Image evaluation method and device, computer readable storage medium and electronic device
By extracting and screening the line feature information of the fundus image and calculating length and gradient, the objectivity and standardization of fundus image clarity evaluation is solved, and the accurate evaluation and standardization of fundus images are achieved.
Patent Information
- Application Number
- CN202110218445.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2021-02-26
- Publication Date
- 2024-11-05
- Estimated Expiration
- 2041-02-26
AI Technical Summary
During the fundus image shooting process, uncertainty in clarity affects the accuracy of the video reading results, and lacks objective and unified evaluation standards, making it difficult to achieve automatic evaluation through computers.
By extracting line feature information of fundus images, including blood vessel and disc edge features, morphological and topological feature screening, length and gradient information are calculated, and the clarity of the image is evaluated based on this information.
The objective and unified evaluation of fundus image clarity is achieved, the accuracy and standardization of evaluation is improved, and it is applicable to all fundus images, reducing the amount of calculation and improving the evaluation efficiency.
Smart Images

Figure CN112907548B_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of image processing technology, and in particular to an image evaluation method and device, a computer-readable storage medium, and an electronic device. Background Art
[0002] In the process of detecting eye lesions, fundus images are a very important basis for detection. However, in the process of taking fundus images, due to the influence of factors such as the shooting environment, the photographer's skills, the patient's cooperation, and the equipment, the clarity of the fundus images finally taken may not meet the requirements of the doctor's later reading, which greatly affects the accuracy of the reading results. In addition, the traditional evaluation of fundus image clarity is subjective and there is no objective or unified standard. Therefore, in the process of automatic evaluation by computer, what features or standards to use to evaluate the clarity of fundus images has become a technical difficulty that needs to be solved urgently. Summary of the invention
[0003] In order to solve the above technical problems, the present application is proposed. The embodiments of the present application provide an image evaluation method and device, a computer-readable storage medium and an electronic device.
[0004] In a first aspect, an embodiment of the present application provides an image evaluation method, comprising: determining line feature information corresponding to a fundus image; and evaluating the clarity of the fundus image based on the line feature information.
[0005] In combination with the first aspect, in some implementations of the first aspect, the line feature information includes blood vessel feature information and / or optic disc edge feature information.
[0006] In combination with the first aspect, in certain implementations of the first aspect, determining line feature information corresponding to the fundus image includes: extracting first line feature information corresponding to the fundus image using a line feature extraction operator; and determining line feature information based on the first line feature information.
[0007] In combination with the first aspect, in certain implementations of the first aspect, after extracting first line feature information corresponding to the fundus image using a line feature extraction operator, the method also includes: performing a screening operation on the first line feature information based on morphological features and topological features corresponding to the first line feature information to generate second line feature information; performing a connection operation on the second line feature information to generate third line feature information corresponding to the second line feature information; wherein, determining the line feature information based on the first line feature information includes: determining the third line feature information corresponding to the first line feature information as the line feature information.
[0008] In combination with the first aspect, in certain implementations of the first aspect, evaluating the clarity of the fundus image based on line feature information includes: calculating length information corresponding to the line feature information to generate first evaluation information; and evaluating the clarity of the fundus image based on the first evaluation information.
[0009] In combination with the first aspect, in certain implementations of the first aspect, length information corresponding to the line feature information is calculated to generate first evaluation information, including: determining a pixel line with a width of a single pixel corresponding to the line feature information based on a preset corrosion operator; and calculating pixel line length information corresponding to the pixel line to generate first evaluation information.
[0010] In combination with the first aspect, in certain implementations of the first aspect, evaluating the clarity of the fundus image based on the line feature information includes: calculating gradient information corresponding to the line feature information to generate second evaluation information; and evaluating the clarity of the fundus image based on the second evaluation information.
[0011] In combination with the first aspect, in certain implementations of the first aspect, gradient information corresponding to the line feature information is calculated to generate second evaluation information, including: determining a gradient area corresponding to the line feature information based on a grayscale image of the fundus image, wherein the gradient area is an area whose grayscale meets a preset grayscale condition; calculating gradient information corresponding to the gradient area to generate second evaluation information.
[0012] In combination with the first aspect, in certain implementations of the first aspect, before determining the line feature information corresponding to the fundus image, the method also includes: performing a preprocessing operation on the fundus image to generate an area to be evaluated corresponding to the fundus image; wherein, determining the line feature information corresponding to the fundus image includes: determining the line feature information corresponding to the fundus image based on the area to be evaluated.
[0013] In combination with the first aspect, in certain implementations of the first aspect, a preprocessing operation is performed on the fundus image to generate an area to be evaluated corresponding to the fundus image, including: extracting an area of interest of the fundus image to generate a first evaluation area; and determining the area to be evaluated based on the first evaluation area.
[0014] In combination with the first aspect, in certain implementations of the first aspect, before determining the area to be evaluated based on the first evaluation area, the method also includes: performing a normalization operation on the first evaluation area to generate a second evaluation area; wherein, determining the area to be evaluated based on the first evaluation area includes: determining the second evaluation area corresponding to the first evaluation area as the area to be evaluated.
[0015] In a second aspect, an embodiment of the present application provides an image evaluation device, including: a line feature information determination module, used to determine line feature information corresponding to a fundus image; and an evaluation module, used to evaluate the clarity of the fundus image based on the line feature information.
[0016] In a third aspect, an embodiment of the present application provides a computer-readable storage medium, which stores a computer program, and the computer program is used to execute the image evaluation method mentioned in any of the above embodiments.
[0017] In a fourth aspect, an embodiment of the present application provides an electronic device, comprising: a processor; a memory for storing instructions executable by the processor; and the processor is used to execute the image evaluation method mentioned in any of the above embodiments.
[0018] The image evaluation method provided in the embodiment of the present application achieves the purpose of evaluating the clarity of the fundus image by determining the line feature information corresponding to the fundus image and evaluating the clarity of the fundus image based on the line feature information. Since the line feature information can accurately characterize the clarity of the fundus image, the embodiment of the present application is conducive to the standardization of the fundus image, so that the fundus image used for reading the film reaches the preset clarity standard. In addition, since the line feature information is a common feature of the fundus image, it can be applied to all fundus images and has universal applicability. BRIEF DESCRIPTION OF THE DRAWINGS
[0019] By describing the embodiments of the present application in more detail in conjunction with the accompanying drawings, the above and other purposes, features and advantages of the present application will become more apparent. The accompanying drawings are used to provide a further understanding of the embodiments of the present application and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the present application and do not constitute a limitation of the present application. In the accompanying drawings, the same reference numerals generally represent the same components or steps.
[0020] Figure 1 Shown is a schematic diagram of a scenario applicable to an embodiment of the present application.
[0021] Figure 2 Shown is another schematic diagram of a scenario applicable to the embodiment of the present application.
[0022] Figure 3 Shown is a flowchart of an image evaluation method provided by an exemplary embodiment of the present application.
[0023] Figure 4 The figure is a schematic diagram of a process of determining line feature information corresponding to a fundus image provided by an exemplary embodiment of the present application.
[0024] Figure 5 The figure shows a fundus image extracted by a meridian feature extraction operator provided by an exemplary embodiment of the present application.
[0025] Figure 6 Shown is a schematic diagram of a process for determining line feature information corresponding to a fundus image provided by another exemplary embodiment of the present application.
[0026] Figure 7 Shown is a partially enlarged fundus image obtained through a screening operation provided by an exemplary embodiment of the present application.
[0027] Figure 8 Shown is a fundus image obtained through a connection operation provided by an exemplary embodiment of the present application.
[0028] Fig. 9 The figure is a schematic diagram of a flow chart of evaluating the clarity of a fundus image based on line feature information provided by an exemplary embodiment of the present application.
[0029] Fig.10 The figure shows a flow chart of calculating the length information corresponding to the line feature information to generate the first evaluation information provided by an exemplary embodiment of the present application.
[0030] Fig.11 The figure is a schematic diagram of a flow chart of evaluating the clarity of a fundus image based on line feature information provided by another exemplary embodiment of the present application.
[0031] Fig.12 The figure is a schematic diagram of a flow chart of calculating gradient information corresponding to line feature information to generate second evaluation information provided by an exemplary embodiment of the present application.
[0032] Fig.13 Shown is a schematic flow chart of an image evaluation method provided by another exemplary embodiment of the present application.
[0033] Fig.14 The figure is a schematic diagram of a flow chart of performing a preprocessing operation on a fundus image to generate an area to be evaluated corresponding to the fundus image, provided by an exemplary embodiment of the present application.
[0034] Fig.15 The figure is a schematic flow chart of another exemplary embodiment of the present application for performing preprocessing operations on a fundus image to generate an area to be evaluated corresponding to the fundus image.
[0035] Fig.16 Shown is a schematic structural diagram of an image evaluation device provided by an exemplary embodiment of the present application.
[0036] Fig.17 Shown is a schematic structural diagram of an image evaluation device provided by another exemplary embodiment of the present application.
[0037] Fig.18 Shown is a schematic structural diagram of an electronic device provided by an exemplary embodiment of the present application. DETAILED DESCRIPTION
[0038] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.
[0039] Figure 1 The figure shows a schematic diagram of a scenario to which the embodiment of the present application is applicable. Figure 1 As shown, the scenario to which the embodiment of the present application is applicable includes a server 1 and an image acquisition device 2, wherein a communication connection relationship exists between the server 1 and the image acquisition device 2.
[0040] Specifically, the image acquisition device 2 is used to acquire a fundus image including a clarity to be evaluated, and the server 1 is used to determine line feature information corresponding to the fundus image based on the fundus image of the clarity to be evaluated acquired by the image acquisition device 2, and then evaluate the clarity of the fundus image based on the line feature information. That is, this scenario implements an image evaluation method.
[0041] because Figure 1 The above-mentioned scenario shown utilizes the server 1 to implement the image evaluation method. Therefore, the scenario can not only improve the adaptability of the scenario, but also effectively reduce the amount of calculation of the image acquisition device 2 .
[0042] It should be noted that the present application is also applicable to another scenario. Figure 2 Another schematic diagram of a scenario applicable to the embodiment of the present application is shown. Specifically, the scenario includes an image processing device 3, wherein the image processing device 3 includes an image acquisition module 301 and a computing module 302, and there is a communication connection relationship between the image acquisition module 301 and the computing module 302.
[0043] Specifically, the image acquisition module 301 in the image processing device 3 is used to acquire a fundus image including the clarity to be evaluated, and the calculation module 302 in the image processing device 3 is used to determine the line feature information corresponding to the fundus image based on the fundus image of the clarity to be evaluated acquired by the image acquisition module 301, and then evaluate the clarity of the fundus image based on the line feature information. That is, this scenario implements an image evaluation method.
[0044] because Figure 2 The above-mentioned scene shown utilizes the image processing device 3 to implement the image evaluation method, and does not require data transmission operations with related devices such as a server. Therefore, Figure 2 The above scenario shown can ensure the real-time performance of the image evaluation method.
[0045] Figure 3The figure is a flow chart of an image evaluation method provided by an exemplary embodiment of the present application. Specifically, the image evaluation method provided by the embodiment of the present application is applied to fundus images including a clarity to be evaluated.
[0046] like Figure 3 As shown, the image evaluation method provided in the embodiment of the present application includes the following steps.
[0047] Step 10: Determine line feature information corresponding to the fundus image.
[0048] It should be noted that the fundus image mentioned in step 10 refers to the fundus image whose clarity needs to be evaluated, and the line feature information refers to feature information that can be used to evaluate the clarity of the fundus image, including vascular feature information and / or optic disc edge feature information. Since the line feature information is feature information common to fundus images, the image evaluation method provided in the embodiment of the present application can be applied to all fundus images and has universal applicability.
[0049] Step 20: Evaluate the clarity of the fundus image based on the line feature information.
[0050] In actual application, the line feature information corresponding to the fundus image is first determined, and then the clarity of the fundus image is evaluated based on the line feature information.
[0051] The image evaluation method provided in the embodiment of the present application achieves the purpose of evaluating the clarity of the fundus image by determining the line feature information corresponding to the fundus image and then evaluating the clarity of the fundus image based on the line feature information. In addition, the embodiment of the present application effectively solves the problem of the lack of preset clarity standards (i.e., clarity standards that can assist doctors in diagnosing eye lesions) of fundus images in clinical application scenarios, is conducive to the standardization of fundus images, and facilitates the control of fundus images used for doctors to read films to reach preset clarity standards.
[0052] Figure 4 The figure is a schematic diagram of a process for determining line feature information corresponding to a fundus image provided by an exemplary embodiment of the present application. Figure 3 The present application is extended based on the embodiment shown Figure 4 The embodiment shown is described below in detail. Figure 4 The embodiment shown and Figure 3 The differences and similarities of the illustrated embodiments are not described in detail.
[0053] like Figure 4 As shown, in the image evaluation method provided in the embodiment of the present application, the step of determining the line feature information corresponding to the fundus image includes the following steps.
[0054] Step 11: extract first line feature information corresponding to the fundus image using a line feature extraction operator.
[0055] Exemplarily, the line feature extraction operator mentioned in step 11 includes any one or any combination of the following: Laplace operator, corner detection algorithm, Zuniga-Haralick positioning operator, Hessian matrix, Log operator. It should be noted that the embodiment of the present application does not impose a unified limitation on the line feature extraction operator.
[0056] Figure 5 The figure shows the fundus image extracted by the meridian feature extraction operator provided by an exemplary embodiment of the present application. Figure 5 The first-line feature information mentioned in the above embodiment is described. Specifically, the first-line feature information includes continuous main vessel feature information and discontinuous branch vessel feature information. Extracting the first-line feature information can effectively remove information that is irrelevant to the evaluation of fundus image clarity. Therefore, the embodiment of the present application can not only reduce the amount of calculation, but also improve the accuracy of the line feature information, thereby improving the accuracy of the clarity of the evaluated fundus image.
[0057] Step 12: determine line feature information based on the first line feature information.
[0058] In actual application, firstly, the first line feature information corresponding to the fundus image is extracted by using the line feature extraction operator, and the line feature information is determined based on the first line feature information, and then the clarity of the fundus image is evaluated based on the line feature information.
[0059] The image evaluation method provided in the embodiment of the present application achieves the purpose of determining the line feature information corresponding to the fundus image by using a line feature extraction operator to extract the first line feature information corresponding to the fundus image, and then determining the line feature information based on the first line feature information. Since the line feature information extraction operation can effectively extract information related to evaluating the clarity of the fundus image, the embodiment of the present application can not only reduce the amount of calculation, but also improve the accuracy of the line feature information, thereby improving the accuracy of the clarity of the evaluated fundus image.
[0060] Figure 6 The figure is a schematic diagram of a process for determining line feature information corresponding to a fundus image provided by another exemplary embodiment of the present application. Figure 4 The present application is extended based on the embodiment shown Figure 6 The embodiment shown is described below in detail. Figure 6 The embodiment shown and Figure 4 The differences and similarities of the illustrated embodiments are not described in detail.
[0061] like Figure 6As shown, in the image evaluation method provided in the embodiment of the present application, after the step of extracting the first line feature information corresponding to the fundus image using the line feature extraction operator, the following steps are also included.
[0062] Step 13: Perform a screening operation on the first line feature information based on the morphological features and topological features corresponding to the first line feature information to generate second line feature information.
[0063] Exemplarily, based on the first line feature information, the angle, length, and other parts that do not conform to the corresponding morphological features and topological features are removed to generate the second line feature information. It should be noted that the angle, length, and other information can be determined manually or with the help of a network training model, and the embodiments of the present application do not impose a unified limitation on this.
[0064] Figure 7 The figure shows a partially enlarged fundus image obtained through a screening operation provided by an exemplary embodiment of the present application. Figure 7 The second line feature information mentioned in the above embodiment is described. The second line feature information is generated by performing a screening operation based on the first line feature information. Specifically, the non-vascular feature information and / or non-optic disc edge feature information in the first line feature information is deleted to generate the second line feature information.
[0065] Step 14: performing a connection operation on the second line feature information to generate third line feature information corresponding to the second line feature information.
[0066] Figure 8 The fundus image obtained by the connection operation provided by an exemplary embodiment of the present application is shown below. Figure 8 The third line feature information mentioned in the above embodiment is described. The third line feature information is generated by performing a connection operation based on the second line feature information. Specifically, the blood vessel feature information and / or optic disc edge feature information that meets the connection conditions in the second line feature information is connected to generate the third line feature information.
[0067] Exemplarily, the blood vessel characteristic information and / or optic disc edge characteristic information meeting the connection condition refers to discontinuous blood vessel characteristic information and / or optic disc edge characteristic information.
[0068] Combined with the above Figure 7 and Figure 8 It can be clearly seen that the third line feature information further improves the accuracy of the line feature information, thereby improving the accuracy of the clarity of the evaluated fundus image, while also reducing the amount of calculation and improving the calculation speed.
[0069] Furthermore, in the image evaluation method provided in the embodiment of the present application, the step of determining the line feature information based on the first line feature information (ie, step 12) includes the following steps.
[0070] Step 15: determine the third line feature information corresponding to the first line feature information as the line feature information.
[0071] In actual application, the line feature extraction operator is first used to extract the first line feature information corresponding to the fundus image, and then the first line feature information is screened based on the morphological features and topological features corresponding to the first line feature information to generate the second line feature information, and then the second line feature information is connected to generate the third line feature information corresponding to the second line feature information, and the third line feature information corresponding to the first line feature information is determined as the line feature information, and the clarity of the fundus image is evaluated based on the line feature information.
[0072] The image evaluation method provided in the embodiment of the present application achieves the purpose of determining the line feature information corresponding to the fundus image by filtering the first line feature information based on the morphological features and topological features corresponding to the first line feature information to generate the second line feature information, then connecting the second line feature information to generate the third line feature information corresponding to the second line feature information, and determining the third line feature information corresponding to the first line feature information as the line feature information. Since the first line feature information is filtered, and the connection operation is performed based on the filtering result, and then the line feature information is determined based on the connection result, the accuracy of the line feature information can be further improved, thereby improving the accuracy of the clarity of the fundus image being evaluated, while also reducing the amount of calculation and improving the calculation speed.
[0073] Fig. 9 FIG. 1 is a flow chart of evaluating the clarity of a fundus image based on line feature information provided by an exemplary embodiment of the present application. Figure 3 The present application is extended based on the embodiment shown Fig. 9 The embodiment shown is described below in detail. Fig. 9 The embodiment shown and Figure 3 The differences and similarities of the illustrated embodiments are not described in detail.
[0074] like Fig. 9 As shown, in the image evaluation method provided in the embodiment of the present application, the step of evaluating the clarity of the fundus image based on line feature information includes the following steps.
[0075] Step 21, calculating the length information corresponding to the line feature information to generate first evaluation information.
[0076] Exemplarily, the length information mentioned in step 21 is the sum of lengths calculated for the line feature information, or it may be a numerical value after mathematical transformation of the sum of lengths, such as performing logarithmic operations and / or exponential operations on the sum of lengths to generate first evaluation information.
[0077] Step 22: Evaluate the clarity of the fundus image based on the first evaluation information.
[0078] Exemplarily, the clarity of the fundus image is evaluated based on the sum of lengths calculated based on the line feature information, or the clarity of the fundus image is evaluated based on a numerical value after mathematical transformation of the sum of lengths calculated based on the line feature information.
[0079] In actual application, the line feature information corresponding to the fundus image is first determined, and then the length information corresponding to the line feature information is calculated to generate first evaluation information, and the clarity of the fundus image is evaluated based on the first evaluation information.
[0080] The image evaluation method provided in the embodiment of the present application achieves the purpose of evaluating the clarity of the fundus image based on the line feature information by calculating the length information corresponding to the line feature information to generate the first evaluation information, and evaluating the clarity of the fundus image based on the first evaluation information. Since the clearer the fundus image is, the more first line feature information is extracted by the line feature extraction operator, so the sum of the lengths corresponding to the line feature information is larger, and the calculated clarity value is also larger, which also reflects that the fundus image is clearer. On the contrary, the more blurred the fundus image is, the less first line feature information is extracted by the line feature extraction operator, and after the screening operation, the second line feature information left is very small, so the sum of the lengths corresponding to the line feature information is smaller, and the calculated clarity value is also smaller, which also reflects that the fundus image is more blurred. Therefore, based on the length information corresponding to the line feature information, the clarity of the fundus image can be accurately determined, so that the fundus image used for reading the film reaches the preset clarity standard.
[0081] Fig.10 FIG. 1 is a flow chart of calculating the length information corresponding to the line feature information provided by an exemplary embodiment of the present application to generate the first evaluation information. Fig. 9 The present application is extended based on the embodiment shown Fig.10 The embodiment shown is described below in detail. Fig.10 The embodiment shown and Fig. 9 The differences and similarities of the illustrated embodiments are not described in detail.
[0082] like Fig.10 As shown, in the image evaluation method provided in the embodiment of the present application, the step of calculating the length information corresponding to the line feature information to generate the first evaluation information includes the following steps.
[0083] Step 211, determining a pixel line with a width of a single pixel corresponding to line feature information based on a preset corrosion operator.
[0084] It should be noted that the line feature information mentioned in the above embodiment is a surface with features such as area and width. The line feature information is converted into a pixel line with a width of a single pixel based on a preset corrosion operator in order to accurately calculate the pixel line length information corresponding to the pixel line. It should be noted that the embodiment of the present application does not impose a unified limitation on the method of calculating the length information corresponding to the line feature information, as long as the length information corresponding to the line feature information can be conveniently and accurately calculated.
[0085] Step 212: Calculate pixel line length information corresponding to the pixel line to generate first evaluation information.
[0086] Exemplarily, the pixel line length information mentioned in step 212 is the sum of lengths calculated for pixel lines with a width of a single pixel, or it may be a value obtained by mathematically transforming the sum of lengths, such as performing logarithmic operations and / or exponential operations on the sum of lengths to generate first evaluation information.
[0087] In actual application, the line feature information corresponding to the fundus image is first determined, and then the pixel line with a width of a single pixel corresponding to the line feature information is determined based on a preset corrosion operator, and the pixel line length information corresponding to the pixel line is calculated to generate first evaluation information, and then the clarity of the fundus image is evaluated based on the first evaluation information.
[0088] The image evaluation method provided in the embodiment of the present application achieves the purpose of calculating the length information corresponding to the line feature information to generate the first evaluation information by determining the pixel line with a width of a single pixel corresponding to the line feature information based on a preset corrosion operator, and calculating the pixel line length information corresponding to the pixel line to generate the first evaluation information, thereby providing a prerequisite for improving the accuracy of the determined length information.
[0089] Fig.11 FIG. 1 is a flow chart of another exemplary embodiment of the present invention for evaluating the clarity of a fundus image based on line feature information. Figure 3 The present application is extended based on the embodiment shown Fig.11 The embodiment shown is described below in detail. Fig.11 The embodiment shown and Figure 3 The differences and similarities of the illustrated embodiments are not described in detail.
[0090] like Fig.11 As shown, in the image evaluation method provided in the embodiment of the present application, the step of evaluating the clarity of the fundus image based on line feature information includes the following steps.
[0091] Step 23: Calculate the gradient information corresponding to the line feature information to generate second evaluation information.
[0092] Exemplarily, the gradient information mentioned in step 23 is obtained by performing partial derivative calculation on the pixel points corresponding to the line feature information, or it may be a numerical value after a mathematical transformation of the partial derivative value, such as performing a square sum operation and / or an absolute value operation and / or a sum operation, and / or a logarithmic operation and / or an exponential operation on the partial derivative value, to generate second evaluation information.
[0093] Step 24 : Evaluate the clarity of the fundus image based on the second evaluation information.
[0094] Exemplarily, the clarity of the fundus image is evaluated based on the partial derivative values of the pixel points corresponding to the line feature information, or the clarity of the fundus image is evaluated based on the numerical values after mathematical transformation of the partial derivative values of the pixel points corresponding to the line feature information.
[0095] In actual application, the line feature information corresponding to the fundus image is first determined, and then the gradient information corresponding to the line feature information is calculated to generate second evaluation information, and the clarity of the fundus image is evaluated based on the second evaluation information.
[0096] The image evaluation method provided in the embodiment of the present application achieves the purpose of evaluating the clarity of the fundus image based on the line feature information by calculating the gradient information corresponding to the line feature information to generate the second evaluation information, and evaluating the clarity of the fundus image based on the second evaluation information. Since the clearer the fundus image, the more complex and changeable the pixel value of the image will be, the greater the change in the image contrast corresponding to the line feature information, the greater the calculated gradient information will be, and the greater the value of the clarity will be, which in turn reflects that the fundus image is clearer. On the contrary, the more blurred the fundus image, the more single the pixel value of the image will be, the smaller the change in the image contrast corresponding to the line feature information, the smaller the calculated gradient information will be, and the smaller the value of the clarity will be, which in turn reflects that the fundus image is more blurred. Therefore, the clarity of the fundus image can be accurately determined based on the gradient information corresponding to the line feature information, so that the fundus image used for reading the film reaches the preset clarity standard.
[0097] Fig.12 FIG. 1 is a flow chart of calculating the gradient information corresponding to the line feature information provided by an exemplary embodiment of the present application to generate the second evaluation information. Fig.11 The present application is extended based on the embodiment shown Fig.12 The embodiment shown is described below in detail. Fig.12 The embodiment shown and Fig.11 The differences and similarities of the illustrated embodiments are not described in detail.
[0098] like Fig.12As shown, in the image evaluation method provided in the embodiment of the present application, the step of calculating the gradient information corresponding to the line feature information to generate the second evaluation information includes the following steps.
[0099] Step 231 : determining the gradient region corresponding to the line feature information based on the grayscale image of the fundus image.
[0100] Exemplarily, the gradient region is a region whose grayscale meets a preset grayscale condition, for example, the gradient region is a region whose grayscale is greater than a preset grayscale threshold.
[0101] It should be noted that the preset grayscale condition mentioned in step 231 is determined by the area of the pixel corresponding to the line feature information.
[0102] Step 232: Calculate the gradient information corresponding to the gradient region to generate second evaluation information.
[0103] Exemplarily, the gradient information mentioned in step 232 is obtained by calculating the partial derivative of the gradient area, or it may be a numerical value after a mathematical transformation of the partial derivative value, such as performing a square sum operation and / or an absolute value operation and / or a sum operation, and / or a logarithmic operation and / or an exponential operation on the partial derivative value to generate second evaluation information.
[0104] In actual application, the line feature information corresponding to the fundus image is first determined, and then the gradient area corresponding to the line feature information is determined based on the grayscale image of the fundus image, wherein the gradient area is an area whose grayscale meets the preset grayscale condition, and the gradient information corresponding to the gradient area is calculated to generate second evaluation information, and then the clarity of the fundus image is evaluated based on the second evaluation information.
[0105] The image evaluation method provided in the embodiment of the present application determines the gradient region corresponding to the line feature information based on the grayscale image of the fundus image, wherein the gradient region is a region whose grayscale meets the preset grayscale condition, and calculates the gradient information corresponding to the gradient region to generate the second evaluation information, thereby achieving the purpose of calculating the gradient information corresponding to the line feature information to generate the second evaluation information. Since the embodiment of the present application further improves the accuracy of the determined gradient information by filtering out the gradient region whose grayscale meets the preset grayscale condition based on the line feature information and calculating the gradient information based on the gradient region.
[0106] Fig.13 FIG. 1 is a flow chart of an image evaluation method provided by another exemplary embodiment of the present application. Figure 3 The present application is extended based on the embodiment shown Fig.13 The embodiment shown is described below in detail. Fig.13 The embodiment shown and Figure 3 The differences and similarities of the illustrated embodiments are not described in detail.
[0107] like Fig.13 As shown, in the image evaluation method provided in the embodiment of the present application, before the step of determining the line feature information corresponding to the fundus image, the following steps are also included.
[0108] Step 30: preprocess the fundus image to generate an area to be evaluated corresponding to the fundus image.
[0109] Exemplarily, preprocessing the fundus image may include: performing denoising on the fundus image. Digital images are often affected by noise interference from imaging devices and external environments during digitization and transmission. These noises may be generated during transmission, and noise is an important cause of image interference. Therefore, before extracting line feature information from the fundus image, it is necessary to remove the noise in the fundus image to avoid interference caused by noise and improve the accuracy of line feature information extraction.
[0110] Furthermore, in the image evaluation method provided in the embodiment of the present application, the step of determining the line feature information corresponding to the fundus image (ie, step 10) includes the following steps.
[0111] Step 40: determining line feature information corresponding to the fundus image based on the area to be evaluated.
[0112] In actual application, the fundus image is first preprocessed to generate an area to be evaluated corresponding to the fundus image, and line feature information corresponding to the fundus image is determined based on the area to be evaluated, and then the clarity of the fundus image is evaluated based on the line feature information.
[0113] The image evaluation method provided in the embodiment of the present application optimizes the fundus image by performing a preprocessing operation on the fundus image before determining the line feature information corresponding to the fundus image to generate an area to be evaluated corresponding to the fundus image, so that its features of interest can be more easily extracted and identified in subsequent processing.
[0114] Fig.14 FIG. 1 is a flow chart of performing a preprocessing operation on a fundus image to generate an area to be evaluated corresponding to the fundus image, provided by an exemplary embodiment of the present application. Fig.13 The present application is extended based on the embodiment shown Fig.14 The embodiment shown is described below in detail. Fig.14 The embodiment shown and Fig.13 The differences and similarities of the illustrated embodiments are not described in detail.
[0115] like Fig.14 As shown, in the image evaluation method provided in the embodiment of the present application, the step of performing a preprocessing operation on the fundus image to generate an area to be evaluated corresponding to the fundus image includes the following steps.
[0116] Step 31 : extracting a region of interest of the fundus image to generate a first evaluation region.
[0117] Exemplarily, the fundus image may include background areas such as name, gender, time, etc. that are unrelated to the line feature information. It is necessary to remove the background area and the invisible fundus area caused by certain reasons (such as eyelid occlusion or large fundus areas that are invisible due to camera lens stains) and extract the area of interest in the fundus image that includes the line feature information.
[0118] Step 32: determine the area to be evaluated based on the first evaluation area.
[0119] In actual application, the region of interest of the fundus image is first extracted to generate a first evaluation region, and the region to be evaluated is determined based on the first evaluation region. Then, the line feature information corresponding to the fundus image is determined based on the region to be evaluated, and then the clarity of the fundus image is evaluated based on the line feature information.
[0120] The image evaluation method provided in the embodiment of the present application achieves the purpose of preprocessing the fundus image to generate the area to be evaluated corresponding to the fundus image by extracting the area of interest of the fundus image to generate the first evaluation area, and determining the area to be evaluated based on the first evaluation area. When collecting fundus images, due to the differences in shooting equipment and shooting techniques of the photographer, the quality of the fundus image may change significantly, making it difficult to process and identify the fundus image later. Therefore, before determining the line feature information corresponding to the fundus image, the area of interest of the fundus image can be extracted so that its features of interest can be more easily extracted and identified in subsequent processing, thereby providing a prerequisite for improving the accuracy of the determined line feature information.
[0121] Fig.15 FIG. 1 is a flow chart of another exemplary embodiment of the present application for performing a preprocessing operation on a fundus image to generate an area to be evaluated corresponding to the fundus image. Fig.14 The present application is extended based on the embodiment shown Fig.15 The embodiment shown is described below in detail. Fig.15 The embodiment shown and Fig.14 The differences and similarities of the illustrated embodiments are not described in detail.
[0122] like Fig.15 As shown, in the image evaluation method provided in the embodiment of the present application, before the step of determining the area to be evaluated based on the first evaluation area, the following steps are also included.
[0123] Step 33: performing a normalization operation on the first evaluation area to generate a second evaluation area.
[0124] For example, normalization is mainly to achieve the unification of differentiated exposure, color and brightness of fundus images, unify the grayscale value range of different images, thereby improving the generalization ability of massive images and making the technology commercialization possible.
[0125] Furthermore, in the image evaluation method provided in the embodiment of the present application, step 32 of determining the area to be evaluated based on the first evaluation area includes the following steps.
[0126] Step 34: determine the second evaluation area corresponding to the first evaluation area as the area to be evaluated.
[0127] In actual application, the region of interest of the fundus image is first extracted to generate a first evaluation region, the first evaluation region is normalized to generate a second evaluation region, and the second evaluation region corresponding to the first evaluation region is determined as the region to be evaluated, and then the line feature information corresponding to the fundus image is determined based on the region to be evaluated, and then the clarity of the fundus image is evaluated based on the line feature information.
[0128] The image evaluation method provided in the embodiment of the present application further optimizes the region of interest of the fundus image by performing a normalization operation on the first evaluation area to generate a second evaluation area, and determines the second evaluation area corresponding to the first evaluation area as the area to be evaluated, thereby providing a prerequisite for further improving the accuracy of the determined line feature information.
[0129] Fig.16 FIG. 1 is a schematic diagram of the structure of an image evaluation device provided by an exemplary embodiment of the present application. Fig.16 As shown, the image evaluation device provided in the embodiment of the present application includes:
[0130] A line feature information determination module 100, used to determine line feature information corresponding to the fundus image;
[0131] The evaluation module 200 is used to evaluate the clarity of the fundus image based on the line feature information.
[0132] In an embodiment of the present application, the line feature information determination module 100 is further used to extract first line feature information corresponding to the fundus image using a line feature extraction operator, and determine the line feature information based on the first line feature information.
[0133] In one embodiment of the present application, the line feature information determination module 100 is also used to perform a screening operation on the first line feature information based on the morphological features and topological features corresponding to the first line feature information to generate second line feature information, perform a connection operation on the second line feature information to generate third line feature information corresponding to the second line feature information, and determine the third line feature information corresponding to the first line feature information as the line feature information.
[0134] In an embodiment of the present application, the evaluation module 200 is further used to calculate the length information corresponding to the line feature information to generate first evaluation information, and evaluate the clarity of the fundus image based on the first evaluation information.
[0135] In one embodiment of the present application, the evaluation module 200 is further used to determine a pixel line with a width of a single pixel corresponding to the line feature information based on a preset corrosion operator, and calculate the pixel line length information corresponding to the pixel line to generate first evaluation information.
[0136] In an embodiment of the present application, the evaluation module 200 is further used to calculate gradient information corresponding to the line feature information to generate second evaluation information, and evaluate the clarity of the fundus image based on the second evaluation information.
[0137] In one embodiment of the present application, the evaluation module 200 is also used to determine the gradient area corresponding to the line feature information based on the grayscale image of the fundus image, wherein the gradient area is an area whose grayscale meets a preset grayscale condition, and calculate the gradient information corresponding to the gradient area to generate second evaluation information.
[0138] Fig.17 The figure shows a schematic diagram of the structure of an image evaluation device provided by another exemplary embodiment of the present application. Fig.16 The present application is extended based on the embodiment shown Fig.17 The embodiment shown is described below in detail. Fig.17 The embodiment shown and Fig.16 The differences and similarities of the illustrated embodiments are not described in detail.
[0139] like Fig.17 As shown, the image evaluation device provided in the embodiment of the present application further includes:
[0140] The preprocessing module 300 is used to perform preprocessing operations on the fundus image to generate an area to be evaluated corresponding to the fundus image, and then determine the line feature information corresponding to the fundus image based on the area to be evaluated using the line feature information determination module 100.
[0141] In an embodiment of the present application, the preprocessing module 300 is further used to extract a region of interest of the fundus image to generate a first evaluation region, and determine a region to be evaluated based on the first evaluation region.
[0142] In an embodiment of the present application, the preprocessing module 300 is further configured to perform a normalization operation on the first evaluation area to generate a second evaluation area, and determine the second evaluation area corresponding to the first evaluation area as the area to be evaluated.
[0143] It should be understood that Figure 16 to Figure 17The operations and functions of the line feature information determination module 100, the evaluation module 200 and the pre-processing module 300 in the provided image evaluation device can refer to the above Figures 3 to 15 The image evaluation method provided is not described here in detail to avoid repetition.
[0144] Below, reference Fig.18 To describe an electronic device according to an embodiment of the present application. Fig.18 Shown is a schematic structural diagram of an electronic device provided by an exemplary embodiment of the present application.
[0145] like Fig.18 As shown, the electronic device 40 includes one or more processors 401 and a memory 402 .
[0146] The processor 401 may be a central processing unit (CPU) or other forms of processing units having data processing capabilities and / or instruction execution capabilities, and may control other components in the electronic device 40 to perform desired functions.
[0147] The memory 402 may include one or more computer program products, which may include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. The volatile memory may include, for example, a random access memory (RAM) and / or a cache memory (cache), etc. The non-volatile memory may include, for example, a read-only memory (ROM), a hard disk, a flash memory, etc. One or more computer program instructions may be stored on the computer-readable storage medium, and the processor 401 may run the program instructions to implement the image evaluation method of each embodiment of the present application described above and / or other desired functions. Various contents such as fundus images including clarity to be evaluated may also be stored in the computer-readable storage medium.
[0148] In one example, the electronic device 40 may further include: an input device 403 and an output device 404, and these components are interconnected via a bus system and / or other forms of connection mechanisms (not shown).
[0149] The input device 403 may include, for example, a keyboard, a mouse, etc.
[0150] The output device 404 can output various information to the outside, including the determined definition information, etc. The output device 404 can include, for example, a display, a speaker, a printer, a communication network and a remote output device connected thereto, and the like.
[0151] Of course, to simplify, Fig.18Only some of the components related to the present application in the electronic device 40 are shown, and components such as a bus, an input / output interface, etc. are omitted. In addition, according to specific application conditions, the electronic device 40 may also include any other appropriate components.
[0152] In addition to the above-mentioned methods and devices, an embodiment of the present application may also be a computer program product, which includes computer program instructions, which, when executed by a processor, enable the processor to execute the steps of the image evaluation method according to various embodiments of the present application described above in this specification.
[0153] The computer program product may be written in any combination of one or more programming languages to write program codes for performing the operations of the embodiments of the present application, including object-oriented programming languages, such as Java, C++, etc., and conventional procedural programming languages, such as "C" language or similar programming languages. The program code may be executed entirely on the user computing device, partially on the user device, as an independent software package, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server.
[0154] In addition, an embodiment of the present application may also be a computer-readable storage medium on which computer program instructions are stored. When the computer program instructions are executed by a processor, the processor executes the steps of the image evaluation method according to various embodiments of the present application described above in this specification.
[0155] The computer readable storage medium can adopt any combination of one or more readable media. The readable medium can be a readable signal medium or a readable storage medium. The readable storage medium can include, for example, but is not limited to, a system, device or device of electricity, magnetism, light, electromagnetic, infrared, or semiconductor, or any combination of the above. More specific examples (non-exhaustive list) of readable storage media include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.
[0156] The basic principles of the present application are described above in conjunction with specific embodiments. However, it should be noted that the advantages, strengths, effects, etc. mentioned in the present application are only examples and not limitations, and it cannot be considered that these advantages, strengths, effects, etc. are required by each embodiment of the present application. In addition, the specific details disclosed above are only for the purpose of illustration and ease of understanding, not for limitation, and the above details do not limit the present application to being implemented by adopting the above specific details.
[0157] The block diagrams of the devices, apparatuses, equipment, and systems involved in this application are only illustrative examples and are not intended to require or imply that they must be connected, arranged, and configured in the manner shown in the block diagram. As will be appreciated by those skilled in the art, these devices, apparatuses, equipment, and systems can be connected, arranged, and configured in any manner. Words such as "including", "comprising", "having", etc. are open words, referring to "including but not limited to", and can be used interchangeably with them. The words "or" and "and" used here refer to the words "and / or" and can be used interchangeably with them, unless the context clearly indicates otherwise. The words "such as" used here refer to the phrase "such as but not limited to", and can be used interchangeably with them.
[0158] It should also be noted that in the apparatus, device and method of the present application, each component or each step can be decomposed and / or recombined. Such decomposition and / or recombination should be regarded as equivalent solutions of the present application.
[0159] The above description of the disclosed aspects is provided to enable any person skilled in the art to make or use the present application. Various modifications to these aspects will be readily apparent to those skilled in the art, and the general principles defined herein may be applied to other aspects without departing from the scope of the present application. Therefore, the present application is not intended to be limited to the aspects shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.
[0160] The above description has been given for the purpose of illustration and description. In addition, this description is not intended to limit the embodiments of the present application to the forms disclosed herein. Although multiple example aspects and embodiments have been discussed above, those skilled in the art will recognize certain variations, modifications, changes, additions and sub-combinations thereof.
Claims
1. An image evaluation method, characterized in that: include: Determining line feature information corresponding to the fundus image, wherein the line feature information includes blood vessel feature information and / or optic disc edge feature information; evaluating the clarity of the fundus image based on the line feature information; Wherein, determining the line feature information corresponding to the fundus image includes: Extracting first line feature information corresponding to the fundus image using a line feature extraction operator, wherein the first line feature information includes continuous trunk vessel feature information, discontinuous branch vessel feature information, and non-vessel feature information; performing a screening operation on the first line feature information based on the morphological features and topological features corresponding to the first line feature information, deleting the non-vascular feature information and / or non-optic disc edge feature information in the first line feature information, so as to generate second line feature information; performing a connection operation on the second line feature information to generate third line feature information corresponding to the second line feature information; determining the third line feature information corresponding to the first line feature information as the line feature information; The filtering operation on the first line feature information based on the morphological features and topological features corresponding to the first line feature information to generate the second line feature information includes: Based on the first line feature information, removing a portion whose angle and length do not conform to the morphological features and topological features corresponding to the first line feature information, so as to generate the second line feature information; The step of performing a connection operation on the second line feature information includes: The interrupted blood vessel feature information and / or optic disc edge feature information in the second line feature information is connected.
2. The image evaluation method according to claim 1, characterized in that The step of evaluating the clarity of the fundus image based on the line feature information comprises: Calculating length information corresponding to the line feature information to generate first evaluation information; The clarity of the fundus image is evaluated based on the first evaluation information.
3. The image evaluation method according to claim 2, characterized in that: The calculating the length information corresponding to the line feature information to generate first evaluation information includes: Determine a pixel line having a width of a single pixel corresponding to the line feature information based on a preset corrosion operator; The pixel line length information corresponding to the pixel line is calculated to generate the first evaluation information.
4. The image evaluation method according to claim 1, characterized in that: The step of evaluating the clarity of the fundus image based on the line feature information comprises: Calculating gradient information corresponding to the line feature information to generate second evaluation information; The clarity of the fundus image is evaluated based on the second evaluation information.
5. The image evaluation method according to claim 4, characterized in that The calculating the gradient information corresponding to the line feature information to generate second evaluation information includes: Determine a gradient region corresponding to the line feature information based on a grayscale image of the fundus image, wherein the gradient region is a region whose grayscale meets a preset grayscale condition; The gradient information corresponding to the gradient region is calculated to generate the second evaluation information.
6. An image evaluation device, characterized in that include: A line feature information determination module, used to determine line feature information corresponding to the fundus image, wherein the line feature information includes blood vessel feature information and / or optic disc edge feature information; An evaluation module, configured to evaluate the clarity of the fundus image based on the line feature information; Wherein, determining the line feature information corresponding to the fundus image includes: Extracting first line feature information corresponding to the fundus image using a line feature extraction operator, wherein the first line feature information includes continuous trunk vessel feature information, discontinuous branch vessel feature information, and non-vessel feature information; performing a screening operation on the first line feature information based on the morphological features and topological features corresponding to the first line feature information, deleting the non-vascular feature information and / or non-optic disc edge feature information in the first line feature information, so as to generate second line feature information; performing a connection operation on the second line feature information to generate third line feature information corresponding to the second line feature information; determining the third line feature information corresponding to the first line feature information as the line feature information; The filtering operation on the first line feature information based on the morphological features and topological features corresponding to the first line feature information to generate the second line feature information includes: Based on the first line feature information, removing a portion whose angle and length do not conform to the morphological features and topological features corresponding to the first line feature information, so as to generate the second line feature information; The step of performing a connection operation on the second line feature information includes: The interrupted blood vessel feature information and / or optic disc edge feature information in the second line feature information is connected.
7. A computer-readable storage medium, characterized in that: The storage medium stores a computer program, and the computer program is used to execute the image evaluation method according to any one of claims 1 to 5.
8. An electronic device, characterized in that: The electronic device comprises: processor; a memory for storing instructions executable by the processor; The processor is used to execute the image evaluation method described in any one of claims 1 to 5.
