relates to an analysis system.

  In the field of ophthalmology, an examination for observing the state of the surface of the eyeball is performed. In particular, in recent years, dry eye has been diagnosed by observing the behavior of the tear film formed on the surface of the eyeball, and a device for the purpose of observing the tear film has been developed. Has been. (See Patent Documents 1 and 2)

  The surface of the eyeball is roughly divided into a cornea and a conjunctiva. In the conjunctiva portion, the conjunctiva is covered with the bulbar conjunctiva, and is associated with the sclera at the corneal boundary to form many wrinkles, thereby smoothing the eye movement. This wrinkle shows a mesh-like structure in young people, but the mesh portion is divided with aging and appears different. When the eyeball is observed using the fluorescent staining solution, a portion having a high fluorescent staining density appears along the mesh-like wrinkles. This high fluorescent staining area is the site where tears accumulate, and knowing its shape, size, and direction is useful for obtaining information on lesions such as dry eye that present symptoms depending on the behavior of the tear film. is there.

  The ophthalmologic apparatus described in Patent Document 1 is composed of a light projecting optical system and a light receiving optical system, and includes a slit lamp microscope and a CCD camera for observing the eye to be examined, and a cornea of the eye to be examined instilled with a fluorescent staining solution. Photograph the surface with a CCD camera. Then, the thickness of the tear film is analyzed from the concentration of the photographed fluorescent staining solution. However, this invention only measures the fluorescence intensity from tear fluid quantitatively under various measurement conditions, and does not quantitatively analyze the surface state of the cornea.

  In the ophthalmologic apparatus described in Patent Document 2, light is projected onto a predetermined point on the outermost tear layer (lipid layer) of the subject's eye, and an interference pattern of reflected light reflected on the front and back surfaces of the lipid layer is photographed with a CCD camera. Is done. An image photographed by the CCD camera is displayed on a monitor, and the examiner simply makes a diagnosis by observing the state and state of the lipid layer from the image displayed on the monitor. In this ophthalmologic apparatus, the examiner subjectively determines the state of the tear film of the eye to be examined by comparing the image observed by the CCD camera with a reference image serving as an index. Therefore, the evaluation criteria are not unified by the examiner, and universal diagnosis cannot be expected.

  Moreover, the main observation site | part of any apparatus is a cornea, It does not carry out the quantitative analysis by specifying the wrinkle of a conjunctiva.

  On the other hand, in the beauty field, wrinkle analysis is performed as a main element of evaluation relating to skin (for example, Patent Documents 3 to 6).

  In the apparatus described in Patent Document 3, a gaze area is photographed and analyzed centering on a luminance histogram due to a shadow formed by lighting skin wrinkles, but the skin wrinkles themselves are identified and extracted individually. Do not mean.

  In the devices described in Patent Documents 4 and 5, in order to evaluate not only skin wrinkles but also skin-specific characteristic elements such as skin texture, the elements related to the positional relationship between the illumination system and the skin wrinkles are emphasized. The degree of skin roughness is quantitatively evaluated using a light / dark enhancement imaging apparatus. These techniques are specialized for the skin, as is apparent from the measurement technique.

In the device described in Patent Literature 6, the frequency component of skin wrinkles is separated by performing filtering using Fourier transform on the gaze region, and the gaze region is separated. It does not identify the wrinkle of the and extract its direction component.
JP-A-6-277179 JP 7-136120 A Japanese Patent Laid-Open No. 2-46833 Japanese Patent Laid-Open No. 6-189842 Japanese Patent Application Laid-Open No. 7-231883 JP-A-7-116146

  The present invention has been made in view of the above-described circumstances, and an object of the present invention is to use a digital image obtained by photographing an arbitrary gaze area on an eyeball stained with a fluorescent substance, and gaze at the gaze area. To provide an eyeball surface analysis system capable of identifying and extracting a streak region represented by wrinkles in a region and quantitatively analyzing and displaying the size and direction of the streak region is there.

  In order to solve the above problems, an eyeball surface analysis system according to the present invention is a system for analyzing the surface state of the anterior ocular segment of a subject's eye stained with a solution containing a labeled fluorescent substance, and the fluorescently stained anterior eye A function of acquiring a digital image of a partial image, a function of removing disturbance information included in the acquired image data, a function of specifying a streak area based on grayscale information for each pixel included in the image data, and image data A function for dividing the image into predetermined ranges, a function for calculating the fluorescence density centroid based on the grayscale image corresponding to the fluorescence intensity in each divided area, and the position information of the streak area identified as the density centroid for each divided area A function for extracting a segmented area based on the function, a function for generating a vector related to a streak area for each extracted area, a function for analyzing a vector related to the generated streaky area, And having a function of outputting the analysis result of the vector according to the center of gravity and / or generated streak regions.

  As described above, the eyeball surface analysis system according to the present invention extracts a streak region on the surface of the eyeball by using a digitized image of the anterior segment of the eye to be examined that has been fluorescently stained, and performs various quantifications. Since it has a function to calculate numerical values that can be analyzed and statistically processed and to display the results, it is possible to perform secular changes for the same subject, quantitative comparison by measurement before and after drug administration, and before and after surgery, or subject A quantitative comparison of the differences between them is possible.

  Embodiments according to the present invention will be described below with reference to the drawings. In this embodiment, extraction / analysis is performed on wrinkles formed in the eyeball conjunctiva as a streak-like region.

  FIG. 1 is a basic processing flow in the present invention, FIG. 2 is a first embodiment incorporating an anterior ocular segment observation optical system incorporating a system according to the present invention, and FIG. 3 is incorporated in the apparatus of FIG. FIG. 4 shows the basic configuration of the anterior ocular segment observation optical system, and FIG. 4 shows the system configuration in this embodiment.

  The anterior ocular segment observation optical system 30 in FIG. 3 irradiates the eye C to be examined stained with a fluorescent staining material from the illumination light sources 31a and 31b and reflects the luminous flux reflected by the eye C to be examined. Is guided to the image sensor 37 through the magnifying lens 32, the half mirror 33, and the focusing lens 36. A light source 35 and a lens 34 for irradiating a light beam used for positioning the optical system at an appropriate position with respect to the eye C or fixing the eye C in the optical axis direction of the optical system are disposed. ing. Here, it is desirable that the wavelengths of the illumination light sources 31a and 31b are suitable for light emission of the fluorescent staining material.

  The apparatus 20 shown in FIG. 2 has a mechanism that can move the head unit 21 in the vertical direction with respect to the movable unit 24 by rotating the movable unit 24 and the joystick 26 having a mechanism that can slide in the front-rear and left-right directions. And the head portion 21 can be moved three-dimensionally with respect to the base portion. The anterior ocular segment observation optical system 30 in FIG. 3 is housed inside the head unit 21. The raw image signal output from the image sensor 37 is output to the display device 22 via a signal processing device (not shown) housed in the device 20. After the subject comes into contact with the jaw cradle 25, the examiner operates the joystick 26 while confirming the anterior eye portion image output to the display device 22, and the head portion 21 moves toward the subject. Align it so that it is in the proper position. After the alignment is completed, by pressing the push button 28, a digitized image is acquired in an image storage device (not shown) housed in the device 20.

  The flow from the acquisition of the anterior segment image to the display will be described below with reference to FIGS. The image sensor 37 shown in FIG. 4 corresponds to 37 in FIG. 3 described above, and always captures an image signal. The control device 41 monitors whether or not the push button 28 has been pressed. When the push button 28 has not been pressed, a raw image is output to the display device 22 via the image processing device 44 according to the flow indicated by the dotted line. To do.

  When the push button 28 is pressed, the control device 41 digitizes the image signal by the A / D converter 42 and stores it as image data in the frame memory 43 according to the flow indicated by the solid line. Thereafter, the stored image data signal is transferred to the image processing device 44. The process from when the push button 28 is pressed until it is transferred to the image processing means 44 via the frame memory 43 corresponds to the anterior segment digital image data acquisition step shown in step 1 of FIG. FIG. 5 shows an example in which the transferred image data is displayed as it is. The size of the image is 480 pixels vertically and 640 pixels horizontally.

  Subsequently, the process proceeds to the stage of removing background unevenness due to the disturbance shown in Step 2 of FIG. The disturbance described here is on the anterior ocular segment caused by the light from the illumination light sources 31a and 31b reflected by the room illumination represented by fluorescent lamps or indirectly by the apparatus or the like and projected onto the anterior ocular segment. Refers to the light that is locally irradiated. The above-mentioned image of FIG. 5 is an image of mainly the conjunctiva portion of the fluorescently stained eyeball. From this image, the wrinkles on the conjunctiva are filled with the fluorescent dye and appear as a streak pattern. The arcuate dark area at the top of the image is the cornea. However, even outside the cornea region, the vicinity of the center of the image is relatively bright, the portions near both ends of the image are dark, and wrinkles are also observed in the surrounding dark portions. This difference in brightness is background unevenness caused by the above-described disturbance. Each pixel constituting the image sensor 37 shown in FIG. 4 acquires luminance information caused by the background unevenness as well as luminance information corresponding to the concentration of the fluorescent substance contained in the tear fluid. Luminance information related to background unevenness may be erroneously extracted when a wrinkle region is subsequently extracted. Therefore, it is preferable to remove the luminance information.

  First, the background fluorescent staining density information is removed based on the idea that the wrinkle region is superimposed on a uniform background. In this removal process, the average value of the image luminance information in a wide area compared to the size of the wrinkle area or blood vessel, that is, the average fluorescent staining density of the selected area set around each pixel for all pixels of the image is obtained. FIG. 6 shows an example of the shape of the selected area. FIG. 6A shows a square region of N × N pixels centered on the reference pixel. In this case, the average fluorescent staining density in the reference pixel is a value obtained by dividing the sum of the luminance of (N × N) pixels by (N × N). FIG. 6B shows an example in which a circular area having a diameter of N pixels is designated. Even in this case, the average fluorescent staining density can be obtained by dividing the sum of the luminances of the pixels included in the selected region by the number of pixels. By subtracting the obtained background average fluorescent staining density information of each pixel from the density information of the original image data, background unevenness included in the original image data is removed, and the brightness of the entire image is made uniform. The shape and size of the selected area for obtaining the average fluorescent staining density affects the time required for the background unevenness removal process and the degree of wrinkle original information left in the processed image. For example, when the area is enlarged, a uniform image can be acquired without damaging the information based on wrinkles, but processing takes time. Conversely, if the area is reduced, the processing time can be shortened, but the influence on the luminance information due to wrinkles increases, and the error increases during the subsequent extraction processing. Furthermore, since the size of the wrinkle region formed on the image changes depending on the observation magnification, the shape and size of the selected region must be set in consideration of processing accuracy, processing speed, and observation magnification.

  Subsequently, the fluorescence density change value calculation for each pixel shown in step 3 of FIG. 1 is performed. Since there are many tears in the depressions represented by the wrinkle area on the surface of the eyeball, the fluorescence staining concentration is high, and in other areas where there is not much tear fluid, the fluorescence staining concentration is low compared to the depressions. Therefore, it is estimated that there is a concavo-convex shape boundary at a pixel position where the density change is large with respect to the adjacent region. Based on this idea, differentiation is performed in the vertical and horizontal directions for all the pixels constituting the image from which the background unevenness due to the disturbance is removed, and the square root of the square sum of the obtained differential values in the target pixel. Change value of fluorescent staining concentration. Then, the change values of the fluorescent staining density with the neighboring pixels are compared for all the pixels constituting the image, and the pixel having the large change value is set as a candidate for the boundary of the wrinkle region. Here, in order to improve the position accuracy, it is desirable to set the neighboring pixels to adjacent pixels. In this case, however, the difference between the change values to be compared is reduced, and pixels having the same change value are concentrated. There is a high possibility of erroneous determination in the area to be performed. On the other hand, if the setting of the pixel to be compared is set too far, the recognition accuracy of the wrinkle region is lowered. For this reason, the pixels used for comparison are preferably pixels that are located two pixels away from each other. FIG. 9 shows an example of pixels that are two pixels apart. FIG. 9A is an example showing a one-dimensional case and FIG. 9B is a two-dimensional case.

  Next, the streak area shown in step 4 of FIG. 1 is extracted. The average of the change value of the fluorescent staining density obtained from the pixel group of the boundary candidate of the wrinkle region extracted by the above procedure is used as a threshold value, and a pixel having a fluorescent staining density higher than the threshold value is extracted as a pixel existing in the wrinkle region. .

  The extracted wrinkle region varies in thickness and direction. In addition, in the image shown in FIG. 5 described above, superimposition of blood vessels that are not fluorescently stained and recognized as dark dendritic components directed upward from below is recognized on wrinkles that are recognized as bright portions by fluorescent staining. Thus, a part of the wrinkle may be apparently divided due to the invasion of the blood vessel or the like. Therefore, if a pixel group adjacent to the pixels extracted as a wrinkle region through the above-described steps is treated as one wrinkle as it is, it will be affected by a division factor such as a blood vessel. It is preferable to remove the information related to the blood vessel or the like which becomes an obstacle during the morphological analysis as well as the background unevenness caused by the above-described disturbance.

  For this reason, the extracted image is divided into predetermined shape regions shown in step 5 of FIG. 1 for the purpose of removing information that becomes an obstacle in analyzing the wrinkle shape and improving the efficiency of the subsequent analysis processing. As the predetermined shape for performing the region division, the shape shown in FIG. 6 can be used. When the square area composed of N × N pixels shown in FIG. 6A is adopted, the divided areas are overlapped by making the pixel N on one side the common divisor of both the vertical and horizontal pixels of the entire image. Therefore, the subsequent processing time can be shortened because the entire image can be analyzed. However, it is not necessarily limited to the common divisor. When a circular area having a diameter of N pixels shown in FIG. 6B is employed, it is necessary to overlap some areas in order to analyze the entire image. On the other hand, since only pixels located within the substantially same distance range are selected, a more natural region can be set as compared with a square region that depends on an orthogonal arrangement of pixels. The above-mentioned pixel N is preferably set so that the width dimension of a wrinkle region having a general size formed on an image depending on the imaging magnification is accommodated. Furthermore, it is desirable to be able to calculate and set the optimum pixel N by providing means for inputting or acquiring conditions such as imaging magnification information, image size, and width size of the wrinkle region.

  Subsequently, the fluorescence staining density centroid is calculated for each divided region shown in Step 6 of FIG. Here, the fluorescence staining density centroid is obtained from the fluorescence staining density value of the pixels included in the divided area and the position information of the pixels in the area.

  Next, a segmented region including a streak region shown in step 7 of FIG. 1 is extracted. Only the divided area included in the pixel group extracted as the wrinkle area in the above-described step 4 is extracted as a useful area in the subsequent analysis. FIG. 7 is an example showing the relationship between the pixels in the wrinkle region and the pixels in the density centroid in the divided region. FIGS. 7A and 7B show a state in which the density centroid pixel in the divided area exists in the wrinkle area, and FIGS. 7C and 7E show a case in which the density centroid pixel does not exist in the wrinkle area. It is an example. FIG. 7C shows a case where a density barycentric pixel exists between two wrinkle regions, and FIG. 7D shows a case where there is a connecting point that branches in two directions. Such a straight line approximates the wrinkle region. The divided areas that are difficult to do are excluded from the analysis target area because correct information cannot be obtained even if vector information calculation is performed thereafter. 7E, the wrinkle region shown in FIG. 7E can be approximated by a single straight line, and there are few applicable regions for the divided regions that deviate from the above-described extraction conditions, and the exclusion has almost no effect on the subsequent analysis results. Based on the premise, in order to avoid complicating the extraction conditions, it is excluded as in (C) and (D). FIG. 8 shows a result of extracting a divided area including a wrinkle area by setting one side of the square area shown in FIG. 6A as 20 pixels as the shape of the divided area. In FIG. 8, the divided frame is not displayed in the divided area excluded from the analysis target because the density barycentric pixel does not exist in the wrinkled area.

  Then, vector information of the streak-like region of the extraction region shown in step 8 of FIG. 1 is calculated. Here, the wrinkle area is vectorized for each of the divided areas extracted through step 7 described above. The vectorization described here is to obtain the direction (angle) and size (length) of the wrinkle region in the divided region. The least square error considering the fluorescence staining density value for the wrinkle region in the divided region is minimized by rotating the rotation line segment centered on the fluorescence staining barycentric pixel obtained in step 6 in FIG. 1 at a certain angle. Let the angle be the direction of the wrinkle region vector. The length of the rotation line segment is N pixels, which is the diameter of the maximum inscribed circle in the case of a square area of N × N pixels, and is the diameter of a circle area having a diameter of N pixels. The angle at which the line segment is rotated affects the accuracy and processing speed to be obtained, and therefore it is preferable to set the angle in consideration of the accuracy or the priority of high-speed processing. FIG. 10 shows the concept of calculating the vector angle of the wrinkle region in the divided region. In addition, the length of the wrinkle region in the divided region is the size of the vector. In this case, the wrinkle region having the center of gravity deviated from the center in the divided region is treated as a short line segment because it is close to the boundary. FIG. 11 shows an example in which the length and angle of the obtained wrinkle area vector for each divided area are displayed in a graphic form.

  Further, in FIG. 11, the ratio (area) of the wrinkle area in the divided area based on the average fluorescent staining density value in the divided area is displayed in gray level, and the calculated area in the step 6 in FIG. A point display of the wrinkle region centroid position is made. Thus, the image processing relating to the wrinkle region is completed, and the information acquired by the processing is transferred to the calculation means indicated by 45 in FIG.

  From here, the density centroid / vector information calculation data analysis shown in step 9 of FIG. 1 is performed. The arithmetic unit shown by 45 in FIG. 4 has an average fluorescent staining density, a vector direction (angle), and a vector size (representing the characteristics of the wrinkle region for each divided region obtained through step 8 in FIG. The distribution of the total number of extracted divided areas is analyzed for (length). The result is transferred to the display device 22 shown in FIG.

  Finally, the analysis result output shown in step 10 of FIG. 1 is performed. The display device indicated by 22 in FIG. 4 displays the analysis result transferred through step 9 in FIG. FIG. 12 shows a display example of the analysis result. The “average density in block” displays the distribution of the average fluorescent staining density obtained for each divided region, and the distribution state of the area ratio of the wrinkle region to the divided region can be grasped. The “identification area average density” displays the distribution of the average fluorescent staining density for the wrinkled area in each divided area, and the depth distribution state of the wrinkled area can be grasped. “Inclination” indicates the distribution of the inclination (angle) of the wrinkle area obtained for each divided area, and it can be understood that there are many wrinkle areas extending in the 0 ° (horizontal) direction. “Linearity” is the ratio of the length of the wrinkled area to one side or diameter of the divided area, and indicates the distribution of the length of the wrinkled area. There are many relatively long (continuous) wrinkled areas. You can see what you are doing.

  Information on the area, depth, slope, and length of the wrinkle region changes with age, and observing the progress can be used for early detection of conjunctival relaxation, which is often observed in the elderly. In the present embodiment, the entire extracted divided region is expressed by the above-described analysis numerical value. However, when attention is paid to a certain divided region, a new analysis result is obtained by comparing / correlating the analytical value with a neighboring divided region. Needless to say, it can be acquired.

  Further, in the above-described embodiment, the acquisition of the anterior ocular segment image by the configuration of the anterior ocular segment observation optical system shown in FIG. 3 is described, but the ocular refractive power, corneal shape, and intraocular pressure used in the ophthalmic examination are described. Needless to say, the function according to the present invention can be mounted as an accessory function by using an observation optical system for alignment with the eye to be examined, which is employed in an inspection apparatus such as corneal observation.

  13 and 14 show a second embodiment in which the system according to the present invention is incorporated. FIG. 13 shows a configuration in which a CCD camera 61 is attached to the slit lamp microscope 50 in order to acquire an anterior ocular segment image. FIG. 14 shows a system configuration in which an image photographed by the CCD camera 61 in FIG. 13 is taken into a computer 71 via an image signal transmission device 62 (video cable) and processed.

  The slit lamp microscope 50 shown in FIG. 13 illuminates the eye C through the prism 59 with the light beam emitted from the illumination unit 51. The examiner is composed of a movable portion 56 slidable with respect to the base portion 57, a magnifying lens unit 52, a joint 53, and an eyepiece portion 54 after the subject abuts his face on the jaw support 58. The column 55, which is attached to the observation optical system and moves up and down with respect to the movable portion 56, is moved by operating the joystick 60 to align the eye C to be examined. The illumination unit 51 is rotatable about the axis A, and is configured to be able to irradiate illumination light from an oblique direction of the eye C to be retracted from the optical path of the observation optical system. The magnifying lens unit 52 has a light beam separating member inside, and guides a part of the light beam of the eye observation image observed at the eyepiece 54 to the CCD camera 61. An eye observation image received by the CCD camera is output via an image signal transmission device 62 (video cable).

  FIG. 14 shows the flow of processing the image information output from the CCD camera 61 shown in FIG. A dotted line frame indicates a function configured by the computer 71. A computer 71 is connected to a signal generator 74 that generates a signal for instructing image capture, a display device 72, and a printer 73. An image signal is always output from the CCD camera 61. The examiner gives an image capture instruction by the signal generator 74 while observing the eye C from the eyepiece 54 of the slit lamp microscope 50 of FIG. When the control device 711 receives the signal from the signal generation device 74, it transfers the image signal to the A / D conversion device 712. Here, the digitized image signal is transferred to the image processing device 714 via the image data storage device 713. The above corresponds to the stage of digital image data acquisition shown in 1 of FIGS. In this embodiment, the A / D conversion device 712 is assumed to be an image processing board that can be built in or added to the computer 71, but is provided outside the computer 71 and transfers digitized image data to the computer 71. It can be replaced with a configuration to

  Since the subsequent processing procedure is the same as that of the first embodiment, description thereof is omitted. The final analysis result is displayed on a monitor 72 connected to the computer 71. Further, the analysis result may be output to a printer 73 connected to the computer 71.

  In this second embodiment, a slit lamp microscope generally used in ophthalmic examinations is used for acquiring an anterior segment image, and the acquired image is processed and analyzed by a computer. Observation is possible without having to prepare

  FIG. 15 shows a system configuration of a third embodiment in which a system according to the present invention in which the CCD camera 61 of FIG. 13 is replaced with a digital camera 81 in the configuration of the second embodiment is incorporated. Here, the image data storage device 82 is assumed to be temporary storage means (such as a compact flash memory) that can be attached to and detached from the digital camera 81. The image data stored in the image data storage device 82 is transferred to the image data storage device 713 of the computer 83 via the image data capturing device 831 (such as a compact flash memory slot) of the computer 83. It should be noted that the signal input device 74 shown in FIG. By directly acquiring digital image data in this way, the configuration relating to A / D conversion in 711 and 712 in FIG. 14 can be omitted.

  In this third embodiment, by replacing the second embodiment with a digital camera, not only the components were reduced, but the anterior segment image acquisition (imaging) was performed in the laboratory and stored. Image data can be read by a computer installed in a separate room, and the installation space in the examination room can be reduced. In the above-described embodiment, the detachable temporary storage means is taken as an example of the image data storage device used for acquiring the image data from the digital camera. However, the data can be transferred by connecting the digital camera and the computer with a cable. Needless to say, it is included.

  Further, by providing the device 20 described in the first embodiment with a video terminal for directly outputting an anterior ocular segment observation image, the slit lamp microscope 61 in the second embodiment can be substituted, and the device 20, the digital camera 81 of the third embodiment can be substituted by providing a video terminal or an image data recording device for outputting digitally processed image data, and the present invention can be used without preparing new equipment. It becomes possible to realize the function concerning.

  Furthermore, in the examples, the wrinkle region on the conjunctiva is the object of observation and analysis, but the analysis seems to be medically useful, such as the extraction and confirmation of linear wounds on the cornea due to the wearing of contact lenses, etc. Is also conceivable, and is not limited to the configuration described in this specification.

It is the figure which showed the processing flow which concerns on this invention. It is the figure which showed the 1st Example. It is the figure which showed the basic composition of the anterior ocular segment observation optical system. It is the figure which showed the processing flow which concerns on a 1st Example. It is the image which showed the eyeball (wrinkle area | region on the conjunctiva) dye | stained fluorescence. It is the figure which showed the example of the shape of the selection area | region. It is the figure which showed the relationship of the wrinkle area | region with the average fluorescence density gravity center of a division area. It is the image after screen division and extraction processing execution. It is the figure which showed the pixel position of two adjacent pixels. It is explanatory drawing regarding calculation of the inclination (angle) of a vector. It is an image after performing analysis of vector information of a wrinkle region. It is the figure which showed the example of a display concerning a quantitative analysis result. It is the figure which showed a part of acquisition means of the anterior ocular segment observation image which concerns on a 2nd Example. It is the figure which showed the processing flow which concerns on a 2nd Example. It is the figure which showed the processing flow which concerns on a 3rd Example.

Claims (6)

A system for analyzing the surface state of the anterior segment of an eye to be examined stained with a solution containing a labeled fluorescent substance,
The ability to acquire a digital image of the anterior segment image that has been fluorescently stained;
A function of removing disturbance information included in the acquired image data, a function of specifying a streak area based on grayscale information for each pixel included in the image data,
A function of dividing image data into a predetermined range;
A function of calculating a fluorescence density centroid based on a grayscale image corresponding to the fluorescence intensity in each divided region;
A function of extracting a divided region based on the density centroid for each divided region and the position information of the specified streak region;
A function for generating a vector related to a streaky region for each extracted region;
An anterior segment having a function of analyzing a vector related to the generated streaky region and a function of outputting an analysis result of the concentration centroid and / or a vector related to the generated streaky region Surface analysis system.   The anterior ocular surface analysis system according to claim 1, wherein the streak region is stained based on an uneven state such as wrinkles or scratches formed on the anterior ocular surface.   The anterior segment surface analysis system according to claim 1, wherein the analysis of the generated vector is statistical processing based on a length and / or direction of the vector.   4. The anterior eye according to claim 1, wherein the output outputs a number distribution of a ratio of a length of a vector generated for each divided region to a maximum vector assumed in the divided region. 5. Surface analysis system.   5. The anterior ocular surface analysis system according to claim 1, wherein the output outputs a number distribution in each direction of a vector generated for each divided region.   An ophthalmologic apparatus having the anterior ocular surface analysis system according to claim 1.

Images