JP6187751B2 - Image processing program - Google Patents

Description

About images processing program that processes the image that the eye to be examined is photographed.

  As an example of an ophthalmologic imaging apparatus that images a subject's eye, an apparatus that scans the fundus of a subject's eye with light and receives reflected light from the fundus is known (for example, Patent Document 1). .

JP 2011-115301 A

Incidentally, when the view the list of the imaging results of ophthalmologic imaging apparatus, it is conceivable to use a reduced image of the fundus image. However, the display mode that makes it easy for the examiner to grasp has not been sufficiently studied.

In view of the above circumstances, the imaging position of each fundus image, and an object to provide a is not easy bur image processing program grasping the examiner through the list display using the reduced image of the fundus image.

An image processing program according to the first aspect of the present disclosure is executed by a processor of an image processing device, and thereby a reduced image of a captured image of an eye to be inspected is based on information indicating an imaging position of the captured image with respect to the eye to be inspected. A list display step of displaying a list on a display screen of a display device, a wide-area image display step of displaying on the display screen a subject eye image captured in a wider range than each of the captured images, and the captured image An image processing program for causing the image processing apparatus to execute an analysis processing step of performing an analysis process and obtaining an analysis result of the captured image , wherein the list display step displays the object to be displayed displayed in the wide area image display step. At the photographing position of each photographed image in the optometry image, a reduced image of each photographed image is superimposed on the eye image to be examined. Arrangement is, when any one of said reduced image displayed on the display screen in the list display step is selected by the examiner, the selection captured image is the captured image corresponding to the reduced image selected The image processing apparatus is further caused to execute a selected photographed image display step of displaying the analysis result of the selected photographed image while displaying at a display magnification higher than that of the reduced image.
An image processing program according to the second aspect of the present disclosure is executed by a processor of an image processing device, thereby reducing a reduced image of a captured image of the subject's eye based on information indicating the imaging position of the captured image with respect to the subject's eye. A list display step for displaying a list on the display screen of the display device, and a wide-area image display step for displaying on the display screen a subject eye image captured in a wider range than each of the captured images. An image processing program to be executed, wherein in the list display step, a reduced image of each photographed image is placed at the photographing position of each photographed image in the eye image displayed by the wide-area image display step. Each of the reduced images displayed on the display screen is further selected by the examiner. Accordingly, the image processing step of performing image processing on the captured image corresponding to the reduced image selected, to be executed by the image processing apparatus.
An image processing program according to the third aspect of the present disclosure is executed by a processor of an image processing device, thereby reducing a reduced image of a captured image of the subject's eye based on information indicating the imaging position of the captured image with respect to the subject's eye. A list display step for displaying a list on the display screen of the display device, and a wide-area image display step for displaying on the display screen a subject eye image captured in a wider range than each of the captured images. An image processing program to be executed, wherein in the list display step, a reduced image of each photographed image is placed at the photographing position of each photographed image in the eye image displayed by the wide-area image display step. Each of the images displayed on the display screen is displayed on the display screen to indicate whether or not the image processing has already been performed. The additional information display step of adding to the reduced image, to be executed by the image processing apparatus.

According to the present invention, there is an effect that it is easy for an examiner to grasp the imaging position of each fundus image through a list display using a reduced image of the fundus image .

It is the block diagram which showed schematic structure of PC which is one Embodiment of this invention. It is the block diagram which showed schematic structure of the ophthalmic imaging device in this embodiment. It is the schematic diagram which showed the controller displayed on a monitor. It is the schematic diagram which showed the thumbnail list window. It is the schematic diagram which showed the wide area | region list window. It is a flowchart which shows an analysis data generation process. It is a flowchart which shows a part of image adjustment process. It is a flowchart which shows the continuation of FIG. 7 of an image adjustment process. It is a figure for demonstrating the position alignment of the image in an image adjustment process. It is the schematic diagram which showed the ROI setting window. It is a flowchart which shows a ROI setting process. It is a schematic diagram which shows a photoreceptor cell correction window. It is a schematic diagram which shows a follow-up window. It is an example of the transformation pattern of the display mode of a photoreceptor cell.

  Hereinafter, exemplary embodiments of the present invention will be described. First, a schematic configuration of a personal computer 1 (hereinafter referred to as “PC1”) which is an image processing apparatus according to the present embodiment will be described with reference to FIG.

  In the present embodiment, the PC 1 acquires an image of the eye to be inspected captured by the ophthalmic imaging apparatus 100 via at least one of a network and an external memory. The PC 1 processes the acquired image. However, it is not limited to the PC 1 that can operate as an image processing apparatus. For example, the ophthalmic imaging apparatus 100 itself may process an image captured by the ophthalmic imaging apparatus 100. In this case, the ophthalmic imaging apparatus 100 operates as an image processing apparatus.

  As shown in FIG. 1, the PC 1 includes a CPU 2. The CPU 2 is a processing device (processor) for executing various processes of the PC 1. Connected to the CPU 2 are a ROM 3, a RAM 4, an HDD 5, a communication I / F 6, a display control unit 7, an operation processing unit 8, and an external memory I / F 9 via a bus.

  The ROM 3 is a nonvolatile storage medium that stores a program such as BIOS. The RAM 4 is a volatile storage medium that temporarily stores various types of information. The HDD 5 (hard disk drive 5) is a non-volatile storage medium. Note that another storage medium such as a flash ROM may be used as the nonvolatile storage medium. The HDD 5 stores an image processing program for processing an image of the eye to be examined. For example, in the present embodiment, a program that causes the PC 1 to execute the processes shown in the flowcharts of FIGS. 6 to 8 and 11 is stored in the HDD 5. Further, the HDD 5 stores image data captured by the ophthalmic imaging apparatus 100. In the following description, for the sake of convenience, it is assumed that data of an image obtained by imaging the same eye to be examined is stored in the HDD 5 or the like.

  The communication I / F 6 connects the PC 1 to an external device such as the ophthalmic imaging apparatus 100. The PC 1 according to the present embodiment can acquire image data captured by the ophthalmic imaging apparatus 100 via the communication I / F 6. In the present embodiment, an image acquired via the communication I / F 6 is stored in the HDD 5. The external memory I / F 9 connects the external memory 15 to the PC 1. For the external memory 15, various storage media such as a USB memory and a CD-ROM can be used. The PC 1 according to the present embodiment can also acquire data of an image captured by the ophthalmic imaging apparatus 100 via the external memory 15. For example, the user attaches the external memory 15 to the ophthalmic imaging apparatus 100 and stores data of an image captured by the ophthalmic imaging apparatus 100 in the external memory 15. Next, the user attaches the external memory 15 to the PC 1 and causes the PC 1 to read image data stored in the external memory 15. As a result, the PC 1 acquires image data captured by the ophthalmic imaging apparatus 100.

  Here, with reference to FIG. 2, a schematic configuration of the ophthalmic imaging apparatus 100 of the present embodiment will be described. In the present embodiment, a case where a laser optometry apparatus with wavefront compensation (AO-SLO) is used as the ophthalmic imaging apparatus 100 will be described. The ophthalmic imaging apparatus 100 includes a fundus imaging optical system 101, a wavefront sensor 102, a wavefront compensation device 103, a target presentation optical system 104, and a second imaging unit 105. The detailed configuration of the ophthalmologic imaging apparatus 100 can be exemplified by the content described in JP2013-070941A.

  The fundus imaging optical system 101 scans the illumination light beam (laser light) two-dimensionally on the fundus of the eye to be examined. Further, the fundus imaging optical system 101 receives reflected light (reflected light beam) of the illumination light beam reflected by the fundus and acquires an image of the eye to be examined (that is, a fundus image). Thereby, the fundus imaging optical system 101 images the fundus with high resolution (high resolution) and high magnification. In the present embodiment, it is assumed that imaging is performed at an angle of view of about 1.5 degrees in order to perform observation at the cell level. The fundus imaging optical system 101 can change the imaging region by moving the scanning range of the illumination light beam up, down, left, and right. In the present embodiment, the fundus imaging optical system 101 continuously performs imaging within the same range. In the present embodiment, for example, the fundus imaging optical system 101 continuously captures for about 3 seconds and acquires about 150 fundus images. That is, in the fundus imaging optical system 101, one image group composed of a plurality of continuous still images is acquired by a series of imaging. Note that the fundus imaging optical system 101 can be configured, for example, as a scanning laser ophthalmoscope using a confocal optical system.

  Data of an image captured by the fundus imaging optical system 101 is acquired by the PC 1 by the method described above. The image data acquired by the PC 1 includes gradation information and coordinate information as data for forming an image. In addition, in the present embodiment, the image data includes, for example, the ID of the eye to be examined, a time stamp indicating the imaging date and time of the image, information indicating the imaging site on the fundus, and the fixation target presentation position at the time of imaging The information etc. which are shown are included.

  In the ophthalmologic imaging apparatus 100 according to the present embodiment, when a fundus image is captured, the wavefront aberration due to the eye to be examined is compensated using the wavefront sensor 102 and the wavefront compensation device 103. The wavefront sensor 102 is an element for detecting wavefront aberration including low-order aberration and high-order aberration. In the present embodiment, the wavefront sensor 102 receives a reflected light beam reflected from the fundus and detects a wavefront aberration caused by the eye to be examined. As the wavefront sensor 102, for example, a Hartmann Shack detector, a wavefront curvature sensor that detects a change in light intensity, or the like can be used.

  The wavefront compensation device 103 relays illumination light irradiated to the eye to be examined by the fundus imaging optical system 101. At this time, the wavefront compensation device 103 deforms the reflection surface of the illumination light based on the detection result of the wavefront sensor 102, for example. Accordingly, the wavefront compensation device 103 controls the wavefront of the illumination light to compensate for wavefront aberration caused by the eye to be examined. As the wavefront compensation device 103, for example, a reflective LCOS (Liquid Crystal On Silicon), a deformable mirror, or the like can be used.

  The optotype presenting optical system 104 presents a fixation target to the eye to be examined when the ophthalmologic imaging apparatus 100 images the fundus. In the ophthalmologic imaging apparatus of the present embodiment, the optotype presenting optical system 104 can switch the fixation target presenting position. In the present embodiment, the target presentation positions are set to nine locations, three in each of the vertical and horizontal directions of the eye to be examined. The range in which illumination light can be irradiated on the fundus is changed by switching the presentation position of the visual target and guiding the line of sight of the eye to be examined. Note that the visual line of the eye to be examined may be guided by moving the fixation target in the visual target presenting optical system 104.

  The second imaging unit 105 acquires a fundus image (that is, a wide area image) having a wider angle of view than the fundus imaging optical system 101. The fundus image acquired by the second imaging unit 105 is used as an image for designating or confirming the imaging position of the fundus by the fundus imaging optical system 101, for example. The second imaging unit 105 can use an existing fundus camera observation / imaging optical system, a scanning laser ophthalmoscope (SLO) optical system, and the like. Although details will be described later, the apparatus according to the present embodiment causes the PC 1 to acquire not only a group of images captured by the fundus imaging optical system 101 but also a fundus image captured by the second imaging unit 105. At this time, the PC 1 acquires at least a fundus image having a wide angle of view captured with the same target position as the fixation target and the image group acquired by the PC 1. Thereby, a fundus image having a wide angle of view can be stored in the HDD 5 of the PC 1 and the external memory 15.

  Returning to FIG. 1, the description of the image processing apparatus 1 will be continued. The display control unit 7 controls the display on the monitor 13. The operation processing unit 8 is connected to an operation unit 14 (for example, a keyboard, a mouse, etc.). The operation processing unit 8 detects an operation of the operation unit 14 by the user and outputs an operation signal to the CPU 2. Thereby, the user's operation on the operation unit 14 is received by the CPU 2. In the present embodiment, an external monitor 13 and an operation unit 14 are used. However, at least a part of the monitor 13 and the operation unit 14 may be incorporated in the PC 1.

  In the present embodiment, the operation of the operation unit 14 by the user is performed on various GUIs displayed on the monitor 13. As a type of GUI, the monitor 13 displays a controller 20 that mainly accepts user operations.

  Here, a schematic configuration of the controller 20 will be described with reference to FIG. A setting window 30 is displayed on the controller 20. The setting window 30 has a data list 31 and a control box 32. The data list 31 displays a list of images stored in the HDD 5 and the external memory 15. The data list 31 displays a list of file names (for example, “image group 1” and “image group 2”) of each image group. In the present embodiment, the image acquired by the fundus imaging optical system 101 acquired by the PC 1 from the ophthalmic imaging apparatus 100 is managed with one file name assigned to each image group. As described above, each image group includes a plurality of still images captured continuously by the fundus imaging optical system 101.

  As shown in FIG. 3, in the data list 31, a list of file names of analysis result images generated from the image group is displayed under the file name of the image group. For example, in FIG. 3, “analysis result image 1a” and “analysis result image 1b” generated from “image group 1” are displayed under “image group 1”. Although details will be described later, the analysis result image is a still image used for photoreceptor cell analysis of the eye to be examined. Thus, in this embodiment, a list of file names of the image group and the analysis result image is displayed in the data list 31.

  A check box 31 a is provided at the head of each file name displayed in the data list 31. In the apparatus of the present embodiment, a user's check operation (for example, click with a mouse) performed on each check box 31a is accepted. The user can select an image to be displayed or the like by checking the check box 31a.

  The control box 32 has a plurality of buttons and input fields. Although details will be described later, a window is expanded in the controller 20 or a parameter used for photoreceptor cell analysis is changed according to a user's operation on the buttons and the input field. For example, when the “display” button 32b is operated in a state where any image in the data list 31 is selected, an image display window 40 as shown in FIG.

  In the image display window 40, the image selected in the data list 31 is displayed. In the present embodiment, as shown in FIG. 3, when a plurality of images are selected in the data list 31, any one image is displayed in the image display window 40. At this time, the image displayed in the image display window 40 can be switched by operating the “page feed” button 41. When an analysis result image is selected in the data list 31, the result of the analysis process performed using the analysis result image is displayed in the window 40. When an image group is selected in the data list 31, each still image included in the image group is continuously displayed in the order in which it was captured. That is, a moving image is displayed on the image display window 40. Note that when the “display” button 32 b is operated in a state where a plurality of images are selected in the data list 31, a plurality of selected images may be displayed on the monitor 13.

  Next, the operation of the PC 1 will be described with reference to FIG.

<Select image>
As described above, based on a user operation on the check box 31 a of the data list 31, the CPU 2 selects an image to be subjected to image processing such as analysis or display processing on the monitor 14. In addition, the PC 1 of the present embodiment also has a method for selecting an image used for processing such as analysis. For example, an image used for processing such as analysis can be selected from the thumbnail list window 50 shown in FIG. 4 and the wide area list window 60 shown in FIG.

  In the thumbnail list window 50 shown in FIG. 4, a list of thumbnail images (an example of an image index) of an image group whose file names are shown in the data list 31 is displayed. When an operation for the “read” button 32 a of the control box 32 is accepted by the apparatus, the thumbnail list window 50 is displayed by the CPU 2. When the “display type switching” button 51 is operated by the user, the type of thumbnail image displayed on the screen is switched by the CPU 2. For example, when the “display type switching” button 51 is operated while the thumbnail images of the image group are displayed as shown in FIG. 4, the thumbnail images displayed in the thumbnail list window 50 are the analysis result images. Switch to thumbnail image. In the present embodiment, a reduced image of any one of the still images included in the image group is used as the thumbnail image indicating the image group on which the photoreceptor cell analysis processing has not been performed. For example, a reduced image of the first image captured first in each image group can be used as a thumbnail image. In addition, for example, a reduced image of the analysis result image generated in the analysis process is used as the thumbnail image of the image group in which the photoreceptor cell analysis processing has already been performed. The thumbnail image may be stored in advance in the HDD 5 or the like. Further, when the “read” button 32a is operated, the PC 1 may create a thumbnail image.

  As shown in FIG. 4, the CPU 2 causes the thumbnail list window 50 to display information indicating the history of photoreceptor cell analysis processing (for example, presence or absence of analysis processing) together with the thumbnail image. In FIG. 4, the display of “analyzed” indicates that the analysis of the image group corresponding to the displayed thumbnail image has already been performed, and the analysis result is stored in the HDD 5 or the like. The display of “unanalyzed” indicates that the analysis of the image group corresponding to the displayed thumbnail image has not been performed. Thereby, when the thumbnail list window 50 is displayed, the user can easily grasp whether or not the analysis processing has been performed on the image group indicated by the thumbnail image. In this embodiment, the display of “analyzed” and “unanalyzed” is always displayed together with each thumbnail image, but only one of the displays of “analyzed” and “unanalyzed” is displayed. It may be displayed.

  The wide area list window 60 shown in FIG. 5 is switched from the thumbnail list window 50 when the “display format switching” button 52 is operated in the thumbnail list window 50. The CPU 2 displays the fundus image W with a wide angle of view (wide area fundus image W) over the entire thumbnail display area of the wide area list window 60. The wide fundus image W is a fundus image captured by the second imaging unit 105 of the ophthalmic imaging apparatus 100. As described above, in the present embodiment, the wide fundus image W is an image acquired in advance by the PC 1 from the ophthalmic imaging apparatus 100.

  The CPU 2 displays the thumbnail image superimposed on the wide area fundus image W. At this time, the CPU 2 determines the position at which the thumbnail image is arranged according to the positional relationship of the imaging parts of the image (image group or analysis result image) indicated by the thumbnail image. For example, in FIG. 5, the CPU 2 matches the position where the thumbnail images of each image group are arranged in the wide fundus image W with the imaging part (imaging range) of each image group. The alignment of the thumbnail image with respect to the fundus image having a wide angle of view can be performed based on, for example, information indicating the imaging region on the fundus included in each image data. In addition, as information indicating the imaging region in the fundus, for example, position information with respect to the image center of a fundus image having a wide angle of view, position information with respect to the macula, and the like can be used.

  In the present embodiment, when there are a plurality of image groups obtained by imaging the same part, the thumbnail images of the respective image groups are displayed on the same position of the wide-area fundus image W. When a plurality of thumbnail images are overlapped, the file name or the like (an example of an image index) of the image group indicated by each thumbnail image is displayed around the thumbnail images. By performing a user selection operation (for example, clicking with a mouse) on the file name on the screen, the CPU 2 can select individual image groups even when a plurality of thumbnail images overlap.

  As shown in FIG. 5, the wide area list window 60 is provided with check boxes 61a to 61c for selectively displaying thumbnail images according to the analysis status. In FIG. 5, the “display all” check box 61 a is selected. When the “analyzed only” check box 61b is selected, the CPU 2 displays a thumbnail image indicating the analyzed image group. In this case, the CPU 2 does not display thumbnail images indicating unanalyzed image groups. When the “unanalyzed only” check box 61c is selected, the CPU 2 displays a thumbnail image indicating an unanalyzed image group. In this case, the CPU 2 does not display thumbnail images indicating the analyzed image group. For this reason, in the PC 1, the user can easily select one of the unanalyzed image group and the analyzed image group.

  Also, in the wide area list window 60, thumbnail image list display is performed for each fixation target presenting position that was presented when the image group was captured. In the wide area list window 60, a fixation target position selection / display box 62 is provided. In the present embodiment, the fixation target position selection box 62 has check boxes for a total of 9 squares, 3 squares each in the vertical and horizontal directions. The check boxes of 9 squares correspond to the fixation target presentation positions in the visual target presentation optical system 104 of the ophthalmic imaging apparatus 100, respectively. The user can designate a thumbnail image displayed on the screen by checking (selecting) any of the check boxes. When any one of the check boxes is checked (selected), the CPU 2 displays a thumbnail image of the image group captured at the fixation position corresponding to the check position on the screen. For example, as shown in FIG. 5A, when the check box at the center of the nine squares is checked, the thumbnail images of the image group captured when the fixation target presentation position is the center are displayed. The Also, as shown in FIG. 5B, when the check box on the right side of the upper row of 9 squares is checked, the thumbnail of the image group captured when the fixation target is presented at the upper right presentation position. An image is displayed.

  Further, as shown in FIGS. 5A and 5B, in this embodiment, the wide area fundus image W displayed in the wide area list window 60 is also selected in conjunction with the check on the check box. The CPU 2 displays a wide area fundus image W imaged at a fixation position corresponding to a check position on the wide area list window 60 from a plurality of fundus images previously captured at each fixation position.

  The “display type switching” button 63 has the same role as the “display type switching” button 51 of the thumbnail list window 50. Further, when the “display format switching” button 64 is operated by the user in the wide area list window 60, the display is switched to the thumbnail list window 50 by the CPU 2.

  As described above, in the wide area list window 60 of the present embodiment, on the wide area fundus image W, thumbnail images (an example of an image index) indicating each image group or analysis result image at a position corresponding to the imaging position of each image. ) Are arranged. For this reason, it is easy for the user to know which position of the eye to be imaged the image group indicated by the thumbnail image. For this reason, the user can easily select an image group or the like used for image processing.

  In addition, even when the imaging position of the eye to be examined is the same, there are cases where it is desired to separately handle image groups or the like having different fixation target presentation positions. For example, in AO-SLO, even if two or more images are taken of the same part, the contents of each image may be different if the positions of the fixation target at the time of capturing each image are different. On the other hand, in the wide area list window 60, thumbnail images are switched and displayed for each fixation target position at the time of imaging such as an image group indicated by each thumbnail image. Therefore, in the PC 1, even if image groups and the like captured at different fixation positions are stored in the HDD 5 and the like, it is easy for the user to select a desired image group and the like.

  In addition, you may display the image group etc. from which the fixation target presentation position at the time of imaging overlaps on one wide area wide fundus image W. In this case, for example, the position of each image in the wide fundus image W is determined using information indicating the fixed position of the fixation target included in the image data and information indicating the imaging position.

  In the present embodiment, when the thumbnail image display is switched for each fixation position, the wide fundus image W displayed in the wide area list window 60 is captured at the same fixation position as the image group indicated by the thumbnail image. Switch to the selected image. Therefore, it is possible to make the user more appropriately grasp the imaging position of the image group or the like indicated by the thumbnail image.

  In the wide area list window 60, as in the thumbnail list window 50, the CPU 2 displays information indicating the history of photoreceptor cell analysis processing together with thumbnail images. Therefore, the PC 1 can make the user easily understand whether or not the analysis processing has been performed on the image group indicated by the thumbnail image.

  In the present embodiment, the thumbnail image and the file name are exemplified as the image index indicating the image (image group or analysis result image), but other icons for specifying the image such as an icon and a shooting date, image quality, reliability, etc. Information may be used for the image index.

<Image analysis>
When one or more image groups are selected by the above-described method, when the “analysis” button 32c (see FIG. 3) of the control box 32 is operated by the user, the CPU 2 performs analysis data generation processing (see FIG. 6). ) Is executed. In the analysis data generation process, an addition average image is generated from each image group. Further, an analysis process is performed on each of the addition average images, and an analysis result is derived.

  Here, the analysis data generation processing will be described with reference to FIG. First, the CPU 2 selects, as an object to be processed this time, an image group that has not been processed by the analysis data generation process among the image groups selected in the data list or the like (S11). Here, the image group selected in the processing of S11 is represented by data set L (image set L) = [L0, L1,..., LN]. Each image is denoted by Ln. However, Ln indicates that an image is captured at an earlier time as the value of the subscript n is smaller.

  Next, the CPU 2 executes image adjustment processing (S12). In the present embodiment, in the image adjustment process, among a plurality of images included in the image group selected by the process of S11, a part or all of the images captured by the ophthalmic imaging apparatus 100 when the fixation is stable. A reference image (first reference image) is generated from the image. In the image adjustment processing (S12) of the present embodiment, the reference image is used as a template when correcting a difference between images by superimposing a part or all of the images in the image group. Although details will be described later, the images in which the difference between the images is adjusted by the image adjustment processing are added and averaged by the subsequent processing of S13.

  Here, the image adjustment processing will be described with reference to FIGS.

  First, the CPU 2 executes the processing from S21 to S28, and performs rough alignment between the images included in the image group (that is, the data set L) selected in the processing of S11. Here, the rough alignment is alignment performed without correcting at least distortion of each image. In this embodiment, rough alignment is performed by translating the fundus image Ln. However, the rough alignment is not limited to parallel movement, and may be, for example, rotational movement and a combination of parallel movement and rotational movement. Coarse alignment is performed on the black image E. The size of the black image E is a horizontal width MW and a vertical width MH. Here, W is the horizontal width of the fundus image Ln, and H is the vertical width of the fundus image Ln. M is a constant of 1 or more (for example, M = 3), and defines a range in which positional deviation between images is allowed. Further, the CPU 2 stores each image after the coarse alignment and the shift amount (details will be described later) of each image in the RAM 4. A data set of each image subjected to coarse alignment is denoted by G = [G0, G1,..., GN].

  In the process of S21, the CPU 2 performs initial setting of the reference image R (second standard image) (S21). The reference image R is used as a reference when the fundus image Ln is roughly aligned. Although details will be described later, in the present embodiment, the reference image R is updated every time the fundus image Ln is aligned with the reference image R. Hereinafter, the reference image updated n times is denoted as Rn. The reference image R0 that is initially set in the present embodiment is an image in which the image L0 is arranged on the black image E so that the center of gravity of the black image E overlaps with the black image E, as shown in FIG.

  After executing S21, the CPU 2 repeatedly executes the processes from S22 to S28 to roughly align the images included in the data set L one by one with the reference image Rn. First, the image Ln that has not been subjected to rough alignment and has the earliest imaging time is selected by the CPU 2 as an image to be aligned next (S22). For example, when the image Lk is aligned in the process of S22 to S28 performed immediately before, the image Lk + 1 is selected by the CPU 2 in the next process of S22. In the process of S22 immediately after the process of S21 is executed, the image L0 is selected by the CPU2.

  In the next process of S23, the image Ln selected in the immediately preceding process of S22 (hereinafter referred to as “selected image”) is aligned with the reference image Rn by translation (S23). Various image processing methods can be used as the alignment method. For example, a method is conceivable in which the selected image Ln is shifted pixel by pixel with respect to the reference image Rn, and the selected image Ln is aligned with the position where the data of the two images most closely matches (the position with the highest correlation). In addition, after extracting common feature points from the reference image Rn and the selected image Ln, a method of aligning the selected image Ln at a position where the feature points overlap each other can be considered.

  In the present embodiment, alignment is performed by sequentially calculating a correlation value between the selected image Ln and the reference image Rn while shifting the selected image Ln by one pixel unit with respect to the reference image Rn. Here, the maximum correlation value is 1, and the larger the value, the higher the correlation between images. Next, the CPU 2 generates an image Gn by duplicating the selected image Ln moved to the position where the correlation with Rn is the highest on the black image E (S24). For example, when aligning the image L0, the image L0 completely matches the fundus image portion included in the reference image R0 at the time of initial setting. Therefore, the image G0 is the same as the reference image R0 (see FIG. 9A). 9B, in the image G1, the image L1 is moved to the black image E after the L1 is moved so that the overlapping range of the fundus image portion L′ 0 and the image L1 in the image G0 overlaps. Has been duplicated. The generated image Gn is stored in the RAM 4 by the CPU 2.

  At this time, the CPU 2 obtains the center-of-gravity position cn of the selected image Ln moved to the position where the correlation with Rn is the highest (S24). Further, the CPU 2 stores the positional deviation amount (shift amount) dn = [dxn, dyn] of the selected image Ln in the RAM 4 (S25). In the present embodiment, the positional deviation amount dn indicates the displacement of the imaging range between the selected image Ln and the selected image Ln−1 captured immediately before Ln. Since the displacement of the imaging range is caused by fixation micromotion, the positional deviation amount dn is the magnitude and direction of fixation micromotion that occurs between the time when the selected image Ln−1 is captured and the time when the selected image Ln is captured. Represents. Therefore, based on the shift amount dn, the CPU 2 can detect the movement of the eye to be examined during imaging. In the present embodiment, the positional deviation amount dn of the selected image Ln is a displacement between the selected image Ln and the image Ln−1 captured at a time before Ln, but is captured at a time later than Ln. It can also be a displacement from the image. dxn and dyn respectively indicate a horizontal direction component and a vertical direction component in the positional deviation amount. In the present embodiment, the positional deviation amount dn can be obtained, for example, from the difference between the center of gravity position cn and the center of gravity position cn−1 acquired in advance.

  Next, the CPU 2 updates the reference image Rn (S27). In the present embodiment, the CPU 2 generates an updated reference image Rn + 1 from the reference image Rn and the selected image Gn after rough alignment. For example, the addition average image of the reference image Rn and the image Gn can be used as the updated reference image Rn + 1. In this case, the gradation value rn + 1 of the pixel at an arbitrary position in the updated reference image Rn + 1 can be expressed by the following equation (1), for example.

rn + 1 = {(n × rn) + gn} / n + 1 (1)
However, rn, r0, and gn indicate the tone values of the pixels at the same position as the arbitrary position in the reference image Rn, the reference image R0 at the time of initial setting, and the image Gn, respectively. As a result, as shown in FIG. 9C, the fundus image portion R′n of the reference image Rn and the fundus image portion L′ n of the image Gn are added and averaged to the updated reference image Rn + 1.

  Next, the CPU 2 determines whether or not the alignment of all the images included in the data set L has been completed (S28). If there is an uncompleted image in the data set L (S28: No), the CPU 2 returns to the process of S22 and repeatedly executes the processes of S22 to S28. On the other hand, when the alignment of all the images included in the data set L is completed (S28: Yes), the process proceeds to S29. Note that, in the present embodiment, in the processing from S22 to S28, among the images included in the data set L, alignment is performed from an image with an earlier imaging time to a reference image, but from an image with an later imaging time to a reference image. May be aligned.

  In the process of S29, the CPU 2 divides the data set G = [G0, G1,..., Gn] of the coarsely aligned image group and temporally corresponds to the fixation state of the eye to be examined. A plurality of data sets F1 = [G0, G1,..., Ga], F2 = [Ga + 1,..., Gb],. Here, a method of dividing the data set G in the present embodiment will be described. In the present embodiment, the data set is divided using the shift amount dn obtained in the process of S26. For example, in the present embodiment, the shift amount dn corresponding to the image Gn included in the data set G is integrated in the order of subscripts (that is, in the order of capturing the images Ln). Note that, as described above, the shift amount dn represents the size of the position shift of the imaging range caused by fixation fixation that occurs while the ophthalmic imaging apparatus 100 captures two consecutive images. Therefore, the integrated value S indicates the amount of positional deviation in the imaging range due to fixation fixation from a certain point in time. When the integrated value S of the deviation amount gn exceeds a predetermined threshold Θ, the data set Fm composed of the image Gn in which the integrated value S includes the deviation amount gn is divided from the data set G. The integrated value of the deviation amount is initialized (to zero), and the same processing is repeated for the remaining data sets G, so that a plurality of data sets F1, F2,. ., Fq is generated. Here, it is considered that the number of images Gn included in the data set Fm indicates the degree of fixation stability of the eye to be examined when the image Gn included in the data set Fm is captured. It is considered that the number of images that can be captured by the ophthalmic imaging device 100 increases as the fixation is more stable, that is, as the temporal change in the imaging range is smaller, while the imaging range is shifted by Θ. Because. Therefore, the data sets F1, F2,... Fq generated by the process of S29 respectively correspond temporally to the fixation state of the eye to be examined. In the present embodiment, the threshold Θ is set to about 1/8 of the size of the image Ln (that is, (Θx, Θy) ≈ (W / 8, H / 8)). However, the threshold Θ can be set as appropriate in relation to the required accuracy. In the present embodiment, it is assumed that the data set is divided when either the x component or the y component of the integrated value S exceeds the threshold value Θx or Θy. In the present embodiment, the positional deviation amount dn is the displacement of the imaging range of two images captured successively, but is not limited to this. For example, the displacement of the imaging range between the selected image and the first image of the data set F including the selected image may be used. In this case, the data set can be divided when the positional deviation amount dn exceeds the threshold Θ.

  Turning to FIG. 8, the description of the flowchart will be continued. Next, the CPU 2 selects the data set Fs including the most images from the data sets F1, F2,... Fq created by the process of S29 (S30). Therefore, in the process of S30, a plurality of images captured when the fixation of the eye to be examined is most stable are selected.

  Next, the CPU 2 acquires the gravity center position C of the data set Fs selected in the process of S30 (S31). The centroid position C can be obtained from the centroid position cn of the fundus image portion of each image Gn included in the data set Fs. For example, the center-of-gravity position C can be obtained by dividing the integrated value of the center-of-gravity cn of each image by the number of images.

  Incidentally, each image Gn included in the data set Fs has a width MW and a height MH in the present embodiment. In the process of S32, the CPU 2 cuts out each image Gn included in the data set Fs with a horizontal width W and a vertical width H around the center of gravity position C (S32). As a result, a data set Oa = [Oa1, Oa2,..., Oap] composed of images On having a width W and a height H is created.

  Next, the CPU 2 averages each image included in the data set Oa to generate a reference image Ob (S33). In the reference image Ob, distortion caused by fixation micromotion included in each image of the data set Oa is averaged. The reference image Ob is used as a template for image processing in the next matching process (S34).

  Next, the CPU 2 executes a fitting process (S34). In the matching process, the CPU 2 corrects distortion of each image included in the data set Oa using the reference image Ob as a reference (template). Various methods can be used to correct the distortion. For example, a local region of each image included in the data set Oa is converted so as to match the reference image Ob. Such correction methods are described in the literature (e.g. A. Dubra, & Z. Harvey, Registration of 2D Images from Fast Scanning Ophthalmic Instruments. Carlos. O. & S. sorzano et al, Elastic Registration of Biological Images Using Vector-Spline Regularicalization etc.). As a result, an image data set O = [O1, O2,... After executing the matching process, the process returns to the analysis data generation process (see FIG. 6), and the CPU 2 continues the process from S13.

  Returning to FIG. 6, the description will be continued. The CPU 2 adds and averages the images included in the image data set O to create a still image (S13).

  Next, the CPU 2 executes an optical distortion correction process (S14). Thereby, the optical image distortion of the eye to be examined and the ophthalmic imaging apparatus 100 is corrected from the still image created in the process of S13.

  Next, the CPU 2 executes a reliability acquisition process (S15). In the process of S15, the CPU 2 acquires the reliability of the still image whose image distortion has been corrected. The reliability is information indicating the reliability (or validity) of the analysis result derived from the analysis using the still image. The reliability may be information indicating whether the image is highly reliable, or may be information indicating the degree of reliability (for example, a numerical value). The reliability is a guideline for the user to select an image to be observed and compared. In general, the higher the still image quality, the higher the reliability. Therefore, for example, the CPU 2 can acquire the reliability from information such as the contrast and brightness of the still image. For example, the greater the contrast, the higher the reliability. Thus, for example, the CPU 2 may acquire the reliability using the contrast distribution of the still image.

  By the way, the factors that deteriorate the image quality of still images (factors that reduce the reliability) include those caused by imaging conditions such as fixation fixation and device settings, pupil diameter, eye aberration, and intermediate transparency. Those caused by individual differences in the eye to be examined such as turbidity of the light body are included. If the low reliability is caused by the situation at the time of imaging, the imaging may be performed again. On the other hand, when the low reliability is caused by the individual difference of the eye to be examined, there are cases where the user wants to select an image used for observation / comparison even for a still image with low reliability. Therefore, for example, based on at least one of the pupil diameter, aberration information, and information indicating the opacity of the intermediate translucent body, the CPU 2 acquires the reliability considering the individual difference of the eye to be examined in the process of S15. You may let them. In the present embodiment, as information indicating the pupil diameter and the degree of opacity of the intermediate translucent body, values measured in advance by an ophthalmologic apparatus other than the ophthalmologic apparatus 100 can be used. Further, the degree of turbidity of the intermediate translucent body can be obtained from the profile of the PSF (point spread function) image at the photographing position. In this case, for example, a PSF image at the same photographing position as the fundus image acquired by the ophthalmic imaging apparatus 100 can be transferred to the PC 1 in advance.

  Next, the CPU 2 executes photoreceptor cell analysis processing (S16). In the photoreceptor cell analysis process of the present embodiment, the CPU 2 detects photoreceptor cells from the still image that has been corrected by the optical distortion correction process (S14). A photoreceptor cell point is set to the detected photoreceptor cell with respect to the still image corrected by the optical distortion correction process (S14). Thereby, in this embodiment, an analysis result image is created. The analysis result image may be an image that can be used for analysis, inspection, comparison with another image, and the like, and the photoreceptor cell point does not necessarily have to be set. Also, using the analysis result image, the photoreceptor density, hexagonal cell appearance rate, regular hexagonal cell appearance rate, etc. in the entire analysis result image are calculated. The image data of the analysis result image and the various analysis results calculated are stored in the HDD 5 (S17).

  After execution of S17, the CPU 2 determines whether or not all the image groups selected by the user from the data list 31 or the like have been processed (S18). When an unprocessed image group remains (S18: No), the CPU 2 returns to the process of S11 and repeatedly executes the processes of S11 to S18. On the other hand, if all the image groups have been processed (S18: Yes), the CPU 2 ends the analysis data generation process.

  As described above, in the PC 1 according to the present embodiment, the data set Fs (image set Fs) including a plurality of eye images that are continuously captured when the fixation is stable is combined with the reference image. Therefore, it is acquired (S30). The plurality of eye images captured continuously when the fixation is stable has fewer differences between the images than the eye images captured when the fixation is unstable. For this reason, it is easy to generate a good reference image Ob by combining a plurality of eye images included in the data set Fs acquired in the process of S30. For example, the distortion in the direction along the retina of the eye to be examined and the distortion in the direction intersecting the retina due to fixation micromotion are easily suppressed in the reference image Ob. Therefore, it is easy to appropriately perform image processing (for example, distortion correction and alignment of the eye image to be examined) using the reference image Ob generated by the PC 1 as a template. Therefore, according to the PC 1, it is possible to obtain a reference image Ob suitable as a template for image processing.

  Further, in the PC 1 of the present embodiment, for the data set Fs including a plurality of eye images that are aligned with each other in the process of S23, an area that is combined with the reference image (range to be cut out from each eye image). Set by the CPU 2 (S32). Since the positional deviation of the region combined with the reference image in each eye image in the data set Fs is corrected, a good reference image Ob is easily generated in the PC1.

  Further, an area combined with the reference image is set in the CPU 2 around the center of gravity position of the data set Fs that has been aligned by the process of S23 (S32). For this reason, in each eye image to be examined included in the data set Fs, the region combined with the reference image tends to be widened. Therefore, a better reference image is likely to be generated. Note that the region combined with the reference image is not limited to the region set around the center of gravity of the data set Fs as in the present embodiment.

  In the process of S31, at least a data set having the largest number of eye images taken continuously is acquired from the plurality of data sets F1, F2,... Fq. The data set in which the fixation of the eye to be examined is more stable at the time of imaging increases the number of eye images to be examined included in the image set. For this reason, it is easy to create a good reference image from a data set having the largest number of eye images to be examined continuously.

  By the way, if each of the test eye images is aligned with the reference image Rn (second reference image) in the process of S23, if there is a large difference between the imaging positions of the test eye image Ln and the reference image Rn, the test eye There is a possibility that the image Ln and the reference image Rn are not properly aligned. For example, when there are few overlapping areas in the eye image Ln and the reference image Rn, the reliability of alignment is lowered. Therefore, it may be difficult to generate a good reference image.

  On the other hand, in this embodiment, every time one eye image is aligned, the reference image Rn is updated using the aligned eye image Ln. Therefore, the reference image Rn includes information on a plurality of eye images Ln to be examined having different imaging positions. For this reason, it is easy to ensure an overlapping area between the eye image Ln to be examined and the reference image Rn. Therefore, the alignment of the eye image Ln to be examined with respect to the reference image Rn is easily performed well.

  The reference image Rn may be updated using at least one of the predetermined number of eye images Ln each time a predetermined number of eye images are aligned. In this case, the update frequency of the reference image Rn can be reduced as compared with the present embodiment. Therefore, the reference image is easily generated in a short time.

  Further, in the present embodiment, when the eye image Ln to be examined is aligned with the reference image Rn (S23), the reference image Rn is continuously updated with the eye image Ln in the previous update (S27). The captured eye image Ln-1 is included. For this reason, the eye images Ln-1 and Ln that are continuously captured are not easily affected by the shift of the imaging position due to the fixation movement. For this reason, the area | region which overlaps with the to-be-tested eye image Ln and the reference image Rn is easy to be ensured. Therefore, in PC1, it is easier to better align the eye image Ln to be examined with respect to the reference image Rn.

<Setting ROI>
The PC 1 of the present embodiment is provided with functions for correcting and reanalyzing analysis data using the analysis result image created by the above analysis data generation process (see FIG. 6). For example, the reanalysis can be performed after changing the analysis target in the analysis result image according to the user's instruction. For example, the user can specify a range (that is, ROI: Region of Interest) used for analysis in the analysis result image and perform reanalysis. When one or more analysis result images are selected in the data list 31 or the like and the “ROI setting” button 32d of the control box 32 is operated, the CPU 2 displays an ROI setting window 70 as shown in FIG. Display.

  As shown in FIG. 10, in the ROI setting window 70, the user can specify the ROI on the analysis result image displayed in the image display area T. The CPU 2 sets the ROI within the range specified by the user. In the present embodiment, the ROI set range is indicated by being surrounded by a one-dot chain line.

  When a plurality of analysis result images are selected in advance in the data list 31 or the like, the CPU 2 displays another selected analysis result image in the image display region T based on the operation of the “page feed” button 71. . When the “analysis” button 72 is operated, the CPU 2 reanalyzes the analysis result image selected in the data list 31 or the like. In the present embodiment, in the reanalysis, processing similar to the photoreceptor cell analysis processing included in the analysis data generation processing (see FIG. 6) is performed. When the ROI is set in the analysis result image, the fundus tissue included in the ROI is an analysis target. Therefore, according to PC1, by setting a ROI by excluding a part where a photoreceptor cell is difficult to detect (for example, a blood vessel or the like), an appropriate analysis result can be easily obtained by reanalysis. The same reanalysis is performed even if another “analysis” button prepared in the control box 32 or the like is operated.

  In the present embodiment, when a plurality of analysis result images are selected in advance in the data list 31 or the like, ROIs can be set for a plurality of images at once. When the ROI is instructed from the analysis result image by the user, the CPU 2 executes ROI setting processing (see FIG. 11).

  In the ROI setting process shown in FIG. 11, first, the CPU 2 sets an ROI within a range designated by the user for the analysis result image being displayed (S40). Next, the CPU 2 determines whether or not a plurality of analysis result images are selected in advance in the data list 31 or the like (S41). When only one analysis result image is selected (S41: No), the CPU 2 skips the processing from S42 to S46 and ends the ROI setting processing. On the other hand, when a plurality of analysis result images are selected in advance (S41: Yes), the process proceeds to S42.

  In the process of S42, the CPU 2 selects one image that has not been processed in S43 and later (S42). Next, the CPU 2 determines whether or not the image in which the ROI is set in the process of S40 and the image selected in the process of S42 are captured at the same fixation target presentation position (S43). . This determination can be performed, for example, by comparing information indicating the fixation position of the fixation target included in each image data. When the present position of the fixation target at the time of imaging is different in each image (S43: No), the process proceeds to S46 described later.

  On the other hand, when the present position of the fixation target at the time of imaging is the same (S43: Yes), the process proceeds to S44. In the process of S44, the CPU 2 determines whether the fundus tissue in the ROI set in the process of S40 is included in the image selected in the process of S42 (S44). For example, the determination in S44 is performed based on the sequentially calculated correlation value while shifting the region in the ROI in the analysis result image being displayed by at least one of translation and rotation with respect to the image selected in S42. Can do. For example, when the maximum correlation value exceeds a predetermined threshold value, it is determined that the fundus tissue in the ROI set in the process of S40 is included in the image selected in the process of S42. In the process of S44, the determination in S44 may be performed based on the pattern matching result of the feature points extracted from the region in the ROI in the analysis result image being displayed and the image selected in the process of S42. it can.

  In the process of S44, when the fundus tissue in the ROI set in the analysis result image being displayed is not included in the image selected in the process of S42 (S44: No), the process proceeds to S46. To do. On the other hand, when the fundus tissue in the ROI set in the process of S40 is included in the image selected in the process of S42 (S44: Yes), the CPU 2 changes the image selected in the process of S42. On the other hand, the ROI is set (S45). In the process of S45, the ROI is set for the same part as the part where the ROI is set in the currently displayed analysis result image.

  In the process of S46, the CPU 2 determines whether or not all the images currently selected in the data list 31 or the like have been processed (S46). If the image currently selected in the data list 31 or the like includes an image that has not been subjected to the processing of S43 and subsequent steps (S46: No), the process proceeds to S42, and the processing of S42 and subsequent steps is executed again. On the other hand, if the processing from S43 onward is performed on all the images selected in the data list 31 or the like (S46: Yes), the processing is terminated.

<Correction of photoreceptor detection results>
Moreover, in PC1 of this embodiment, the object analyzed in an analysis result image can be changed also by correcting the detection result of the photoreceptor cell in an analysis result image. Correction of the photoreceptor cell detection result is performed on the photoreceptor cell correction window 80 shown in FIG. When one or more analysis result images are selected in the data list 31 or the like and the “photocell point correction” button 32 e of the control box 32 is operated, the CPU 2 displays the photocell point correction window 80 on the controller 20. .

  As shown in FIG. 12A, any of the analysis result images selected in the data list 31 or the like is displayed in the image display area T in the photoreceptor cell correction window 80. The role of the “page forward” button 81 is the same as that of the other “page forward” buttons described above.

  As shown in FIG. 12B, when the “add” button 83 is operated by the user, the user can specify the position to add the photoreceptor cell point in the analysis result image. When the user designates the position to add the photoreceptor cell point, the CPU 2 sets the icon Ia at the position designated by the user on the analysis result image. On the other hand, as shown in FIG. 12C, when the “delete” button 84 is operated by the user, the user can designate the photoreceptor cell point to be deleted in the analysis result image. When the photoreceptor cell point to be deleted is instructed by the user, the CPU 2 sets the icon Ib at the position of the photoreceptor cell point deleted by the user.

  When the “analysis” button 82 is operated in a state where either the icon Ia or the icon Ib is set, the CPU 2 corrects the photoreceptor cell point of the displayed analysis result image. For example, the CPU 2 adds a photoreceptor point at the position where the icon Ia is set. On the other hand, the CPU 2 deletes the photoreceptor cell point at the position where the icon Ib is set. The image in which the photoreceptor cell point is corrected is stored in the HDD 5 as a new analysis result image. Further, the CPU 2 performs a process similar to the photoreceptor analysis process included in the analysis data generation process (see FIG. 6) on the new analysis result image. As a result, various analysis results for the new analysis result image are output.

  Similar to the case where a plurality of analysis result images are selected and the ROI is set, even when a plurality of analysis result images are selected and the detection result of the photoreceptor cell point is corrected, one analysis result image is selected. The CPU 2 may reflect the correction of the performed photoreceptor cell point to another analysis result image having the same photoreceptor cell point.

<Change of axial length information>
In the PC 1 of the present embodiment, reanalysis of the eye to be examined can be performed by changing information on the axial length of the eye to be examined. The user inputs the axial length of the eye to be examined in the axial length input box 32g (see FIG. 3) of the control box 32, selects one or more analysis result images in the data list 31 or the like, and then clicks the “Analysis” button. 32c is operated. As a result, the CPU 2 reanalyzes the selected analysis result image based on the axial length input to the axial length input box 32g. Here, processing similar to the photoreceptor cell analysis processing included in the analysis data generation processing (see FIG. 6) is performed with the changed axial length. By changing the information about the axial length of the eye to be examined, the estimated value of the size of the photoreceptor cell is changed in the photoreceptor analysis process. Therefore, by inputting an accurate axial length measured by an axial length measuring device or the like into the axial length input box 32g, a more accurate analysis result such as visual cell density can be obtained.

  In the present embodiment, the case where reanalysis is performed by changing the axial length of the eye to be examined has been described, but other methods may be adopted as long as the correct visual cell size can be used in the reanalysis. For example, when reanalyzing, the corneal curvature radius of the eye to be examined may be changed. For example, it is possible to allow the user to input the corneal curvature radius as if the axial length was input to the axial length input box 32g in the present embodiment. It should be noted that only one of the axial length and the corneal curvature radius may be changed, but if both can be changed, a more accurate visual cell size can be used in reanalysis. Therefore, in this case, a more appropriate analysis result can be obtained.

<Follow-up display>
The PC 1 of the present embodiment is provided with a follow-up display function that can display images with different imaging dates and times on the same screen (that is, simultaneously) on one screen. The follow-up display allows the user to compare how the specific position of the fundus has changed over time.

  The user can set the base image by operating the “base image setting” button 32f of the control box 32 (see FIG. 3). The base image is an image used as a reference for comparison. In the present embodiment, first, the user operates the base image setting button 32 and then selects a file name of any analysis result image arranged in the data list 31. Thereby, the image of the file name selected by the user is set as the base image by the CPU 2. Next, the user checks one or more check boxes 31a for analysis result images other than the base image. Thereby, the image checked by the user is set by the CPU 2 as a comparative image to be displayed simultaneously with the base image. The base image and the comparison image may be selected from the wide area list window 60 shown in FIG. Since the thumbnail images of the analysis result images captured at the same imaging position are displayed at the same position in the wide area fundus image W, the user can easily select the base image and the comparison image.

  When the “analysis” button 32c is operated by the user with the base image and the comparison image set, the CPU 2 displays a follow-up display window 90 (see FIG. 13). As shown in FIG. 13, the follow-up display window 90 displays a base image and at least one of comparison images. In FIG. 13, the analysis result image 2 is displayed as a comparative image. The comparison image is displayed in the follow-up display window 90 in a state of being aligned with the base image. The alignment is performed by the CPU 2 rotating and translating the comparison image with respect to the base image. For example, when the correlation between the base image and the comparison image is obtained, the comparison image may be moved to a position where the correlation value is highest.

  In the follow-up display window 90, the analysis results of the base image and the comparison image are displayed by the CPU 2, respectively. For example, the analysis result of each image stored in the HDD 5 in advance may be displayed. However, if the imaging positions of the base image and the comparative image are shifted, it is difficult to accurately compare the analysis results such as the photoreceptor cell density between the base image and the comparative image. Therefore, in this embodiment, a common ROI may be set for the base image and the comparison image, and reanalysis may be performed. For example, as in the ROI setting window 70 (see FIG. 10) described above, the user may be able to specify the ROI on the base image of the follow-up display window 90 (or on the comparison image). For example, when the “analysis” button 92 is operated after the ROI is specified for the base image, the CPU 2 executes the above-described ROI setting process for the base image and the comparison image. Thereby, ROI can be set to a common area | region with a base image and a comparison image. An area in which a common fundus tissue is shown may be searched, and an ROI may be set in an area common to each image. When a common ROI is set for each of the base image and the comparison image, the CPU 2 obtains an analysis result such as a cell density and outputs the analysis result to a screen or the like. This makes it easier for the user to compare analysis results such as the photoreceptor cell density between the base image and the comparative image.

  Further, as shown in FIG. 13, the follow-up display window 90 displays the reliability of each image. Using the reliability as a guide, the user can select an image to be compared. Therefore, it is easy for the user to properly compare the images.

  Further, as shown in FIG. 14, in the base image and the comparative image, hexagonal cells may be displayed in a manner different from cells having other shapes. For example, hexagonal cells may be colored differently from cells of other shapes, or may be hatched. In the normal retina, hexagonal cells are regularly arranged. It is known that the corners of hexagonal cells increase and decrease in the process of losing shape due to abnormalities such as lesions. Or you may display a regular hexagonal cell in the aspect different from cells other than a regular hexagonal cell among hexagonal cells.

  As described above, in the PC 1 according to the present embodiment, when the ROI setting window 70, the follow-up window 90, and the like are displayed, the target to be analyzed by the photoreceptor cell analysis process is changed in a plurality of fundus images. The CPU 2 receives an instruction from the user via the operation unit 14 and the operation processing unit 8. When an object related to an instruction received by the CPU 2 is included in an overlapping portion of a plurality of images (S44: Yes), when a photoreceptor cell analysis process is performed, each of the plurality of images is received from the user. The analysis result reflecting the instruction is output by the CPU 2. As described above, the objects to be analyzed in the plurality of images are collectively changed in accordance with an instruction from the user. Therefore, when analyzing a plurality of images having overlapping portions at least in part of each other, the burden on the user who gives an instruction to change the analysis target is easily suppressed. In particular, in the present embodiment, the CPU 2 receives an instruction from the user using any one of the images displayed on the ROI setting window 70, the follow-up window 90, and the like. For this reason, the user can easily give an instruction to change the analysis target.

  In this embodiment, when the ROI instructed by the user is received by the CPU 2, the fundus tissue in the ROI set for each of the plurality of images is analyzed by the CPU 2. Therefore, analysis within a range desired by the user can be performed on each of the plurality of images while suppressing the burden on the user.

  As mentioned above, although demonstrated based on embodiment, this invention is not limited to the said embodiment, It can deform | transform variously.

  For example, in the above-described embodiment, the movement of the eye to be inspected at the time of imaging is detected by causing the CPU 2 to obtain the shift amount of the imaging position between images that are continuously captured. However, the movement of the eye to be examined during imaging may be detected by other methods. For example, while one image group is imaged by the fundus imaging optical system 101 of the ophthalmologic imaging apparatus 100, the second imaging unit 105 also continuously captures a wide fundus image (frontal image of the fundus). The PC 1 is caused to acquire a wide-range fundus image continuously captured together with the image group. In addition, the CPU 2 may detect the movement of the subject's eye at the time of capturing the image group based on the movement of the characteristic part such as the blood vessel and the macula indicated by the wide fundus image captured continuously. Further, when the eye to be examined moves, the aberration detected by the wavefront sensor 102 changes. Thus, for example, while one image group is captured by the fundus imaging optical system 101, the ophthalmic imaging apparatus 100 sequentially acquires aberrations detected by the wavefront sensor 102 (mainly wavefront aberrations due to the eye to be examined). deep. The PC 1 is caused to acquire an aberration detection result while the image group is captured together with the image group. In addition, the PC 1 may cause the CPU 2 to detect the movement of the eye to be examined at the time of capturing the image group based on the detection result of the aberration. Whichever method is used, it is possible to detect the movement of the eye to be examined at the time of capturing an image group without requiring a special device.

  In the analysis data generation process of the above embodiment, a set of analysis data (analysis result image and analysis data such as photoreceptor cell density) is obtained from one image group. However, a set of analysis data may be created from a plurality of image groups. For example, when a plurality of image groups obtained by imaging the same position of the eye to be examined are selected in the data list 31 or the like on the same imaging day, the selected plurality of image groups are regarded as one image group and analyzed. Data generation processing (see FIG. 6) may be executed. If there are a plurality of pieces of analysis data for the same imaging date, it is difficult for the user to determine which data to use. On the other hand, when analysis data generation processing is performed by regarding a plurality of image groups captured at the same position of the eye to be examined as one image group on the same imaging date, Analysis data can be obtained when is stable.

  In addition, the reference image Ob1 generated from the first image group may be used as a template for performing alignment or the like of the second image group having a different shooting date and time from the first image group. In this case, for example, the CPU 1 corrects the distortion after the data set f2 in which fixation is stable in the second image group is aligned with the reference image Ob1. Further, the CPU 2 cuts out the data set f2 within the range of the reference image Ob1, and generates an analysis result image. Thereby, since the analysis result image generated from the first image group and the analysis result image generated from the second image group use the same reference image as a template, the user can compare the analysis results. Easy to do.

  In the above-described embodiment, the case where the reliability of the image is obtained with respect to the image subjected to the averaging process in S13 has been described, but the present invention is not limited to this. For example, you may obtain | require the reliability of the image before adding an image.

Further, in the above embodiment, when ROIs are set for a plurality of analysis result images by the ROI setting process (see FIG. 11), the ROI set region set by the user for one analysis result image is displayed. CPU2 also searched from other analysis result images.
In addition, the CPU 2 sets an ROI in a region searched in another analysis result image. However, the CPU 2 does not necessarily need to search the other analysis result image for the region where the ROI set by the user for one analysis result image is set. For example, if the difference in the imaging range between the analysis result images is sufficiently small, the CPU 2 sets the ROI of another analysis result image at the same position (coordinates) on the image as the ROI instructed by the user. May be.

  In the above-described embodiment, the case where the fundus image captured by the AO-SLO is processed in the PC 1 as the ophthalmic imaging apparatus 100 has been described. However, according to the present invention, in addition to AO-SLO, images captured by various devices capable of capturing the eye to be examined can be processed by the PC 1. For example, an optical tomographic interferometer (OCT) that acquires a tomographic image of the anterior segment or the fundus may be used as the ophthalmic imaging apparatus 100.

1 PC
2 CPU
5 HDD
13 Monitor 15 External memory Ob Reference image W Wide area fundus image

Claims (3)

By being executed by the processor of the image processing apparatus,
A list display step of displaying a reduced image of a captured image of the subject's eye on a display screen of a display device based on information indicating a photographing position of the subject image with respect to the subject's eye;
A wide-area image display step for displaying on the display screen a test eye image captured in a wider range than each of the captured images;
An analysis process for performing an analysis process on the captured image and obtaining an analysis result in the captured image;
An image processing program for causing the image processing apparatus to execute
In the list display step, a reduced image of each photographed image is placed on the eye image to be superposed at the photographing position of each photographed image in the eye image displayed by the wide-area image display step. ,
When any one of the reduced images displayed on the display screen in the list display step is selected by an examiner, a selected captured image that is the captured image corresponding to the selected reduced image is selected from the reduced image. An image processing program for causing the image processing apparatus to further execute a selected photographed image display step of displaying the analysis result of the selected photographed image while displaying at a high display magnification. By being executed by the processor of the image processing apparatus,
A list display step of displaying a reduced image of a captured image of the subject's eye on a display screen of a display device based on information indicating a photographing position of the subject image with respect to the subject's eye;
An image processing program for causing the image processing apparatus to execute a wide-area image display step for displaying on the display screen a subject eye image captured in a wider range than each of the captured images,
In the list display step, a reduced image of each photographed image is placed on the eye image to be superposed at the photographing position of each photographed image in the eye image displayed by the wide-area image display step. ,
Furthermore, an image processing step of performing image processing on the captured image corresponding to the selected reduced image by selecting any one of the reduced images displayed on the display screen by the examiner, An image processing program to be executed by an image processing apparatus. By being executed by the processor of the image processing apparatus,
A list display step of displaying a reduced image of a captured image of the subject's eye on a display screen of a display device based on information indicating a photographing position of the subject image with respect to the subject's eye;
An image processing program for causing the image processing apparatus to execute a wide-area image display step for displaying on the display screen a subject eye image captured in a wider range than each of the captured images,
In the list display step, a reduced image of each photographed image is placed on the eye image to be superposed at the photographing position of each photographed image in the eye image displayed by the wide-area image display step. ,
An image processing program for causing the image processing apparatus to execute an additional information display step of adding a display indicating whether or not the image processing has already been performed to each of the reduced images displayed on the display screen.