JP2021062101A - Image processing device, image processing method, and image processing program - Google Patents

Abstract

To accurately detect a candidate of a vortex vein.SOLUTION: An image processing device comprises: an acquisition unit for acquiring chorioidea vessel data generated from eyeground image data; a determination unit which sets a first region in the eye ground image data and determines the position of a vortex vein on the basis of the feature amount of the chorioidea vessel data around the first region; and an output unit for outputting the position of the vortex vein.SELECTED DRAWING: Figure 4

Description

The present invention relates to an image processing apparatus, an image processing method, and an image processing program.

A method for extracting a crossing / branching site of a blood vessel from a fundus image is disclosed (see Patent Document 1 below). It is required to detect the position of the vortex vein from the fundus image.

Japanese Unexamined Patent Publication No. 2011-056068

The image processing apparatus of the first disclosure technique sets an acquisition unit for acquiring choroidal blood vessel data generated from fundus image data and a first region in the fundus image data, and choroidal blood vessel data around the first region. It is provided with a determination unit for determining the vortex vein position based on the feature amount of the above, and an output unit for outputting the vortex vein position.

The image processing method of the second disclosure technique is an image processing method executed by an image processing apparatus having a processor that executes a program and a storage device that stores the program, and the processor generates from fundus image data. Acquisition process for acquiring the choroidal blood vessel data, and determination process for setting the first region in the fundus image data and determining the vortex vein position based on the feature amount of the choroidal blood vessel data around the first region. And the output process of outputting the decision result by the decision process.

The image processing program of the third disclosure technique sets an acquisition process for acquiring choroidal blood vessel data generated from the fundus image data and a first region in the fundus image data in the processor, and sets the first region in the fundus image data around the first region. A determination process for determining the position of the vortex vein and an output process for outputting the determination result by the determination process are executed based on the feature amount of the choroidal blood vessel data.

FIG. 1 is an explanatory diagram showing a system configuration example of an ophthalmic system. FIG. 2 is an explanatory diagram showing an example of binarized image data obtained by binarizing the fundus image data. FIG. 3 is a block diagram showing a hardware configuration example of a computer (image processing device and terminal). FIG. 4 is a block diagram showing a functional configuration example of the image processing device. FIG. 5 is a flowchart showing an example of a procedure for detecting a vortex vein by an image processing device. FIG. 6 is an explanatory diagram showing an example of an output screen in the output device. FIG. 7 is an enlarged view of a part of the search area A of the binarized image data C according to the first embodiment. FIG. 8 is an explanatory diagram showing a scan result in which the region of interest P is raster-scanned in the exclusion region B in the search region A. FIG. 9 is a flowchart showing a detailed processing procedure example 1 of the vortex vein position determining process (step S503) shown in FIG. FIG. 10 is an enlarged view of a part of the search area A of the binarized image data C according to the second embodiment. FIG. 11 is a flowchart showing a detailed processing procedure example 2 of the vortex vein position determining process (step S503) shown in FIG. FIG. 12 is an enlarged view of a part of the search area A of the binarized image data C according to the third embodiment. FIG. 13 is a flowchart showing a detailed processing procedure example 3 of the vortex vein position determining process (step S503) shown in FIG.

<System configuration example of ophthalmic system>
FIG. 1 is an explanatory diagram showing a system configuration example of an ophthalmic system. The ophthalmology system 100 includes an ophthalmology device 101, an image processing device 102, and a terminal 103. The ophthalmic apparatus 101, the image processing apparatus 102, and the terminal 103 are communicably connected via a network 104 such as a LAN (Local Area Network), a WAN (Wide Area Network), or the Internet.

The ophthalmic apparatus 101 images the patient's eye 105 to be inspected and generates fundus image data 110. The ophthalmologic apparatus 101 is, for example, a scanning laser ophthalmoscope (SLO) that scans a laser beam on the eye 105 to be inspected and generates fundus image data 110 based on the reflected light from the fundus. The ophthalmic apparatus 101 transmits the generated fundus image data 110 to the image processing apparatus or the terminal.

The eye 105 to be inspected has a large number of choroidal blood vessels 106 inside. The choroid is a thin film constituting the eyeball wall of the eye 105 to be inspected, and the choroidal blood vessel 106 is a blood vessel existing in the choroid. A vortex vein 107 exists in the deep part of the choroid (near the boundary between the choroid and the sclera) where a large number of choroidal blood vessels 106 having different thicknesses are present. Anatomically, the vortex vein 107 is a vascular site where the choroidal blood vessels 106 are concentrated, and is a drainage channel for blood that has flowed into the eyeball. The vortex vein 107 is part of the choroidal blood vessel 106. There are 3 to 7 vortex veins 107, which are located around the fundus (near the equator of the eyeball). It is assumed that the ophthalmic apparatus 101 has an angle of view capable of photographing the area around the fundus where the vortex vein 107 exists.

The image processing device 102 detects the vortex vein 107 by image analysis from the fundus image data 110 from the ophthalmologic device 101. The terminal 103 displays the detection result from the image processing device 102. Although the ophthalmic apparatus 101, the image processing apparatus 102, and the terminal 103 have been described as separate apparatus, at least two of them may be configured by one apparatus.

<Example of binarized image data>
FIG. 2 is an explanatory diagram showing an example of binarized image data obtained by binarizing the fundus image data 110. The binarized image data C is image data obtained by binarizing the fundus image data 110 by the image processing device 102. The horizontal direction of the binarized image data C is the X direction, and the direction orthogonal to the X direction is the Y direction. In the present specification, the choroidal blood vessel 106 is indicated by white pixels and the background is indicated by black pixels, but the choroidal blood vessel 106 may be indicated by black pixels and the background may be indicated by white pixels.

In FIG. 2, the search area A is a part of the binarized image data C, and the pixel P is the center position. The blood vessel pixel region D, which is the white pixel region in FIG. 2, is the detection result of the choroidal blood vessel 106 by the binarization process (sometimes referred to as the blood vessel detection result D).

The blood vessel pixel region D includes choroidal blood vessels 106, vortex veins 107, and erroneously detected pixels. In addition, some choroidal blood vessels 106 and vortex veins 107 may not be detected due to reasons such as low contrast of the fundus image data 110.

<Computer hardware configuration example>
FIG. 3 is a block diagram showing a hardware configuration example of a computer (image processing device 102 and terminal 103). The computer 300 has a processor 301, a storage device 302, an input device 303, an output device 304, and a communication interface (communication IF) 305. The processor 301, the storage device 302, the input device 303, the output device 304, and the communication IF 305 are connected by the bus 306.

The processor 301 controls the computer 300. The storage device 302 serves as a work area for the processor 301. Further, the storage device 302 is a non-temporary or temporary recording medium for storing various programs and data. Examples of the storage device 302 include a ROM (Read Only Memory), a RAM (Random Access Memory), an HDD (Hard Disk Drive), and a flash memory.

The input device 303 inputs data. The input device 303 includes, for example, a keyboard, a mouse, a touch panel, a numeric keypad, and a scanner. The output device 304 outputs data. Examples of the output device 304 include a display and a printer. The communication IF305 connects to the network and transmits / receives data.

<Example of functional configuration of image processing device 102>
FIG. 4 is a block diagram showing a functional configuration example of the image processing device 102. The image processing device 102 includes an image acquisition unit 401, a blood vessel acquisition unit 402, a determination unit 403, and an output unit 404. Specifically, the image acquisition unit 401, the blood vessel acquisition unit 402, the determination unit 403, and the output unit 404 are realized by causing the processor 301 to execute a program stored in the storage device 302 shown in FIG. 3, for example. Will be done.

The image acquisition unit 401 acquires the fundus image data 110. The source of the fundus image data 110 may be an ophthalmic apparatus 101, a terminal 103, or a computer (not shown). The blood vessel acquisition unit 402 detects the blood vessel pixel region D of the choroidal blood vessel 106 from the fundus image data 110. The determination unit 403 swirls the region of interest P based on the state of the region of interest P in the blood vessel detection result D by the blood vessel acquisition unit 402 and the remaining blood vessel pixel region in the blood vessel detection result D excluding the region of interest P. Determine as a vein candidate. The output unit 404 outputs the determination result by the determination unit 403. The determination result is the coordinate position of the vortex vein candidate in the binarized image data C. In addition, the determination result may include information indicating whether or not the coordinate position is the position of the blood vessel detection result D.

FIG. 5 is a flowchart showing an example of a detection processing procedure for the vortex vein 107 by the image processing device 102. The image processing device 102 acquires the fundus image data 110 by the image acquisition unit 401 (step S501).

Next, the image processing device 102 acquires the blood vessel pixel region D of the choroidal blood vessel 106 from the fundus image data 110 by the blood vessel acquisition unit 402 (step S502). Specifically, for example, the blood vessel acquisition unit 402 emphasizes the grayscale fundus image data 110 and then performs binarization processing by a method such as setting a predetermined threshold value to convert it into binarized image data.

For example, the blood vessel acquisition unit 402 may be converted into binarized image data C by binarizing the brightness value of each pixel of the fundus image data 110, based on the prediction result obtained by deep learning. It may be converted into binarized image data C. Then, the blood vessel acquisition unit 402 detects the blood vessel pixel region of the binarized image data C as the blood vessel pixel region D of the choroidal blood vessel 106.

Next, the image processing device 102 executes the vortex vein position determination process by the determination unit 403 (step S503). The vortex vein position determination process (step S503) is a process of determining the position of the vortex vein 107 from the blood vessel detection result D. The details of the vortex vein position determination process (step S503) will be described later. The blood vessel detection result D can be said to be choroidal blood vessel information indicating the network and structure of the choroidal blood vessels such as the traveling direction of the choroidal blood vessels, the blood vessel diameter (thickness of the blood vessels), the area of the blood vessels, and the branching / merging of the blood vessels. It is possible to detect the vortex vein position with high accuracy by combining not only the blood vessel traveling direction but also the thickness of the blood vessel and the branch point / confluence point of the blood vessel.

Next, the image processing device 102 outputs the data related to the vortex vein 107 by the output unit 404 (step S504). Specifically, for example, the output unit 404 outputs the data related to the vortex vein 107 to the output device 304 of the image processing device 102 so that the data can be displayed on the terminal 103. Further, the output unit 404 transmits data regarding the vortex vein 107 to the terminal 103. In this case, the terminal 103 outputs the data related to the vortex vein 107 so as to be displayable by the output device 304 of the terminal 103. The data relating to the vortex vein 107 includes data in which the position of the vortex vein 107 is specified in the blood vessel detection result D. Further, the data regarding the vortex vein 107 may include data such as an ID for identifying a plurality of vortex veins and a distance and an angle from the macula, the optic nerve head, and the like.

<Output screen example>
FIG. 6 is an explanatory diagram showing an example of an output screen in the output device 304. The output screen 600 has a first display area 601, a second display area 602, and a third display area 603. In the first display area 601, the grayscale fundus image 611 obtained in the current shooting (2019/07/16) and the binarized image 612 obtained in the current shooting are displayed. The binarized image 612 includes an image 613 of the vortex vein 107. In the second display area 602, a grayscale fundus image 621 obtained in the previous shooting (2019/04/16) and a binarized image 622 obtained in the previous shooting are displayed. The binarized image 622 includes an image 623 of the vortex vein 107.

By comparing the fundus images 611 and 621 with the binarized images 612 and 622, the user binarizes the images 613 and 623 of the vortex vein 107 and their positions, which were not visible in the fundus images 611 and 621. It can be visually recognized in images 612 and 622.

Information about follow-up is displayed in the third display area. The information regarding the follow-up includes the image of the partially enlarged region 631 of the fundus image 611, the image of the partially enlarged region 632 of the binarized image 612, the image of the partially enlarged region 641 of the fundus image 621, and the partially enlarged image of the binarized image 622. The image of region 642 is included. Partial enlargement areas 631, 632, partial enlargement area 641 and partial enlargement area 642 include the position of the vortex vein 107, and the partial enlargement area 632, 642 displays images 613 and 623 of the vortex vein 107.

In addition, the information regarding the follow-up includes the follow-up information 630 of the blood vessel diameter. The blood vessel diameter is the diameter of the choroidal blood vessel 106. The ampulla portion 61 is a blood vessel site where one end of a plurality of choroidal blood vessels 106 is aggregated, that is, a vortex vein 107. The peripheral portion 62 is a plurality of choroidal blood vessels 106 connected to the ampulla portion 61. Since there are a plurality of choroidal blood vessels 106 constituting the peripheral portion 62, the blood vessel diameter of the peripheral portion 62 may be a statistical value (for example, an average value, a maximum value, a minimum value, a median value) of the plurality of choroidal blood vessels 106. Further, the measurement position of the blood vessel diameter is specified, for example, by operating the input device.

Examples 1 to 3 of the vortex vein position determination process (step S503) will be described below. Each of Examples 1 to 3 is a vortex based on the state of the region of interest P in the blood vessel detection result D by the blood vessel acquisition unit 402 and the remaining blood vessel pixel region in the blood vessel detection result D excluding the region of interest P. This is an example of embodying the process of determining the position of a blood vessel.

Example 1 shows Example 1 of the vortex vein positioning process (step S503). Example 1 of the vortex vein position determination process (step S503) is based on the number in which the region of interest P in the blood vessel detection result D and the remaining blood vessel pixel region in the blood vessel detection result D excluding the region of interest P are connected. This is the process of determining the position of the vortex vein.

FIG. 7 is an enlarged view of a part of the search area A of the binarized image data C according to the first embodiment. The search area A is a rectangular area centered on the area of interest P in the binarized image data C. In FIG. 7, 45 × 35 pixels are used as an example. The region of interest P is one or more pixels. In FIG. 7, it is 1 × 1 pixel. The exclusion area B is a rectangular area centered on the area of interest P. The exclusion area B is an area that includes the area of interest P and is included in the search area A. In FIG. 7, 9 × 9 pixels are used as an example. The shape of the exclusion region B is not limited to a rectangle, and may be a polygon other than a rectangle. In this embodiment, the exclusion area B is a single area, but it may be composed of a plurality of rectangular areas. In this case, the point P does not have to be included in the exclusion region.

The region of interest P is raster-scanned in the binarized image data C. Therefore, the search area A and the exclusion area B centered on the area of interest P also follow the raster scan of the area of interest P. The stride of the region of interest P (the distance between the region of interest P before movement and the region of interest P after movement) is such that the exclusion region B before movement and the exclusion region B after movement of the region of interest P overlap or are adjacent to each other. It may be an interval.

As shown in FIG. 2, the remaining white pixel connecting region DW in the blood vessel detection result D excluding the region of interest P before the exclusion region B is set is one. The white pixel connection area DW is an area in which a plurality of white pixels are connected to each other. Here, the connection of pixels of the same color (here, white) means up and down (Y direction) with respect to the pixel of interest. ) When considering whether it is the same color as the surrounding 4 pixels adjacent to the left and right (X direction), and whether it is the same color as the surrounding 8 pixels including the 4 pixels diagonally above right, diagonally above left, diagonally below right, and diagonally below left There are cases to consider. In this embodiment, unless otherwise specified, a method of considering the surrounding 8 pixels with respect to the white pixel of interest is adopted as the definition of connection. In FIG. 2, since all the white pixels are connected, the number of white pixel connecting regions DW is 1.

When the exclusion region B is set as shown in FIG. 7, the blood vessel pixel region D is changed. That is, the white pixel connecting region DW, which was one in FIG. 2, is divided by the exclusion region B, and is divided into five white pixel connecting regions DW of Fa, Fb, Fc, Fd, and Fe as shown in FIG. To. These five white pixel connecting regions DW are connected to the exclusion region B. Therefore, the number of white pixel connecting regions DW connected to the exclusion region B in the region of interest P in FIG. 7 is 5 of Fa, Fb, Fc, Fd, and Fe.

In this way, the determination unit 403 moves the exclusion region B by scanning the region of interest P, and at the position of the region of interest P after the movement, the white pixel connection region DW connected to the exclusion region B. Performs the process of counting the number of. By scanning all of the blood vessel pixels (white pixels) in the search area A, the number of connections can be obtained for each pixel of the blood vessel pixels in the search area A.

FIG. 8 is an explanatory diagram showing a scan result in which the region of interest P is raster-scanned in the exclusion region B in the search region A. The position of each pixel in the scan result 800 corresponds to the position of each pixel in the search area A. In the scan result 800, the black pixels indicate the background of the search area A (pixels indicating the fundus region other than the blood vessels), and the white pixels indicate the pixels of the blood vessel detection result D (pixels indicating the blood vessels) in the search area A.

The numerical value in the pixel is the number N2 of the white pixel connection region DW in the search region A connected to the exclusion region B when the pixel becomes the region of interest P. In the case of FIG. 7, N2 = 5. The determination unit 403 is based on the difference value N obtained by subtracting the number N1 of the white pixel connection region DW of the blood vessel detection result D in the search region A in the state where the exclusion region B is not set from the number N2. P is determined as a vortex vein candidate. In the case of FIG. 7, N1 = 1 (see FIG. 2). Therefore, the difference value N is N2-N1 = 4. The difference value N is the amount of change in the number of white pixel connected regions DW before and after the exclusion region B is set. That is, the larger the amount of change, the easier it is to be determined as a vortex vein candidate.

In the scan result 800, the determination unit 403 determines the region of interest P as a vortex vein candidate based on the difference value N. Specifically, for example, the determination unit 403 determines the pixel having the maximum difference value N as the vortex vein candidate. In the example of FIG. 8, since the numerical value N2 = 6 in the pixel becomes the maximum value and the difference value N = 5, the pixel is determined as a vortex vein candidate. That is, the coordinate position of the pixel is the position of the vortex vein candidate. This indicates that the position of the pixel with a large amount of change has many choroidal blood vessels around the pixel, and it is presumed that the position is likely to satisfy the characteristics of the vortex vein where multiple choroidal blood vessels merge. Because it can be done.

Further, the determination unit 403 may determine a pixel whose difference value N is equal to or greater than the first threshold value as a vortex vein candidate. In the example of FIG. 8, when the first threshold value is "4", the difference value N of the pixels having the numerical values N2 = 5 and 6 in the pixel is N = 4 and 5, so that the pixel is a vortex vein. It will be decided as a candidate. Further, the determination unit 403 may determine pixels having a difference value N up to the upper nth (n is an arbitrary integer of 1 or more) as a vortex vein candidate. In the example of FIG. 8, since there are 6 types of numerical values N2 = 6,5,4,3,2,1 in the pixel, the difference value N is 6 of N = 5,4,3,2,1,0. Kind. In the case of n = 3, since the difference value N = 3 of the pixel of the numerical value N2 = 4 in the pixel is the difference value N up to the top three, the numerical values "4", "5" and "6" in the pixel Pixels are determined as vortex vein candidates.

Further, when there are a plurality of pixels that are candidates for the vortex vein, the determination unit 403 counts the total number of white pixels that are connected to the pixels of the vortex vein candidate. The narrowing down of vortex vein candidates may be performed based on the total number. Specifically, when the total number of white pixels connected to the pixels of the vortex vein candidate is not equal to or more than the second threshold value, the determination unit 403 excludes the pixels of the vortex vein candidate from the vortex vein candidates. The second threshold value may be equal to or greater than the number of pixels in the exclusion area B.

By setting the second threshold value as described above, the exclusion region B corresponds to the size of a general vortex vein 107, so that the detection accuracy of the vortex vein 107 can be improved. The determination unit 403 may determine the region of interest P as a vortex vein candidate simply based on the numerical value N2 in the pixel instead of the difference value N.

The pixel of the region of interest P that is a candidate for a vortex vein is not limited to the white pixel that constitutes the blood vessel detection result D, and may be a black pixel that constitutes a background that is not the blood vessel detection result D. Since the blood vessel detection result D depends on the accuracy of the blood vessel detection process, the region to be detected as the vortex vein 107 may not be detected in the blood vessel detection result D. Therefore, the black pixel may be determined as a vortex vein candidate.

<Swirl vein position determination process (step S503)>
FIG. 9 is a flowchart showing a detailed processing procedure example 1 of the vortex vein position determining process (step S503) shown in FIG. The image processing device 102 designates an undesignated region of interest P in the binarized image data C by the determination unit 403 (step S901). Step S901 is a process of designating the position of the raster scan described above. Step S909: When Yes, the determination unit 403 sets the area to the right of the region of interest P (when the region of interest P is the right end of the binarized image data C, the left end when striding in the Y direction) to the next. It is designated as the region of interest P.

The determination unit 403 may specify the region of interest P by limiting it to white pixels. As a result, the vortex vein candidate can be determined as a white pixel. In this case, since the raster scan of the black pixel can be skipped, the raster scan can be speeded up. Further, when the region of interest P is set at or near the end of the binarized image data C, a search range having a preset size may not be set in the binarized image data C. In such a case, it is not necessary to set the region of interest P. Alternatively, special processing such as reducing the search range may be performed.

The image processing device 102 sets the search area A centered on the area of interest P by the determination unit 403 (step S902). The image processing device 102 detects the blood vessel pixel region of the blood vessel detection result D in the search region A by the determination unit 403, and counts the number N1 thereof (step S903). Specifically, for example, the number N1 of the blood vessel pixel regions of the blood vessel detection result D shown in FIG. 2 is N1 = 1.

As shown in FIG. 7, the image processing apparatus 102 sets the exclusion region B centered on the region of interest P by the determination unit 403 (step S904). Then, the image processing device 102 outputs the correction detection result D2 by the determination unit 403 (step S905). The modified detection result D2 is a region obtained by subtracting the exclusion region B from the blood vessel detection result D.

The image processing device 102 detects the number N2 of the white pixel connection region DW of the correction detection result D2 in the search region A by the determination unit 403 (step S906). In the example of FIG. 7, N2 = 5. Next, the image processing device 102 calculates the difference value N = N2-N1 in the region of interest P by the determination unit 403, and determines whether or not the difference value N is, for example, the first threshold value VM or more. (Step S907).

When the difference value N is not equal to or greater than the first threshold value VM (step S907: No), the region of interest P is excluded from the vortex vein candidates, and the process proceeds to step S909. On the other hand, when the difference value N is equal to or greater than the first threshold value VM (step S907: Yes), the image processing apparatus 102 determines the region of interest P as the vortex vein candidate V by the determination unit 403 (step S908). , Step S909.

The image processing device 102 determines whether or not there is an undesignated region of interest P by the determination unit 403 (step S909). If there is an undesignated region of interest P (step S909: Yes), the process returns to step S901. On the other hand, when there is no undesignated region of interest P (step S909: No), the image processing device 102 ends the vortex vein position determination process (step S503) and proceeds to step S504.

As described above, according to the first embodiment, the vortex vein 107, which is a blood vessel site where the plurality of choroidal blood vessels 106 of the vortex vein 107 are centrally connected, and the plurality of choroidal blood vessels 106 connected to the vortex vein 107. The image processing apparatus 102 can detect the vortex vein candidate V with high accuracy even when the number of the vortex vein candidates is unknown. Further, the image processing device 102 can also estimate the existence position of the choroidal blood vessel 106 connected to the vortex vein candidate V from the position of the blood vessel pixel region of the correction detection result D2.

Example 2 shows Example 2 of the vortex vein positioning process (step S503). Example 2 of the vortex vein position determination process (step S503) is an example in which the region of interest P is determined as a vortex vein candidate based on the number of blood vessel sites around the region of interest P in the blood vessel detection result D. The same components as those in the first embodiment are designated by the same reference numerals, and the description thereof will be omitted.

FIG. 10 is an enlarged view of a part area A of the binarized image data C according to the second embodiment. Two blood vessel detection results D are detected in the search area A. In order to distinguish, in the blood vessel detection result D of FIG. 2, the region connected to the region of interest P inside the region A is referred to as Da, and the region not connected is referred to as Db. When Da and Db are not distinguished, it is referred to as blood vessel detection result D.

The figure E is a circle centered on the region of interest P in the region A. The figure E has a size larger than that of a general vortex vein 107. The shape of the figure E is not limited to a circle, and may be an ellipse or a polygon including a rectangle. The determination unit 403 detects a region (represented by hatching, hereinafter referred to as an intersection region) J where the figure E (boundary) and the blood vessel detection results Da and Db intersect. In FIG. 10, the intersecting region J is composed of a plurality of pixels, but in FIG. 10, the number of connected regions (the sum of the number of regions Da and the number of regions Db) is eight, and these are J1, J2, ...・ Set as J8.

Then, the determination unit 403 searches for the path R connecting the region of interest P and the intersection region J by passing through the inside of the blood vessel detection result D for each intersection region J. In FIG. 10, the route R1 between the intersection region J1 and the region of interest P, the route R2 between the intersection region J2 and the region of interest P, the route R3 between the intersection region J3 and the region of interest P, and the route between the intersection region J4 and the region of interest P. R4, the route R5 between the intersection region J5 and the region of interest P, the route R6 between the intersection region J6 and the region of interest P, and the route R7 between the intersection region J7 and the region of interest P are detected.

Since the intersecting region J8 is a region within the blood vessel detection result Db and the Db is not connected to the blood vessel detection result Da, it cannot be connected to the blood vessel detection result Da with white pixels. Therefore, there is no route between the intersection region J8 and the region of interest P. The path R is, for example, the shortest path by the Dijkstra method, that is, the path that minimizes the number of white pixels. The crossing region J assumes the site of the choroidal blood vessel 106 connected to the vortex vein 107. Therefore, when the number of white pixels of the path R is equal to or less than the third threshold value, it is considered that the choroidal blood vessel 106 is not formed, so that the determination unit 403 does not have to be generated as the path R.

<Swirl vein position determination process (step S503)>
FIG. 11 is a flowchart showing a detailed processing procedure example 2 of the vortex vein position determining process (step S503) shown in FIG. After executing step S901 by the determination unit 403, the image processing device 102 sets the figure E centered on the region of interest P as shown in FIG. 10 (step S1103). The image processing device 102 detects the intersection region J between the figure E and the blood vessel detection result D by the determination unit 403 (step S1104). Specifically, for example, in FIG. 10, the intersecting regions J1 to J8 are detected.

The image processing device 102 searches the path R between the intersection region J and the region of interest P by the determination unit 403 as shown in FIG. 10 (step S1105). Next, the image processing device 102 determines whether or not the number m of the paths R is equal to or greater than the first threshold value VM by the determination unit 403 (step S1106).

When the number m of the paths R is not equal to or greater than the first threshold value VM (step S1106: No), the region of interest P is excluded from the vortex vein candidates, and the process proceeds to step S909. On the other hand, when the number m of the paths R is equal to or greater than the first threshold value VM (step S1106: Yes), the image processing apparatus 102 determines the region of interest P as the vortex vein candidate V by the determination unit 403 (step S1106: Yes). S1107), the process proceeds to step S909.

As described above, according to the second embodiment, since the image processing apparatus 102 includes a process of determining the intersection region J connected to the region of interest P, the vortex is assumed when the region of interest P is a candidate for a vortex vein. The number of choroidal blood vessel 106 candidates connected to the vein candidates can be detected. By searching for the path R with the region of interest P during the process, it is also possible to suppress a decrease in the determination accuracy of the vortex vein candidate by excluding the crossing region J in which the path R cannot be generated from the choroid candidates. It becomes.

Further, when the number m of the paths R is equal to or larger than the first threshold value VM, the region of interest P is determined as the vortex vein candidate, so that the determination accuracy of the vortex vein candidate can be improved. In this way, when the number of the vortex vein 107, which is a blood vessel site where the plurality of choroidal blood vessels 106 of the vortex vein 107 are centrally connected, or the number of the plurality of choroidal blood vessels 106 connected to the vortex vein 107 is unknown. Even if there is, the image processing device 102 can detect the vortex vein candidate V with high accuracy. The image processing device 102 can also estimate the existence position of the choroidal blood vessel 106 connected to the vortex vein candidate V depending on the direction of the path R.

Example 3 shows Example 3 of the vortex vein positioning process (step S503). Example 3 of the vortex vein positioning process (step S503) utilizes the shape feature of the vortex vein 107 that the plurality of choroidal blood vessels 106 connected to the vortex vein 107 are anisotropic with respect to the vortex vein 107. This is an example of determining a vortex vein candidate based on the distribution of blood vessel detection result D (distribution of white pixels). Anisotropy is based on the fact that the number of choroidal blood vessels connected to the vortex vein 107 tends to differ depending on the direction when the vortex vein 107 is centered. The number of choroidal blood vessels on the peripheral side of the fundus tends to be large, and the number of choroidal blood vessels on the central side of the fundus tends to be small.
The same components as those of Examples 1 and 2 are designated by the same reference numerals, and the description thereof will be omitted.

FIG. 12 is an enlarged view of a part area A of the binarized image data C according to the third embodiment. The figure G is a circle centered on the region of interest P in the region A. The figure G has a size larger than that of a general vortex vein 107. A line segment that passes through the region of interest P and is parallel to the X direction is defined as a line segment L1. A line segment that passes through the region of interest P and intersects the line segment L1 is defined as a line segment L2. Let θ be the angle formed by the line segment L1 and the line segment L2. The area of the figure G is bisected by the line segment L1 or the line segment L2.

Let H1 and H2 be the regions divided by the line segment L2 in the figure G. The area of region H1 and the area of region H2 are equal. If the region of interest P exists at the position of the vortex vein 107, the area S1 of the blood vessel detection result D in the division region H1 and the area of the blood vessel detection result D in the division region H2 due to the anisotropy of the vortex vein 107 described above. A bias occurs in S2. The determination unit 403 calculates a value S based on the bias, and if the value S is equal to or less than the fourth threshold value, determines the region of interest P as a vortex vein candidate.

The determination unit 403 changes the angle θ within a predetermined range (for example, 0 degree ≤ θ ≤ 180 degrees), and detects the angle θ at which the bias is equal to or greater than the fourth threshold value. The step width Δθ for changing the angle θ is preset. Further, when the region of interest P is determined as a vortex vein candidate, the blood vessel detection result D corresponding to the choroidal blood vessel 106 is biased to either the divided region H1 or H2 (that is, the white pixel in the divided region H1). There is a bias between the number of pixels in the above and the number of white pixels in the divided region H2). Therefore, the determination unit 403 can identify whether the choroidal blood vessel 106 connected to the vortex vein 107 is abundant in the divided regions H1 and H2.

<Swirl vein position determination process (step S503)>
FIG. 13 is a flowchart showing a detailed processing procedure example 3 of the vortex vein position determining process (step S503) shown in FIG. After executing step S901 by the determination unit 403, the image processing device 102 sets the figure G centered on the region of interest P as shown in FIG. 12 (step S1303). The image processing device 102 uses the determination unit 403 to set a preset initial value for the angle θ with respect to the line segment L1 (step S1034).

As shown in FIG. 12, the image processing apparatus 102 sets a straight line L2 that passes through the region of interest P and forms an angle θ with the line segment L1 by the determination unit 403 (step S1305). Next, the image processing device 102 calculates the areas S1 and S2 of the divided regions H1 and H2 on the straight line L2 in the figure G by the determination unit 403 (step S1306). The area S1 is the number of white pixels in the divided region H1, and the area S2 is the number of white pixels in the divided region H2.

Next, the image processing device 102 calculates the value S based on the bias by the determination unit 403 (step S1307). The value S is calculated by the following formula (1).

S = MIN (S1, S2) / MAX (S1, S2) ... (1)

The value S indicates that the smaller the value, the larger the area difference between the divided regions H1 and H2. If the bias S is not equal to or less than the fourth threshold value St (step S1308: No), the process proceeds to step S1310. When the bias S is equal to or less than the fourth threshold value St (step S1308: Yes), the image processing apparatus 102 determines the region of interest P as a vortex vein candidate by the determination unit 403, and shifts to (step S1309).

When the numerator on the right side of the equation (1) and the denominator are interchanged, the bias S indicates that the larger the value, the larger the area difference between the divided regions H1 and H2. In this case, in step S1308, if the bias S is not equal to or greater than the fourth threshold value St (step S1308: No), the process proceeds to step S1310. When the bias S is equal to or greater than the fourth threshold value St (step S1308: Yes), the image processing apparatus 102 determines the region of interest P as a vortex vein candidate by the determination unit 403, and shifts to (step S1309). Become.

In step S1310, the image processing device 102 determines whether or not the angle search in the region of interest P has been completed by the determination unit 403 (step S1310). As the end condition of the angle search, for example, the angle θ after the step width Δθ addition may exceed a predetermined range, or the region of interest P may be determined as a vortex vein candidate in step S1309. When the angle search in the region of interest P is not completed (step S1310: No), the process proceeds to step S1304, and Δθ is added to the angle θ to update the angle θ. On the other hand, when the angle search in the region of interest P is completed (step S1310: Yes), the process proceeds to step S909.

As described above, according to the third embodiment, the image processing apparatus 102 determines the vortex vein candidate by utilizing the characteristic of the shape of the vortex vein 107 that it is anisotropic with respect to the vortex vein 107. It is possible to improve the detection accuracy of candidates. Further, the image processing device 102 can estimate the direction in which the choroidal blood vessel 106 connected to the vortex vein candidate V exists from the position of the divided region having the larger bias.

Further, in FIG. 13, as the end condition of the angle search in step S1310, the case where the region of interest P is determined as the vortex vein candidate in step S1309 is mentioned, but the case where the region of interest P is determined as the vortex vein candidate is the case. However, if the angle θ after the step width Δθ addition does not exceed the predetermined range, the process may return to step S1304.

That is, the determination unit 403 may count the number of determinations of the vortex vein candidate in step S1309 in the region of interest P. In this case, if the number of determinations is not equal to or greater than the fifth threshold value, the determination unit 403 may exclude the region of interest P from the vortex vein candidates. By tentatively determining the vortex vein candidates for each angle and narrowing down the vortex vein candidates based on the number of tentatively determined times, it is possible to improve the detection accuracy of the vortex vein 107.

Further, in Example 3, the binarized image data C obtained by binarizing the fundus image data 110 was used, but the grayscale image data 110 of the fundus of the eye or the image data obtained by emphasizing the choroidal blood vessels was binarized. It may be used instead of C. In this case, the areas S1 and S2 may be the sum of the brightness values. As a result, the binarization process becomes unnecessary, and the detection speed of the vortex vein can be improved.

Example 4 is an example in which at least two of Examples 1 to 3 are combined. For example, when combining Example 1 and Example 2, the image processing apparatus 102 uses the vortex vein candidate V (referred to as V1) in Example 1 and the vortex vein candidate V (referred to as V2) in Example 2. ) And, the vortex vein candidates common to are output. Further, when combining Example 1 and Example 3, the image processing device 102 is common to the vortex vein candidate V1 in Example 1 and the vortex vein candidate V (referred to as V3) in Example 3. Output vortex vein candidates.

Further, when combining Example 2 and Example 3, the image processing apparatus 102 selects a vortex vein candidate common to the vortex vein candidate V2 in Example 2 and the vortex vein candidate V3 in Example 3. Output. Further, when combining Examples 1 to 3, the image processing apparatus 102 uses the vortex vein candidate V1 in Example 1, the vortex vein candidate V2 in Example 2, and the vortex vein candidate in Example 3. Outputs vortex vein candidates common to V (referred to as V3).

In this way, by combining at least two of Examples 1 to 3, the vortex vein candidates are narrowed down, and the detection accuracy of the vortex vein candidates can be improved.

The present invention is not limited to the above contents, and may be any combination thereof. In addition, other aspects considered within the scope of the technical idea of the present invention are also included in the scope of the present invention.

A search area, B exclusion area, C binarized image data, D blood vessel detection result, 100 ophthalmic system, 101 ophthalmic device, 102 image processing device, 103 terminal, 104 network, 105 eye to be examined, 106 choroidal blood vessel, 107 vortex vein , 110 fundus image data, 300 computer, 301 processor, 302 storage device, 303 input device, 304 output device, 401 image acquisition unit, 402 blood vessel acquisition unit, 403 determination unit, 404 output unit

Claims (9)

An acquisition unit that acquires choroidal blood vessel data generated from fundus image data,
A determination unit that sets the first region in the fundus image data and determines the vortex vein position based on the feature amount of the choroidal blood vessel data around the first region.
An output unit that outputs the vortex vein position and
An image processing device comprising. The image processing apparatus according to claim 1.
The vortex vein position is a candidate position of the vortex vein in which the feature amount satisfies a predetermined condition.
The determination unit determines the vortex vein position from the plurality of candidate positions.
Image processing device. The image processing apparatus according to claim 1 or 2.
A second region including the first region is set, and the second region is set.
The feature amount is the number of choroidal blood vessels connected to the second region.
Image processing device. The image processing apparatus according to claim 1 or 2.
The determination unit sets a figure including the first region, and sets the figure.
The feature amount is the number of intersections between the figure and the choroidal blood vessel.
Image processing device. The image processing apparatus according to claim 1 or 2.
The determination unit sets a figure including the first region, and sets the figure.
The feature amount is the number of routes that follow from the intersection of the figure and the choroidal blood vessel to the first region.
Image processing device. The image processing apparatus according to claim 1 or 2.
The determination unit sets a figure including the first region, sets a line segment that passes through the first region and divides the figure into a first division region and a second division region having the same area, and the first The area of the first choroidal blood vessel in the one division region and the area of the second choroidal blood vessel in the second division region were calculated.
The feature amount is the ratio of the first choroidal blood vessel area to the second choroidal blood vessel area.
Image processing device. The image processing apparatus according to any one of claims 1 to 6.
The determination unit selects the first region from one of the pixels of the choroidal blood vessel.
Image processing device. An image processing method executed by an image processing apparatus having a processor that executes a program and a storage device that stores the program.
The processor
Acquisition process to acquire choroidal blood vessel data generated from fundus image data,
A determination process in which a first region in the fundus image data is set and the position of the vortex vein is determined based on the feature amount of the choroidal blood vessel data around the first region.
Output processing to reissue the decision result by the decision processing and
Image processing method to execute. To the processor
Acquisition process to acquire choroidal blood vessel data generated from fundus image data,
A determination process in which a first region in the fundus image data is set and the position of the vortex vein is determined based on the feature amount of the choroidal blood vessel data around the first region.
Output processing that outputs the determination result by the determination processing and
An image processing program that executes.

Images