Method for automatic measurement of meibomian gland quality

Abstract

The present application relates to a kind of methods for automatically measuring meibomian gland mass, the method comprises: a set of continuous gray images including eye and meibomian gland is collected;According to the gray image of succession, the gray value change rate of each pixel is determined;The minimum circumscribed rectangle of eye in each image is determined, the area in minimum circumscribed rectangle is formed foreground area, the area outside minimum circumscribed rectangle is formed background area;According to the gray value change rate of each pixel in background area, the foreground area is merged, and merged image is obtained;The pre-processing of merged image is obtained after processing image;Wherein, pre-processing includes: median filter processing, normalization processing and removing uneven illumination processing;Based on after processing image, the mass of meibomian gland is measured.The method provided by the present application can realize the automatic evaluation of meibomian gland based on a set of continuous gray images including eye and meibomian gland.

Description

Methods for automatically measuring meibomian gland quality Technical Field

[0001] This invention relates to the field of medical image processing technology, and in particular to a method for automatically measuring meibomian gland quality. Background Technology

[0002] Meibomian gland dysfunction (MGD) is one of the main causes of evaporative dry eye.

[0003] Current diagnostic methods for MGD mainly rely on qualitative assessments by ophthalmologists, lacking automated and quantitative analysis tools.

[0004] Therefore, there is an urgent need for an automated method to assess the quality and function of meibomian glands. Summary of the Invention

[0005] (a) Technical problems to be solved

[0006] To address the aforementioned problems, this invention provides a method for automatically measuring meibomian gland quality.

[0007] (II) Technical Solution

[0008] To achieve the above objectives, the main technical solutions adopted by the present invention include:

[0009] A method for automatically measuring meibomian gland quality, the method comprising:

[0010] Acquire a series of consecutive grayscale images, including those of the eyes and meibomian glands;

[0011] Based on a continuous grayscale image, determine the rate of change of grayscale value for each pixel;

[0012] Determine the minimum bounding rectangle of the eye in each image, and form the foreground region by the area within the minimum bounding rectangle and the background region by the area outside the minimum bounding rectangle.

[0013] Based on the rate of change of gray values ​​of each pixel in the background region, the foreground region is merged to obtain a merged image;

[0014] The merged image is preprocessed to obtain the processed image; the preprocessing includes: median filtering, normalization and removal of uneven illumination.

[0015] Meibomian gland quality was measured based on the processed images.

[0016] Optionally, based on a continuous grayscale image, the rate of change of grayscale value for each pixel is determined, including:

[0017] Determine the grayscale value at each pixel in each image. ,in, For image identification, Pixel identifier;

[0018] Determine the average grayscale value of each pixel. ,in, The total number of images;

[0019] Determine the initial rate of change of grayscale values ​​for each pixel. ;

[0020] Determine the mean of the initial rate of change ,in, This represents the total number of pixels.

[0021] Determine the overall rate of change ;

[0022] Determine the adjustment ratio for each pixel. ;

[0023] Determine the rate of change of grayscale values ​​for each pixel. .

[0024] Optionally, determine the adjustment ratio for each pixel. ,include:

[0025] Determine the adjustment degree of each pixel as follows: ;

[0026] like ,but ;

[0027] like ,but ;

[0028] like ,but ;

[0029] in, To find the maximum value function, This is a function that takes the minimum value.

[0030] Optionally, the foreground region is merged based on the grayscale value change rate of each pixel in the background region to obtain a merged image, including:

[0031] Determine the mean rate of change of grayscale values ​​for all pixels located in the background region. Maximum value and minimum value ;

[0032] For each pixel located in the foreground region, determine the merged grayscale value.

[0033] ,in, For image identification, For the pixel identifier of the foreground area, For image In pixels grayscale value at that location The total number of images. For pixels The rate of change of grayscale value To find the minimum value function, This is the preset maximum adjustment value. ;

[0034] Based on the merged grayscale value of each pixel in the foreground region The merged image is obtained.

[0035] Optionally, the merged images are preprocessed to obtain a processed image, including:

[0036] based on The merged image is then subjected to median filtering to obtain a median-filtered image.

[0037] The median-filtered image is normalized to obtain a normalized image; where any pixel in the normalized image... grayscale value

[0038] ,in, For any pixel in the median filtered image grayscale value, To find the maximum value function, The function is for finding the minimum value;

[0039] The normalized image is processed by a low-pass filter to obtain a low-pass filtered image;

[0040] The difference between the normalized image and the low-pass filtered image is used to determine the processed image.

[0041] Optionally, based on the processed image, meibomian gland quality is measured, including:

[0042] According to the mask The processed image is then processed to detect the meibomian gland region within the processed image.

[0043] The quality of the meibomian glands was measured based on the meibomian gland region.

[0044] Optionally, according to the mask The processed image is then processed to detect the meibomian gland region, including:

[0045] Determine the grayscale value of each pixel contained within the meibomian gland region. ,in, Pixel identifiers for the meibomian gland region;

[0046] Determine the maximum threshold for each pixel and minimum threshold ;

[0047] Determine the grayscale value of each pixel.

[0048] ;in, Maximum threshold for pixels, The minimum threshold for pixels, To find the maximum value function, The function is for finding the minimum value;

[0049] According to the mask The image formed by grayscale values ​​is processed to detect the meibomian gland region.

[0050] Optionally, ;

[0051] in, This represents the overall rate of change.

[0052] Optionally, ;

[0053] in, This represents the overall rate of change.

[0054] Optionally, ;

[0055] in, The three dimensions of mask rotation, The average deviation of the mask in three dimensions, These are the standardized values ​​for the three dimensions.

[0056] (III) Beneficial Effects

[0057] This invention relates to a method for automatically measuring meibomian gland quality. The method includes: acquiring a set of continuous grayscale images including the eye and meibomian glands; determining the grayscale value change rate of each pixel based on the continuous grayscale images; determining the minimum bounding rectangle of the eye in each image, forming a foreground region within the minimum bounding rectangle and a background region outside the minimum bounding rectangle; merging the foreground regions based on the grayscale value change rate of each pixel within the background region to obtain a merged image; preprocessing the merged image to obtain a processed image; wherein the preprocessing includes: median filtering, normalization, and removal of uneven illumination; and measuring meibomian gland quality based on the processed image. The method provided by this invention can achieve automatic assessment of meibomian glands based on a set of continuous grayscale images including the eye and meibomian glands. Attached Figure Description

[0058] Figure 1 is a flowchart illustrating an embodiment of the present application of an automatic method for measuring meibomian gland quality;

[0059] Figure 2 is a schematic diagram of one image of a set of consecutive grayscale images provided in an embodiment of this application;

[0060] Figure 3 is a schematic diagram of meibomian gland region detection provided in an embodiment of this application. Detailed Implementation

[0061] To better explain and facilitate understanding of the present invention, the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0062] Current diagnostic methods for meibomian gland dysfunction (MGD) primarily rely on qualitative assessments by ophthalmologists, lacking automated and quantitative analytical tools. Therefore, there is an urgent need for an automated method to assess the quality and function of the meibomian glands.

[0063] To address this problem, the present invention relates to a method for automatically measuring meibomian gland quality. The method includes: acquiring a set of continuous grayscale images including the eye and meibomian glands; determining the grayscale value change rate of each pixel based on the continuous grayscale images; determining the minimum bounding rectangle of the eye in each image, forming a foreground region within the minimum bounding rectangle and a background region outside the minimum bounding rectangle; merging the foreground regions based on the grayscale value change rate of each pixel within the background region to obtain a merged image; preprocessing the merged image to obtain a processed image; wherein the preprocessing includes: median filtering, normalization, and removal of uneven illumination; and measuring meibomian gland quality based on the processed image. The method provided by this invention can achieve automatic assessment of meibomian glands based on a set of continuous grayscale images including the eye and meibomian glands.

[0064] Referring to Figure 1, the implementation process of the automatic measurement method for meibomian gland quality provided in this embodiment is as follows:

[0065] 101. Acquire a series of continuous grayscale images including the eyes and meibomian glands.

[0066] For example, a high-precision, high-definition camera is used to capture a video of a user's eye. During the acquisition process, the camera's position and parameters must not change, and the user should also avoid changing their position as much as possible. This ensures that the acquisition device and the object being acquired for a series of consecutive grayscale images are identical, thereby guaranteeing the comparability of the automatic meibomian gland quality measurement method provided in this embodiment during subsequent processing. For example, for different images, the content represented by the same pixel position is the same. For example, one image from a series of consecutive grayscale images is shown in Figure 2.

[0067] 102. Based on a continuous grayscale image, determine the rate of change of grayscale value for each pixel.

[0068] The implementation process of this step is as follows:

[0069] 102-1, Determine the grayscale value of each pixel in each image. .

[0070] in, For image identification, Pixel identifier.

[0071] For example, the first pixel in the top left corner is labeled as 1, and the pixel immediately to its right is labeled as 2. The pixels are numbered sequentially from left to right and from top to bottom. In this way, the total number of pixels in each image and the positions of pixels with the same label are the same.

[0072] In step 102-1, the grayscale value of pixel 1 in image 1 is determined. The grayscale value of pixel 1 in image 2 ..., pixel 1 in the image grayscale values ​​in Determine the grayscale value of pixel 2 in image 1, the grayscale value of pixel 2 in image 2, ..., the grayscale value of pixel 2 in image 1. The grayscale value in the image. ...

[0073] 102-2, Determine the average grayscale value of each pixel. .

[0074] in, This represents the total number of images.

[0075] For example, for pixel 1, its grayscale mean .

[0076] 102-3, Determine the initial rate of change of grayscale values ​​for each pixel. .

[0077] For example, for pixel 1, its initial rate of change of grayscale value

[0078] .

[0079] initial rate of change Characterizes a certain pixel Regarding the changes in grayscale values ​​in each image, since the position, parameters, and user position of the acquisition device are kept relatively fixed during image acquisition in step 101, ideally, the grayscale values ​​of a particular pixel... The grayscale values ​​remain unchanged across all images, that is, under ideal conditions. In other words The closer it is to 0, the more accurate the set of continuous grayscale images acquired in step 101 is.

[0080] 102-4, Determine the mean of the initial rate of change .

[0081] in, This represents the total number of pixels.

[0082] 102-5, Determine the overall rate of change .

[0083] Overall rate of change It characterizes the overall grayscale value variation of pixels in a set of continuous grayscale images. The closer the value is to 0, the more accurate the set of consecutive grayscale images acquired in step 101 is.

[0084] 102-6, Determine the adjustment ratio for each pixel. .

[0085] That is, determine the adjustment degree of each pixel as ,according to Determine the adjustment ratio .

[0086] like ,but .

[0087] like ,but .

[0088] like ,but .

[0089] in, To find the maximum value function, This is a function that takes the minimum value.

[0090] 102-7, Determine the grayscale value change rate of each pixel. .

[0091] Adjustment Characterizes a certain pixel The relationship between the changes in grayscale values ​​in individual images and the overall grayscale value changes of pixels in a set of consecutive grayscale images. If the overall grayscale value changes of pixels in a set of consecutive grayscale images are considered unavoidable differences caused by device and user system errors, then for a given pixel... Adjustment degree Characterize a pixel The difference between the change and the systematic error is something that needs to be addressed (i.e., This ensures the comparability of grayscale values ​​between each pixel.

[0092] if This indicates that a certain pixel If the change is indistinguishable from the systematic error, and the two are the same, then there is no need to eliminate it.

[0093] if This indicates that a certain pixel If the change is reduced relative to the systematic error, then the variance is increased to regress it to the standard error; the larger the variance, the more it is regressed.

[0094] if This indicates that a certain pixel If the change in variance increases relative to the systematic error, then the variance is reduced to regress it onto the standard error. The larger the variance, the more it is regressed.

[0095] 103. Determine the minimum bounding rectangle of the eye in each image, and form the foreground region by the area within the minimum bounding rectangle and the background region by the area outside the minimum bounding rectangle.

[0096] This step uses an existing image recognition scheme to obtain the minimum bounding rectangle of the eye in each image. Since the position, parameters, and user position of the acquisition device are kept relatively fixed when acquiring images in step 101, the minimum bounding rectangle in each image does not change much and can be considered unchanged.

[0097] 104. Based on the rate of change of gray values ​​of each pixel in the background area, the foreground area is merged to obtain a merged image.

[0098] The merging process is as follows:

[0099] 104-1, Determine the mean rate of change of grayscale values ​​for all pixels located in the background region. Maximum value and minimum value .

[0100] 104-2, For each pixel located in the foreground region, determine the merged grayscale value. .

[0101] in, For image identification, For the pixel identifier of the foreground area, For image In pixels grayscale value at that location The total number of images. For pixels The rate of change of grayscale value To find the minimum value function, This is the preset maximum adjustment value. .

[0102] In step 104-2, image grayscale errors caused by uncontrollable factors such as user movement are eliminated, further ensuring the accuracy of subsequent image processing.

[0103] The background area will not change due to human factors, so in step 104-2, the pixels of the foreground area will be adjusted based on the grayscale value of the background area.

[0104] It represents the pixels in the foreground region across all images. The average gray value is a base value. It is an adjustment value for the pixels in the foreground area based on the grayscale value of the background area.

[0105] 104-3, based on the merged grayscale value of each pixel in the foreground region. The merged image is obtained.

[0106] The merged image is actually an image containing only the smallest bounding rectangle of the eye, and the values ​​of each pixel in the image are the merged and adjusted values, which is an image that can accurately reflect the user's eyes and meibomian glands.

[0107] 105. Preprocess the merged image to obtain the processed image.

[0108] The preprocessing includes median filtering, normalization, and removal of uneven illumination.

[0109] For example, it can be achieved through the following steps:

[0110] 105-1, based on The merged images are then subjected to median filtering to obtain a median-filtered image.

[0111] 105-2, normalize the median-filtered image to obtain a normalized image. In the normalized image, any pixel... grayscale value .

[0112] in, For any pixel in the median filtered image grayscale value, To find the maximum value function, This is a function that takes the minimum value.

[0113] 105-3. The normalized image is processed by a low-pass filter to obtain a low-pass filtered image.

[0114] 105-4, the difference between the normalized image and the low-pass filtered image is determined as the processed image.

[0115] Low-pass filtered images characterize the differences caused by uneven illumination. By determining the difference between the normalized image and the low-pass filtered image as the processed image, the effect of uneven illumination in the normalized image can be removed, resulting in an image after removing the unevenness, i.e., the processed image. This allows for accurate measurement of meibomian gland quality.

[0116] 106. Measuring meibomian gland quality based on processed images.

[0117] The implementation process of step 106 is as follows:

[0118] 106-1, based on the mask The processed image is then processed to detect the meibomian gland region within the processed image.

[0119] To further eliminate the influence of the meibomian gland margin area, step 106-1 is implemented as follows:

[0120] 1. Determine the grayscale value of each pixel contained in the meibomian gland region. .

[0121] in, Pixel identifier for the meibomian gland region.

[0122] 2. Determine the maximum threshold for pixels. and minimum pixel threshold .

[0123] in, . .

[0124] in, The overall rate of change. To find the maximum value function, This is a function that takes the minimum value.

[0125] 3. Determine the grayscale value for each pixel. .

[0126] Grayscale values ​​of each pixel It is the grayscale value of each pixel. On this basis, the differences between grayscale values ​​are further increased.

[0127] 4. Based on the mask The image formed by grayscale values ​​is processed to detect the meibomian gland region.

[0128] The above processing can increase the difference between the gray values ​​of each pixel in the image formed by gray-scale processing, thereby better increasing the gray value difference of the meibomian gland edge, more accurately identifying the edge of the meibomian gland, and thus obtaining the meibomian gland region formed by the edge of the meibomian gland.

[0129] in addition, .

[0130] in, The three dimensions of mask rotation, The average deviation of the mask in three dimensions, These are the standardized values ​​for the three dimensions.

[0131] For example, , , .

[0132] in, These are the standard values ​​for the three dimensions of mask rotation.

[0133] By creating a mask Sequences can be used to isolate meibomian glands from processed images. For example, a mask. If the obtained value is greater than the preset threshold, the corresponding processed value is 1; otherwise, it is 0. The meibomian gland region is obtained based on the processed value.

[0134] For example, as shown in Figure 3, A schematic diagram of the detection results for the mask and meibomian gland areas.

[0135] 106-2, Measuring meibomian gland quality based on the meibomian gland region.

[0136] 106-3, according to Measuring meibomian gland quality.

[0137] Among them, meibomian gland quality includes the proportion of meibomian glands, meibomian gland length, curvature, width, length, and number of glands.

[0138] For example, the proportion of meibomian glands is ,in, This represents the number of pixels contained in the meibomian gland region. This represents the number of pixels contained within the eyelid.

[0139] The curvature is the length of the meibomian gland divided by the distance between the two ends of the meibomian gland.

[0140] This embodiment provides a method for automatically measuring meibomian gland quality. The method involves acquiring a set of continuous grayscale images including the eyes and meibomian glands; determining the grayscale value change rate of each pixel based on the continuous grayscale images; determining the minimum bounding rectangle of the eye in each image; defining the area within the minimum bounding rectangle as the foreground region and the area outside the minimum bounding rectangle as the background region; merging the foreground regions based on the grayscale value change rate of each pixel within the background region to obtain a merged image; preprocessing the merged image to obtain a processed image; the preprocessing includes median filtering, normalization, and removal of uneven illumination; and measuring the meibomian gland quality based on the processed image. This embodiment provides a method for automatically assessing meibomian glands based on a set of continuous grayscale images including the eyes and meibomian glands.

[0141] It should be clarified that the present invention is not limited to the specific configurations and processes described above and shown in the figures. For the sake of brevity, detailed descriptions of known methods are omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of the present invention is not limited to the specific steps described and shown. Those skilled in the art can make various changes, modifications, and additions, or change the order of steps, after understanding the spirit of the present invention.

[0142] It should also be noted that the exemplary embodiments mentioned in this invention describe methods or systems based on a series of steps or apparatus. However, this invention is not limited to the order of the steps described above; that is, the steps can be performed in the order mentioned in the embodiments, or in a different order, or several steps can be performed simultaneously.

[0143] Finally, it should be noted that the above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for automatically measuring meibomian gland quality, characterized in that, The method includes: acquiring a set of continuous grayscale images including the eyes and meibomian glands; determining the grayscale value change rate of each pixel based on the continuous grayscale images; determining the minimum bounding rectangle of the eyes in each image, forming the foreground region within the minimum bounding rectangle and the background region outside the minimum bounding rectangle; merging the foreground regions based on the grayscale value change rate of each pixel in the background region to obtain a merged image; preprocessing the merged image to obtain a processed image; wherein the preprocessing includes: median filtering, normalization, and removal of uneven illumination; and measuring the meibomian gland quality based on the processed image.

2. The method according to claim 1, characterized in that, Determining the grayscale value change rate of each pixel based on the continuous grayscale images includes: determining the grayscale value at each pixel in each image. ,in, For image identification, Identify pixels; determine the average grayscale value of each pixel. ,in, The total number of images; determine the initial rate of change of grayscale values ​​for each pixel. Determine the mean of the initial rate of change. ,in, Given the total number of pixels; determine the overall rate of change. Determine the adjustment ratio for each pixel. Determine the rate of change of grayscale values ​​for each pixel. 。 3. The method according to claim 2, characterized in that, The adjustment ratio of each pixel is determined. This includes: determining the adjustment degree of each pixel. ;like ,but ;like ,but ;like ,but ;in, To find the maximum value function, This is a function that takes the minimum value.

4. The method according to claim 1, characterized in that, The step of merging the foreground region based on the grayscale value change rate of each pixel in the background region to obtain a merged image includes: determining the mean of the grayscale value change rate of all pixels located in the background region. Maximum value and minimum value For each pixel located in the foreground region, determine the merged grayscale value. ,in, For image identification, For the pixel identifier of the foreground area, For image In pixels grayscale value at that location The total number of images. For pixels The rate of change of grayscale value To find the minimum value function, This is the preset maximum adjustment value. Based on the merged grayscale value of each pixel in the foreground region. The merged image is obtained.

5. The method according to claim 1, characterized in that, The preprocessing of the merged image to obtain the processed image includes: based on The merged image is then subjected to median filtering to obtain a median-filtered image; the median-filtered image is then normalized to obtain a normalized image; wherein, any pixel in the normalized image... grayscale value ,in, For any pixel in the median filtered image grayscale value, To find the maximum value function, To find the minimum value function, the normalized image is processed by a low-pass filter to obtain a low-pass filtered image; the difference between the normalized image and the low-pass filtered image is determined as the processed image.

6. The method according to claim 1, characterized in that, The measurement of meibomian gland quality based on the processed image includes: according to a mask The processed image is processed to detect the meibomian gland region in the processed image; the meibomian gland quality is measured based on the meibomian gland region.

7. The method according to claim 6, characterized in that, According to the mask The processed image is then processed to detect the meibomian gland region within the processed image, including: determining the grayscale value of each pixel contained within the meibomian gland region. ,in, Identify the pixels in the meibomian gland region; determine the maximum threshold for each pixel. and minimum threshold Determine the grayscale value of each pixel. ;in, Maximum threshold for pixels, The minimum threshold for pixels, To find the maximum value function, To find the minimum value function; based on the mask The image formed by the grayscale values ​​is processed to detect the meibomian gland region.

8. The method according to claim 7, characterized in that, ;in, This represents the overall rate of change.

9. The method according to claim 7, characterized in that, ;in, This represents the overall rate of change.

10. The method according to claim 6 or 7, characterized in that, ;in, The three dimensions of mask rotation, The average deviation of the mask in three dimensions, These are the standardized values ​​for the three dimensions.

Images