CN111588345A - Eye disease detection method, AR glasses and readable storage medium - Google Patents
Eye disease detection method, AR glasses and readable storage medium Download PDFInfo
- Publication number
- CN111588345A CN111588345A CN202010558496.XA CN202010558496A CN111588345A CN 111588345 A CN111588345 A CN 111588345A CN 202010558496 A CN202010558496 A CN 202010558496A CN 111588345 A CN111588345 A CN 111588345A
- Authority
- CN
- China
- Prior art keywords
- image
- eye
- diagnosis
- human eye
- diagnostic
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 208000030533 eye disease Diseases 0.000 title claims abstract description 108
- 239000011521 glass Substances 0.000 title claims abstract description 37
- 238000001514 detection method Methods 0.000 title claims abstract description 27
- 238000003745 diagnosis Methods 0.000 claims abstract description 141
- 241000282414 Homo sapiens Species 0.000 claims abstract description 124
- 238000000034 method Methods 0.000 claims abstract description 41
- 238000012545 processing Methods 0.000 claims abstract description 17
- 238000012549 training Methods 0.000 claims description 76
- 230000008569 process Effects 0.000 claims description 10
- 208000022873 Ocular disease Diseases 0.000 claims description 8
- 201000010099 disease Diseases 0.000 claims description 7
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 claims description 7
- 210000001508 eye Anatomy 0.000 description 131
- 208000002177 Cataract Diseases 0.000 description 5
- 238000001228 spectrum Methods 0.000 description 5
- 230000008859 change Effects 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 230000006870 function Effects 0.000 description 4
- 201000004569 Blindness Diseases 0.000 description 3
- 238000004891 communication Methods 0.000 description 3
- 208000010412 Glaucoma Diseases 0.000 description 2
- 241000209149 Zea Species 0.000 description 2
- 235000005824 Zea mays ssp. parviglumis Nutrition 0.000 description 2
- 235000002017 Zea mays subsp mays Nutrition 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 2
- 210000005252 bulbus oculi Anatomy 0.000 description 2
- 235000005822 corn Nutrition 0.000 description 2
- 230000004438 eyesight Effects 0.000 description 2
- 206010047571 Visual impairment Diseases 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000004925 denaturation Methods 0.000 description 1
- 230000036425 denaturation Effects 0.000 description 1
- 229940079593 drug Drugs 0.000 description 1
- 239000003814 drug Substances 0.000 description 1
- 238000013399 early diagnosis Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 210000000744 eyelid Anatomy 0.000 description 1
- 230000002349 favourable effect Effects 0.000 description 1
- 238000002483 medication Methods 0.000 description 1
- 208000030159 metabolic disease Diseases 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 210000001328 optic nerve Anatomy 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000010827 pathological analysis Methods 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 230000002265 prevention Effects 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 102000004169 proteins and genes Human genes 0.000 description 1
- 108090000623 proteins and genes Proteins 0.000 description 1
- 210000001525 retina Anatomy 0.000 description 1
- 238000001356 surgical procedure Methods 0.000 description 1
- 208000024891 symptom Diseases 0.000 description 1
- 208000029257 vision disease Diseases 0.000 description 1
- 230000004382 visual function Effects 0.000 description 1
- 230000004393 visual impairment Effects 0.000 description 1
Images
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B3/00—Apparatus for testing the eyes; Instruments for examining the eyes
- A61B3/10—Objective types, i.e. instruments for examining the eyes independent of the patients' perceptions or reactions
- A61B3/14—Arrangements specially adapted for eye photography
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B3/00—Apparatus for testing the eyes; Instruments for examining the eyes
- A61B3/0016—Operational features thereof
- A61B3/0025—Operational features thereof characterised by electronic signal processing, e.g. eye models
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B3/00—Apparatus for testing the eyes; Instruments for examining the eyes
- A61B3/10—Objective types, i.e. instruments for examining the eyes independent of the patients' perceptions or reactions
- A61B3/14—Arrangements specially adapted for eye photography
- A61B3/145—Arrangements specially adapted for eye photography by video means
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0012—Biomedical image inspection
- G06T7/0014—Biomedical image inspection using an image reference approach
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
- G06T2207/30041—Eye; Retina; Ophthalmic
Landscapes
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Engineering & Computer Science (AREA)
- Medical Informatics (AREA)
- General Health & Medical Sciences (AREA)
- Physics & Mathematics (AREA)
- Animal Behavior & Ethology (AREA)
- Veterinary Medicine (AREA)
- Heart & Thoracic Surgery (AREA)
- Ophthalmology & Optometry (AREA)
- Molecular Biology (AREA)
- Surgery (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Public Health (AREA)
- Multimedia (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Radiology & Medical Imaging (AREA)
- Quality & Reliability (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Signal Processing (AREA)
- Eye Examination Apparatus (AREA)
Abstract
The invention discloses an eye disease detection method, AR glasses and a readable storage medium, wherein the method comprises the following steps: acquiring a human eye image to be detected; processing the human eye image to obtain a diagnostic image, wherein the diagnostic image comprises at least one of an amplitude spectrogram corresponding to the human eye image and a phase spectrogram corresponding to the human eye image; and determining a diagnosis result of the eye according to the diagnosis image, and outputting the diagnosis result. The diagnosis device can automatically diagnose whether the user has the eye disease through the human eye image without manually diagnosing whether the user has the eye disease, so that the detection efficiency of the eye disease of the user is high.
Description
Technical Field
The present invention relates to the field of image processing technologies, and in particular, to an eye disease detection method, AR glasses, and a readable storage medium.
Background
Cataract and the like are eye diseases caused by blindness of human beings, and the eye diseases threaten and damage the visual function of optic nerves, not only reduce the vision of human beings and make the eyes of human beings blurred, but also cause blindness of patients when the eyes are serious. The cataract patient has the symptoms of lens metabolism disorder, the lens becomes turbid due to protein denaturation, and light rays blocked by the turbid lens cannot be projected on retina, so that the vision is blurred. Prevention of further visual impairment through early diagnosis and selection of appropriate medications and surgery is critical to reducing the incidence of blindness.
In the prior art, the detection method of eye diseases such as cataract is mainly to manually observe the eye region by a detector and perform pathological analysis, so as to obtain the detection result of the eye region. The manual detection of eye diseases by doctors leads to low detection efficiency of eye diseases.
Disclosure of Invention
The embodiment of the application provides the eye disease detection method, the AR glasses and the readable storage medium, and the accuracy of eye disease detection is achieved.
The embodiment of the application provides an eye disease detection method, which comprises the following steps:
acquiring a human eye image to be detected;
processing the human eye image to obtain a diagnostic image, wherein the diagnostic image comprises at least one of an amplitude spectrogram corresponding to the human eye image and a phase spectrogram corresponding to the human eye image;
and determining a diagnosis result of the eye according to the diagnosis image, and outputting the diagnosis result.
In one embodiment, the step of processing the image of the human eye to obtain a diagnostic image comprises:
windowing the human eye image;
performing Fourier transform on the windowed human eye image to obtain at least one of the amplitude spectrogram and the phase spectrogram;
determining the diagnostic image from at least one of the magnitude and phase spectrograms.
In an embodiment, the step of determining the diagnostic image according to at least one of the magnitude spectrogram and the phase spectrogram further comprises:
adjusting a low-frequency area in the amplitude spectrogram to a first preset position corresponding to the low-frequency area, and adjusting a high-frequency area in the amplitude spectrogram to a second preset position corresponding to the high-frequency area;
the step of determining the diagnostic image from at least one of the magnitude and phase spectrograms comprises:
determining the diagnostic image according to the adjusted at least one of the magnitude spectrogram and the phase spectrogram.
In one embodiment, the step of determining a diagnosis of the eye from the diagnostic image comprises:
comparing the diagnostic image with a preset reference image to obtain a comparison result, wherein the reference image is determined according to a human eye image of a patient with eye diseases;
and determining the diagnosis result according to the comparison result.
In one embodiment, the step of determining the diagnosis result according to the alignment result comprises:
determining the eye disease corresponding to the reference image as the diagnosis result when the comparison result is that the difference value between the parameter in the diagnosis image and the parameter in the reference image is smaller than or equal to a preset threshold value;
and determining the disease of the uninfected eye as the diagnosis result when the comparison result is that the difference value between the parameters in the diagnosis image and the parameters in the reference image is larger than a preset threshold value.
In an embodiment, before the step of acquiring the image of the human eye to be detected, the method further includes:
training a diagnosis model according to the human eye image of the eye disease patient;
the step of training a diagnostic model based on an image of a human eye of a patient with eye disease comprises:
inputting the diagnostic image to the diagnostic model;
and obtaining a diagnosis result output by the diagnosis model.
In one embodiment, the step of training a diagnostic model based on an image of a human eye of a patient with eye disease comprises:
acquiring a training sample set, wherein the training sample set comprises diagnostic training images corresponding to human eye images of a plurality of eye disease patients;
inputting each diagnosis training image in the training sample set into a preset training model so as to train the training model;
and stopping training the training model when the convergence value of the training model is determined to be smaller than a preset threshold value, and storing the training model which is stopped to be trained as the diagnosis model.
In one embodiment, the step of acquiring an image of a human eye to be detected includes:
acquiring the to-be-detected eye image collected by a camera carried by AR glasses;
the step of processing the human eye image to obtain a diagnostic image comprises:
controlling the AR glasses to process the human eye image to obtain a diagnostic image;
the step of determining a diagnosis result of the eye from the diagnosis image and outputting the diagnosis result includes:
and controlling the AR glasses to determine a diagnosis result of the eye according to the diagnosis image, and controlling the AR glasses to output the diagnosis result.
In one embodiment, the step of controlling the AR glasses to output the diagnosis result includes:
controlling the AR glasses to display the diagnosis result on lenses thereof.
In addition, in order to achieve the above object, the present invention further provides an AR glasses, which includes an image capturing module, a memory, a processor, and an eye disease detection program stored in the memory and executable on the processor, wherein the eye disease detection program, when executed by the processor, implements the steps of the eye disease detection method as described above.
Further, to achieve the above object, the present invention also provides a readable storage medium having stored thereon an eye disease detection program, which when executed by a processor, implements the steps of the eye disease detection method as described above.
According to the eye disease detection method, the AR glasses and the readable storage medium, the human eye image to be detected is obtained through the diagnosis device, and the human eye image is processed to obtain the diagnosis image, wherein the diagnosis image comprises at least one of an amplitude spectrogram and a phase spectrogram corresponding to the human eye image; and determining a diagnosis result of the eye according to the diagnosis image, and outputting the diagnosis result. The diagnosis device can automatically diagnose whether the user has the eye disease through the human eye image without manually diagnosing whether the user has the eye disease, so that the detection efficiency of the eye disease of the user is high.
Drawings
Fig. 1 is a schematic hardware configuration diagram of a diagnostic apparatus according to an embodiment of the present invention;
FIG. 2 is a schematic flow chart of a method for detecting an eye disease according to a first embodiment of the present invention;
FIG. 3 is a schematic flow chart of a method for detecting an eye disease according to a second embodiment of the present invention;
FIG. 4 is a schematic flow chart of a method for detecting an eye disease according to a third embodiment of the present invention;
FIG. 5 is a schematic flow chart of a method for detecting an eye disease according to a fourth embodiment of the present invention;
FIG. 6 is a schematic flow chart of a fifth embodiment of the method for detecting an eye disease of the present invention;
FIG. 7 is a schematic flowchart of a method for detecting an eye disease according to a sixth embodiment of the present invention;
the implementation, functional features and advantages of the objects of the present invention will be further explained with reference to the accompanying drawings.
Detailed Description
It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
The main solution of the embodiment of the invention is as follows: acquiring a human eye image to be detected; processing the human eye image to obtain a diagnostic image, wherein the diagnostic image comprises at least one of an amplitude spectrogram corresponding to the human eye image and a phase spectrogram corresponding to the human eye image; and determining a diagnosis result of the eye according to the diagnosis image, and outputting the diagnosis result.
The diagnosis device can automatically diagnose whether the user has the eye disease through the human eye image without manually diagnosing whether the user has the eye disease, so that the detection efficiency of the eye disease of the user is high.
As an implementation, the diagnostic device may be as shown in fig. 1.
An embodiment of the present invention relates to a diagnostic apparatus including: a processor 101, e.g. a CPU, a memory 102, a communication bus 103. Wherein a communication bus 103 is used for enabling the connection communication between these components.
The memory 102 may be a high-speed RAM memory or a non-volatile memory (e.g., a disk memory). As shown in fig. 1, a memory 102, which is a readable storage medium, may include therein an eye disease detection program; and the processor 101 may be configured to call the eye disease detection program stored in the memory 102 and perform the following operations:
acquiring a human eye image to be detected;
processing the human eye image to obtain a diagnostic image, wherein the diagnostic image comprises at least one of an amplitude spectrogram corresponding to the human eye image and a phase spectrogram corresponding to the human eye image;
and determining a diagnosis result of the eye according to the diagnosis image, and outputting the diagnosis result.
In one embodiment, the processor 101 may be configured to call an eye disease detection program stored in the memory 102 and perform the following operations:
windowing the human eye image;
performing Fourier transform on the windowed human eye image to obtain at least one of the amplitude spectrogram and the phase spectrogram;
determining the diagnostic image from at least one of the magnitude and phase spectrograms.
In one embodiment, the processor 101 may be configured to call an eye disease detection program stored in the memory 102 and perform the following operations:
adjusting a low-frequency area in the amplitude spectrogram to a first preset position corresponding to the low-frequency area, and adjusting a high-frequency area in the amplitude spectrogram to a second preset position corresponding to the high-frequency area;
the step of determining the diagnostic image from at least one of the magnitude and phase spectrograms comprises:
determining the diagnostic image according to the adjusted at least one of the magnitude spectrogram and the phase spectrogram.
In one embodiment, the processor 101 may be configured to call an eye disease detection program stored in the memory 102 and perform the following operations:
comparing the diagnostic image with a preset reference image to obtain a comparison result, wherein the reference image is determined according to a human eye image of a patient with eye diseases;
and determining the diagnosis result according to the comparison result.
In one embodiment, the processor 101 may be configured to call an eye disease detection program stored in the memory 102 and perform the following operations:
determining the eye disease corresponding to the reference image as the diagnosis result when the comparison result is that the difference value between the parameter in the diagnosis image and the parameter in the reference image is smaller than or equal to a preset threshold value;
and determining the disease of the uninfected eye as the diagnosis result when the comparison result is that the difference value between the parameters in the diagnosis image and the parameters in the reference image is larger than a preset threshold value.
In one embodiment, the processor 101 may be configured to call an eye disease detection program stored in the memory 102 and perform the following operations:
training a diagnosis model according to the human eye image of the eye disease patient;
the step of training a diagnostic model based on an image of a human eye of a patient with eye disease comprises:
inputting the diagnostic image to the diagnostic model;
and obtaining a diagnosis result output by the diagnosis model.
In one embodiment, the processor 101 may be configured to call an eye disease detection program stored in the memory 102 and perform the following operations:
acquiring a training sample set, wherein the training sample set comprises diagnostic training images corresponding to human eye images of a plurality of eye disease patients;
inputting each diagnosis training image in the training sample set into a preset training model so as to train the training model;
and stopping training the training model when the convergence value of the training model is determined to be smaller than a preset threshold value, and storing the training model which is stopped to be trained as the diagnosis model.
In one embodiment, the processor 101 may be configured to call an eye disease detection program stored in the memory 102 and perform the following operations:
acquiring the to-be-detected eye image collected by a camera carried by AR glasses;
controlling the AR glasses to process the human eye image to obtain a diagnostic image;
and controlling the AR glasses to determine a diagnosis result of the eye according to the diagnosis image, and controlling the AR glasses to output the diagnosis result.
In one embodiment, the processor 101 may be configured to call an eye disease detection program stored in the memory 102 and perform the following operations:
controlling the AR glasses to display the diagnosis result on lenses thereof.
According to the scheme, the human eye image to be detected is obtained through the diagnosis device, and the human eye image is processed to obtain a diagnosis image, wherein the diagnosis image comprises at least one of an amplitude spectrogram and a phase spectrogram corresponding to the human eye image; and determining a diagnosis result of the eye according to the diagnosis image, and outputting the diagnosis result. The diagnosis device can automatically diagnose whether the user has the eye disease through the human eye image without manually diagnosing whether the user has the eye disease, so that the detection efficiency of the eye disease of the user is high.
Based on the hardware architecture of the above-mentioned diagnostic device, an embodiment of the method for detecting an ocular disease of the present invention is proposed.
Referring to fig. 2, fig. 2 is a diagram illustrating a method for detecting an eye disease according to a first embodiment of the present invention, the method comprising the steps of:
step S10: acquiring a human eye image to be detected;
specifically, the human eye image to be detected is a human eye image corresponding to an eye region, and the eye region is a region for detecting an eye disease, including but not limited to an eyeball, an eyelid, or a skin around the eye. The human eye image to be detected can be a human eye image collected through a camera, and the specific application scene can be a human eye image collected through a camera on AR glasses when the AR glasses are used. The human eye image to be detected can be an original image or an image obtained by carrying out preprocessing such as denoising on the original image; the human eye image to be detected can be a color image or a gray scale image; the human eye image to be detected can be a single-frame image or a multi-frame image. The multi-frame image can be a multi-frame image shot when the orientation of eyes in the eye region is adjusted, and detection of the eye region is facilitated to be carried out on the whole face.
The human eye image corresponding to the eye region can be obtained through the diagnosis device, and the human eye image is analyzed to obtain a diagnosis result. The diagnostic device may be an image capture device with an image capture module, the image capture device may be an AR glasses, and the image capture module may be a high definition camera. The diagnosis device can also be connected with high-definition camera background processing equipment.
Step S20: processing the human eye image to obtain a diagnostic image, wherein the diagnostic image comprises at least one of an amplitude spectrogram corresponding to the human eye image and a phase spectrogram corresponding to the human eye image;
specifically, the diagnostic image is a spectrogram corresponding to a human eye image, wherein the spectrogram may be an amplitude spectrogram and/or a phase spectrogram. The point of the magnitude spectrogram represents the magnitude of the difference between the point and the neighboring point, i.e., the magnitude of the gradient, or the magnitude of the frequency of the point. The amplitude spectrogram is an image representing amplitude information of the human eye image, and if the number of dark points in the amplitude spectrogram is more, the human eye image is softer, otherwise, the human eye image is sharper. The phase spectrogram is an image representing phase information corresponding to the human eye image, and whether the eyeball has a position change or not can be judged according to the phase spectrogram.
Step S30: and determining a diagnosis result of the eye according to the diagnosis image, and outputting the diagnosis result.
Specifically, the diagnosis result is the result of detecting the eye disease, and the diagnosis result includes, but is not limited to, whether the eye is affected, and the type of the affected eye disease. The diagnosis may include information on various eye diseases, such as cataract, glaucoma, or corn blowholes, and the severity of the disease.
And after the diagnosis result is determined, outputting the diagnosis result according to a target output mode, wherein the target output mode comprises at least one of character output, image output and voice output.
Specifically, the target output mode is an output mode for outputting the diagnosis result, and the target output mode comprises character output, image output and/or voice output, wherein the character output represents the diagnosis result in a character form; the image output is to represent the diagnosis result in a picture form; the voice output is to represent the diagnosis result in a voice mode and output a voice template recorded in advance. The form problem of the output diagnosis result is effectively solved, and the diagnosis result is clearly displayed in the modes of characters, pictures and/or voice and the like.
The eye disease detection method of this embodiment may be applied to AR glasses, and the step S10 further includes: acquiring the to-be-detected eye image collected by a camera carried by AR glasses; the step S20 further includes: controlling the AR glasses to process the human eye image to obtain a diagnostic image; the step S30 further includes: and controlling the AR glasses to determine a diagnosis result of the eye according to the diagnosis image, and controlling the AR glasses to output the diagnosis result. Wherein controlling the AR glasses to output the diagnosis result may be controlling the AR glasses to display the diagnosis result on lenses thereof.
In the technical scheme provided by this embodiment, a human eye image to be detected is acquired through a diagnostic device, and the human eye image is processed to obtain a diagnostic image, wherein the diagnostic image includes at least one of an amplitude spectrogram and a phase spectrogram corresponding to the human eye image; and determining a diagnosis result of the eye according to the diagnosis image, and outputting the diagnosis result. The diagnosis device can automatically diagnose whether the user has the eye disease through the human eye image without manually diagnosing whether the user has the eye disease, so that the detection efficiency of the eye disease of the user is high.
Referring to fig. 3, fig. 3 is a diagram illustrating a second embodiment of the method for detecting an eye disease according to the present invention, wherein the step S20 further includes:
step S21: and windowing the human eye image.
Specifically, after a human eye image is acquired, windowing is required to be performed on the human eye image; and determining the diagnostic image by using the human eye image subjected to windowing processing. The windowing process is an image processing method for processing an image for a person by a window function. Due to the fact that the discontinuity of the edges of the human eye images can cause the amplitude spectrogram after Fourier transform to be complex and disordered, namely, a frequency spectrum leakage phenomenon occurs, and the amplitude spectrogram is not easy to analyze. Therefore, when image processing is carried out, a proper window function is added, namely windowing processing is carried out, so that the amplitude spectrogram is clearer and more concentrated in distribution. The window functions for performing the windowing process include, but are not limited to, window functions such as a hanning window, a casser window, and/or a hamming window.
Step S22: performing Fourier transform on the windowed human eye image to obtain at least one of the amplitude spectrogram and the phase spectrogram; determining the diagnostic image from at least one of the magnitude and phase spectrograms.
Specifically, the diagnostic image corresponding to the human eye image is determined by performing fourier transform on the human eye image, generating an amplitude spectrogram according to an amplitude parameter after fourier transform, or generating a phase spectrogram according to a phase parameter after fourier transform, so as to obtain at least one of the amplitude spectrogram and the phase spectrogram, and determining the diagnostic image corresponding to the human eye image according to at least one of the amplitude spectrogram and the phase spectrogram.
In the technical scheme provided by the embodiment, due to the adoption of the technical means of Fourier transform of the human eye image, the corresponding amplitude spectrogram and phase spectrogram of the human eye image are determined; the technical means of windowing the image is adopted, the spectrum leakage phenomenon caused by the discontinuity of the edge of the human eye image is effectively solved, the obtained amplitude parameter is simpler and more standard, and the human eye image subjected to windowing is used for determining the amplitude spectrogram, so that the analysis of the subsequent amplitude spectrogram is facilitated.
Referring to fig. 4, fig. 4 is a diagram illustrating a third embodiment of the method for detecting an eye disease according to the present invention, and based on the second embodiment, before the step S22, the method further includes:
step S23: and adjusting the low-frequency area in the amplitude spectrogram to a first preset position corresponding to the low-frequency area, and adjusting the high-frequency area in the amplitude spectrogram to a second preset position corresponding to the high-frequency area.
Specifically, the low-frequency area represents a region with a smaller frequency in the amplitude spectrogram, corresponds to a place with a smaller gray scale change in the human eye image, belongs to a part inside or outside the edge information of the object in the human eye image, and can represent the outline of the object in the human eye image. The high-frequency area represents a region with a larger frequency in the amplitude spectrogram, and corresponds to a place with a larger gray scale change in the human eye image, generally referred to as edge information, which can represent the details and the texture of the human eye image.
Because the fourier transform has symmetry, the amplitude spectrogram to be determined obtained by processing the human eye image often uses the center coordinate of the spectrum parameter as an origin, and the upper left is symmetrical with the lower right, and the upper right is symmetrical with the lower left. After the adjustment of step S23, the center of the amplitude spectrogram is the average brightness of the human eye image, the frequency is 0, and from the center of the amplitude parameter to the outside, the frequency increases, the highlight indicates that the frequency characteristic is more obvious, and the frequency change direction of the center of the amplitude spectrogram is perpendicular to the direction of the ground object in the human eye image. The adjusted amplitude spectrogram can more clearly display the ratio of the low-frequency area to the high-frequency area, so that the diseased condition of the eye area can be conveniently determined.
The step S22 further includes:
step S221: determining the diagnostic image according to the adjusted at least one of the magnitude spectrogram and the phase spectrogram.
Specifically, the diagnostic image is determined from at least one of the amplitude spectrogram and the phase spectrogram adjusted in step S23.
In the technical scheme provided by this embodiment, since the low-frequency area represents most of the effective information of the human eye image, the low-frequency area is centrally adjusted to a certain position of the amplitude spectrogram, and the problem that the low-frequency area in the amplitude spectrogram is not centrally distributed is effectively solved, so that the effective information is centralized at a certain position, and the adjusted amplitude spectrogram is used for analysis, which is favorable for accurately determining a diagnosis result.
Referring to fig. 5, fig. 5 is a fourth embodiment of the method for detecting an eye disease according to the present invention, and based on any one of the first to third embodiments, the step S30 further includes:
step S31: and comparing the diagnostic image with a preset reference image to obtain a comparison result, wherein the reference image is determined according to the human eye image of the eye disease patient.
Specifically, the preset reference image is a spectrogram corresponding to a human eye image with eye diseases. The reference image includes at least one of an amplitude reference map and a phase reference map. The amplitude reference image is an amplitude spectrum image corresponding to the human eye image with eye diseases, and the phase reference image is a phase spectrum image corresponding to the human eye image with eye diseases.
Step S32: and determining the diagnosis result according to the comparison result.
Specifically, the comparison result between the diagnostic image and the reference image refers to a difference between the diagnostic image and the reference image, wherein the comparison result between the diagnostic image and the preset reference image refers to a comparison result between the amplitude spectrogram and the amplitude reference image, and/or a comparison result between the phase spectrogram and the phase reference image. And comparing the diagnostic image with the reference image to determine a diagnostic result corresponding to the diagnostic image. The comparison result of the comparison between the diagnostic image and the reference image can be the ratio result of the high-frequency area and the low-frequency area in the amplitude spectrogram, and the ratio result of the high-frequency area and the low-frequency area in the amplitude reference image; a comparison between the phase case in the phase spectrogram and the phase case in the phase reference spectrogram can be made.
The step S32 further includes:
step S321: determining the eye disease corresponding to the reference image as the diagnosis result when the comparison result is that the difference value between the parameter in the diagnosis image and the parameter in the reference image is smaller than or equal to a preset threshold value;
specifically, the difference between the parameter in the diagnostic image and the parameter in the reference image, which represents the difference between the diagnostic image and the reference image, may be the difference between the ratio of the high-frequency area to the low-frequency area in the amplitude spectrogram and the ratio of the high-frequency area to the low-frequency area in the amplitude reference image; may be the difference between the phase case in the phase spectrogram and the phase case in the phase reference image. The preset threshold is a numerical value or a numerical range for judging whether the diagnostic image belongs to the eye disease corresponding to the reference image. The eye disease corresponding to the reference image may include various eye diseases, such as cataract, glaucoma or corn.
Step S322: and determining the disease of the uninfected eye as the diagnosis result when the comparison result is that the difference value between the parameters in the diagnosis image and the parameters in the reference image is larger than a preset threshold value.
Specifically, when the difference between the parameter in the diagnostic image and the parameter in the reference image is greater than the preset threshold, it represents that the difference between the diagnostic image and the reference image is large and does not belong to the eye disease type corresponding to the reference image, so as to output the diagnosis result of the eye disease.
In the technical solution provided in this embodiment, a technical means of comparing the parameters in the diagnostic image with the preset parameters in the reference image to obtain a comparison result and determining the diagnostic result according to the comparison result is adopted, and the diagnostic result of the disease is determined by comparing the amplitude spectrogram and the amplitude reference image and/or comparing the phase spectrogram and the phase reference image, so that the disease detection of the eye region is realized.
Referring to fig. 6, fig. 6 is a fifth embodiment of the method for detecting an eye disease according to the present invention, and based on any one of the first to third embodiments, before the step S10, the method further includes:
step S40: and training a diagnosis model according to the human eye image of the eye disease patient.
Specifically, the diagnosis device is provided with a diagnosis model which is obtained by training according to human eye images of a large number of eye disease patients.
The step S30 further includes:
step S33: inputting the diagnostic image to the diagnostic model.
Specifically, the diagnostic model is a functional module trained according to an image of a human eye suffering from eye diseases, and the diagnostic model may be a trained neural network model. The diagnosis model analyzes the input diagnosis image and outputs a diagnosis result corresponding to the diagnosis image, wherein the diagnosis model is obtained by training a large number of human eye images suffering from eye diseases, and the diagnosis model can continuously learn the human eye images suffering from the eye diseases.
Step S34: and obtaining a diagnosis result output by the diagnosis model.
Specifically, the diagnostic model analyzes the diagnostic image, and the diagnostic model sets the output layer as a diagnostic result, so that the diagnostic result can be directly obtained through the diagnostic model.
In the technical scheme provided by the embodiment, due to the adoption of the technical means of the diagnosis model, the diagnosis image is analyzed through the diagnosis model, and the diagnosis result is obtained, so that the efficiency of obtaining the diagnosis result is improved.
Referring to fig. 7, fig. 7 is a sixth embodiment of the method for detecting an eye disease according to the present invention, and based on the fifth embodiment, the step S40 further includes:
s401: acquiring a training sample set, wherein the training sample set comprises diagnostic training images corresponding to human eye images of a plurality of eye disease patients;
specifically, the training sample set is data for training the training model, and the diagnostic training chart included in the training sample set is a spectrogram corresponding to a human eye image with eye disease, and may be an amplitude spectrogram and/or a phase spectrogram.
S402: inputting each diagnosis training image in the acquired training sample set into a preset training model so as to train the training model;
specifically, the training model is a functional module for learning and analyzing human eye images with eye diseases, and becomes a diagnosis model after the training of the training model is stopped.
S403: and stopping training the training model when the convergence value of the training model is determined to be smaller than a preset threshold value, and storing the training model which is stopped to be trained as the diagnosis model.
Specifically, the training model is trained through a large number of human eye images suffering from eye diseases, and is also iteratively calculated through a large number of data, the convergence value of the training model is a difference condition representing an output result after iterative calculation and an expected result, and the preset convergence threshold value is an expected difference condition reached by training of the training model. When the convergence value is smaller than the preset convergence threshold value, the output result of the training model data reaches the expected result, so that the training of the training model is stopped, and the training model with the training stopped is saved as the diagnosis model.
In the technical scheme provided by the embodiment, as the technical means of training the training model to obtain the diagnosis model is adopted, and the training model is trained through a large number of training sample sets, the generated diagnosis model has more comprehensive data, and the accuracy and comprehensiveness of the diagnosis result are ensured.
The invention also provides AR glasses, which comprise an image acquisition module, a memory, a processor and an eye disease detection program which is stored on the memory and can be run on the processor, wherein the eye disease detection program realizes the steps of the eye disease detection method in the embodiment when being executed by the processor.
The present invention also provides a readable storage medium, on which an eye disease detection program is stored, which when executed by a processor implements the steps of the eye disease detection method according to the above embodiment.
The above-mentioned serial numbers of the embodiments of the present invention are merely for description and do not represent the merits of the embodiments.
It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.
Through the above description of the embodiments, those skilled in the art will clearly understand that the method of the above embodiments can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware, but in many cases, the former is a better implementation manner. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a readable storage medium (such as ROM/RAM, magnetic disk, optical disk) as described above, and includes several instructions for enabling a terminal device (such as a mobile phone, a computer, a server, an air conditioner, or a network device) to execute the method according to the embodiments of the present invention.
The above description is only a preferred embodiment of the present invention, and not intended to limit the scope of the present invention, and all modifications of equivalent structures and equivalent processes, which are made by using the contents of the present specification and the accompanying drawings, or directly or indirectly applied to other related technical fields, are included in the scope of the present invention.
Claims (11)
1. An eye disease detection method, comprising:
acquiring a human eye image to be detected;
processing the human eye image to obtain a diagnostic image, wherein the diagnostic image comprises at least one of an amplitude spectrogram corresponding to the human eye image and a phase spectrogram corresponding to the human eye image;
and determining a diagnosis result of the eye according to the diagnosis image, and outputting the diagnosis result.
2. The method of detecting an ocular disease according to any of claim 1, wherein the step of processing the image of the human eye to obtain a diagnostic image comprises:
windowing the human eye image;
performing Fourier transform on the windowed human eye image to obtain at least one of the amplitude spectrogram and the phase spectrogram;
determining the diagnostic image from at least one of the magnitude and phase spectrograms.
3. The method of detecting an ocular disease of claim 2, wherein the step of determining the diagnostic image from at least one of the amplitude and phase spectrograms is preceded by:
adjusting a low-frequency area in the amplitude spectrogram to a first preset position corresponding to the low-frequency area, and adjusting a high-frequency area in the amplitude spectrogram to a second preset position corresponding to the high-frequency area;
the step of determining the diagnostic image from at least one of the magnitude and phase spectrograms comprises:
determining the diagnostic image according to the adjusted at least one of the magnitude spectrogram and the phase spectrogram.
4. The method for detecting an ocular disease according to claim 1, wherein the step of determining a diagnosis result of the eye from the diagnostic image comprises:
comparing the diagnostic image with a preset reference image to obtain a comparison result, wherein the reference image is determined according to a human eye image of a patient with eye diseases;
and determining the diagnosis result according to the comparison result.
5. The method of detecting an ocular disease of claim 4, wherein the step of determining the diagnosis based on the comparison result comprises:
determining the eye disease corresponding to the reference image as the diagnosis result when the comparison result is that the difference value between the parameter in the diagnosis image and the parameter in the reference image is smaller than or equal to a preset threshold value;
and determining the disease of the uninfected eye as the diagnosis result when the comparison result is that the difference value between the parameters in the diagnosis image and the parameters in the reference image is larger than a preset threshold value.
6. The method for detecting an eye disease of claim 1, wherein the step of obtaining the image of the human eye to be detected further comprises, before the step of obtaining the image of the human eye to be detected:
training a diagnosis model according to the human eye image of the eye disease patient;
the step of training a diagnostic model based on an image of a human eye of a patient with eye disease comprises:
inputting the diagnostic image to the diagnostic model;
and obtaining a diagnosis result output by the diagnosis model.
7. The method of detecting an ocular disease of claim 6, wherein the step of training a diagnostic model based on an image of a human eye of a patient with an ocular disease comprises:
acquiring a training sample set, wherein the training sample set comprises diagnostic training images corresponding to human eye images of a plurality of eye disease patients;
inputting each diagnosis training image in the training sample set into a preset training model so as to train the training model;
and stopping training the training model when the convergence value of the training model is determined to be smaller than a preset threshold value, and storing the training model which is stopped to be trained as the diagnosis model.
8. The method for detecting an eye disease according to any one of claims 1 to 7, wherein the step of acquiring an image of a human eye to be detected comprises:
acquiring the to-be-detected eye image collected by a camera carried by AR glasses;
the step of processing the human eye image to obtain a diagnostic image comprises:
controlling the AR glasses to process the human eye image to obtain a diagnostic image;
the step of determining a diagnosis result of the eye from the diagnosis image and outputting the diagnosis result includes:
and controlling the AR glasses to determine a diagnosis result of the eye according to the diagnosis image, and controlling the AR glasses to output the diagnosis result.
9. The method for detecting an ocular disease of claim 8, wherein the step of controlling the AR glasses to output the diagnosis result comprises:
controlling the AR glasses to display the diagnosis result on lenses thereof.
10. AR glasses, characterized in that the AR glasses comprise an image acquisition module, a memory, a processor and an eye disease detection program stored on the memory and executable on the processor, the eye disease detection program, when executed by the processor, implementing the steps of the eye disease detection method according to any one of claims 1 to 9.
11. A readable storage medium, having stored thereon an eye disease detection program which, when executed by a processor, implements the steps of the eye disease detection method according to any one of claims 1 to 9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010558496.XA CN111588345A (en) | 2020-06-18 | 2020-06-18 | Eye disease detection method, AR glasses and readable storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010558496.XA CN111588345A (en) | 2020-06-18 | 2020-06-18 | Eye disease detection method, AR glasses and readable storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111588345A true CN111588345A (en) | 2020-08-28 |
Family
ID=72186220
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010558496.XA Pending CN111588345A (en) | 2020-06-18 | 2020-06-18 | Eye disease detection method, AR glasses and readable storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111588345A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113554641A (en) * | 2021-07-30 | 2021-10-26 | 江苏盛泽医院 | Pediatric pharyngeal image acquisition method and device |
| CN115862300A (en) * | 2023-02-15 | 2023-03-28 | 暨南大学附属第一医院（广州华侨医院） | Early warning method for monitoring stroke risk and intelligent glasses for early warning stroke risk |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101720205A (en) * | 2007-04-10 | 2010-06-02 | 南加利福尼亚大学 | Methods and systems for blood flow measurement using doppler optical coherence tomography |
| CN102955889A (en) * | 2011-08-29 | 2013-03-06 | 中国科学院力学研究所 | Pulse wave reconstruction method for extracting time domain feature points |
| CN103393409A (en) * | 2013-08-22 | 2013-11-20 | 南京大学 | Comprehensive analysis method of multi-dimensional biology opto-acoustic image spectra |
| CN105595958A (en) * | 2014-11-14 | 2016-05-25 | 株式会社理光 | Simultaneous capture of filtered images of the eye |
| CN107911576A (en) * | 2017-11-01 | 2018-04-13 | 北京小米移动软件有限公司 | Image processing method, device and storage medium |
| US20190059719A1 (en) * | 2017-08-22 | 2019-02-28 | Boe Technology Group Co., Ltd. | Method and electronic device for positioning eyeball in retinopathy image |
| CN109685102A (en) * | 2018-11-13 | 2019-04-26 | 平安科技（深圳）有限公司 | Breast lesion image classification method, device, computer equipment and storage medium |
| CN109712119A (en) * | 2018-12-13 | 2019-05-03 | 深圳先进技术研究院 | A kind of magnetic resonance imaging and patch recognition methods and device |
| CN110348431A (en) * | 2019-08-26 | 2019-10-18 | 闫戎 | A kind of disease pre-warning method, device, equipment and system |
-
2020
- 2020-06-18 CN CN202010558496.XA patent/CN111588345A/en active Pending
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101720205A (en) * | 2007-04-10 | 2010-06-02 | 南加利福尼亚大学 | Methods and systems for blood flow measurement using doppler optical coherence tomography |
| CN102955889A (en) * | 2011-08-29 | 2013-03-06 | 中国科学院力学研究所 | Pulse wave reconstruction method for extracting time domain feature points |
| CN103393409A (en) * | 2013-08-22 | 2013-11-20 | 南京大学 | Comprehensive analysis method of multi-dimensional biology opto-acoustic image spectra |
| CN105595958A (en) * | 2014-11-14 | 2016-05-25 | 株式会社理光 | Simultaneous capture of filtered images of the eye |
| US20190059719A1 (en) * | 2017-08-22 | 2019-02-28 | Boe Technology Group Co., Ltd. | Method and electronic device for positioning eyeball in retinopathy image |
| CN107911576A (en) * | 2017-11-01 | 2018-04-13 | 北京小米移动软件有限公司 | Image processing method, device and storage medium |
| CN109685102A (en) * | 2018-11-13 | 2019-04-26 | 平安科技（深圳）有限公司 | Breast lesion image classification method, device, computer equipment and storage medium |
| CN109712119A (en) * | 2018-12-13 | 2019-05-03 | 深圳先进技术研究院 | A kind of magnetic resonance imaging and patch recognition methods and device |
| CN110348431A (en) * | 2019-08-26 | 2019-10-18 | 闫戎 | A kind of disease pre-warning method, device, equipment and system |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113554641A (en) * | 2021-07-30 | 2021-10-26 | 江苏盛泽医院 | Pediatric pharyngeal image acquisition method and device |
| CN113554641B (en) * | 2021-07-30 | 2022-04-12 | 江苏盛泽医院 | Pediatric pharyngeal image acquisition method and device |
| CN115862300A (en) * | 2023-02-15 | 2023-03-28 | 暨南大学附属第一医院（广州华侨医院） | Early warning method for monitoring stroke risk and intelligent glasses for early warning stroke risk |
| CN115862300B (en) * | 2023-02-15 | 2023-11-14 | 暨南大学附属第一医院（广州华侨医院） | Intelligent glasses for monitoring cerebral apoplexy risk early warning method and cerebral apoplexy risk early warning |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108427503B (en) | Human eye tracking method and human eye tracking device | |
| US7116820B2 (en) | Detecting and correcting red-eye in a digital image | |
| KR101848526B1 (en) | Systems and methods for rendering a display to compensate for a viewer's visual impairment | |
| US11389058B2 (en) | Method for pupil detection for cognitive monitoring, analysis, and biofeedback-based treatment and training | |
| JP7197708B2 (en) | Preprocessing method and storage device for fundus image quantitative analysis | |
| CN108985210A (en) | A kind of Eye-controlling focus method and system based on human eye geometrical characteristic | |
| WO2020019286A1 (en) | Blepharoptosis detection method and system | |
| CN112183200B (en) | Eye movement tracking method and system based on video image | |
| US11887299B2 (en) | Image processing system and image processing method | |
| WO2020232855A1 (en) | Method and apparatus for adjusting screen display on the basis of subtle expression | |
| CN111588345A (en) | Eye disease detection method, AR glasses and readable storage medium | |
| CN114937024A (en) | Image evaluation method and device and computer equipment | |
| CN111738241A (en) | Pupil detection method and device based on double cameras | |
| US20220061659A1 (en) | System and method for finding an area of an eye from a facial image | |
| CN116030042B (en) | Diagnostic device, method, equipment and storage medium for doctor's diagnosis | |
| Akhade et al. | Automatic optic disc detection in digital fundus images using image processing techniques | |
| Shaikha et al. | Optic Disc Detection and Segmentation in Retinal Fundus Image | |
| CN113116292B (en) | Eye position measurement method, device, terminal and equipment based on eye appearance image | |
| CN115512410A (en) | Abnormal refraction state identification method and device based on abnormal eye posture | |
| Redi et al. | How to apply spatial saliency into objective metrics for JPEG compressed images? | |
| US10617294B1 (en) | System and method for determining the spherical power of eyes based on measured refractive error | |
| Kabbani et al. | Robust Sclera Segmentation for Skin-Tone Agnostic Face Image Quality Assessment | |
| Rafiq et al. | Reconstruction of scene using corneal reflection | |
| CN116228545B (en) | Fundus color photographic image stitching method and system based on retina characteristic points | |
| CN118196218B (en) | Fundus image processing method, device and equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200828 |
|
| RJ01 | Rejection of invention patent application after publication |