CN118097764A - Left-right eye identification system and method applied to slit lamp, electronic equipment and storage medium - Google Patents
Left-right eye identification system and method applied to slit lamp, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN118097764A CN118097764A CN202410109920.0A CN202410109920A CN118097764A CN 118097764 A CN118097764 A CN 118097764A CN 202410109920 A CN202410109920 A CN 202410109920A CN 118097764 A CN118097764 A CN 118097764A
- Authority
- CN
- China
- Prior art keywords
- detected
- image
- eye
- human eye
- right eye
- 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
- 238000000034 method Methods 0.000 title claims abstract description 34
- 238000007781 pre-processing Methods 0.000 claims abstract description 47
- 238000000605 extraction Methods 0.000 claims abstract description 38
- 210000001747 pupil Anatomy 0.000 claims description 31
- 210000000720 eyelash Anatomy 0.000 claims description 29
- 238000010606 normalization Methods 0.000 claims description 26
- 238000001914 filtration Methods 0.000 claims description 14
- 238000012549 training Methods 0.000 claims description 12
- 230000006870 function Effects 0.000 claims description 10
- 238000012545 processing Methods 0.000 claims description 9
- 238000001514 detection method Methods 0.000 claims description 8
- 238000005070 sampling Methods 0.000 claims description 8
- 238000004590 computer program Methods 0.000 claims description 7
- 230000009466 transformation Effects 0.000 claims description 6
- 238000007689 inspection Methods 0.000 abstract description 11
- 238000010586 diagram Methods 0.000 description 8
- 230000005540 biological transmission Effects 0.000 description 2
- 238000003745 diagnosis Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 230000001575 pathological effect Effects 0.000 description 2
- 210000002159 anterior chamber Anatomy 0.000 description 1
- 230000000712 assembly Effects 0.000 description 1
- 238000000429 assembly Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 210000000795 conjunctiva Anatomy 0.000 description 1
- 210000004087 cornea Anatomy 0.000 description 1
- 210000000695 crystalline len Anatomy 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 210000000554 iris Anatomy 0.000 description 1
- 238000011022 operating instruction Methods 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 210000003786 sclera Anatomy 0.000 description 1
- 210000004127 vitreous body Anatomy 0.000 description 1
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02B—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO BUILDINGS, e.g. HOUSING, HOUSE APPLIANCES OR RELATED END-USER APPLICATIONS
- Y02B20/00—Energy efficient lighting technologies, e.g. halogen lamps or gas discharge lamps
- Y02B20/40—Control techniques providing energy savings, e.g. smart controller or presence detection
Landscapes
- Image Analysis (AREA)
Abstract
The invention discloses a left and right eye identification system, a method, electronic equipment and a storage medium applied to a slit lamp, wherein the left and right eye identification system comprises the following components: the device comprises a human eye image region to be detected identification module, an image preprocessing module, a characteristic extraction module and a characteristic identification module; the human eye image region to be detected identification module is used for receiving and dividing a human eye region to be detected; the image preprocessing module is used for preprocessing the image of the human eye image to be detected, which is received by the human eye image region to be detected identification module; the feature extraction module is used for extracting features of the human eye image to be detected after being preprocessed by the image preprocessing module; the feature recognition module is used for recognizing the human eye features extracted by the feature extraction module so as to judge and output the human eye type to be detected. The invention can improve the efficiency of clinical slit lamp inspection and avoid human error.
Description
Technical Field
The invention belongs to the technical field of human eye identification, relates to an identification system, and in particular relates to a left and right eye identification system and method applied to a slit lamp, electronic equipment and a storage medium.
Background
The slit lamp is a slit lamp microscope, and is an optical instrument most frequently used in ophthalmic clinical examination. A doctor can clearly observe the pathological conditions of the eyes by means of a slit lamp microscope, can determine the positions, the sizes and the depths of the pathological conditions of the eyes, can comprehensively check cornea, conjunctiva, anterior chamber, sclera, iris, pupil, crystalline lens, vitreous body and the like, and is an important checking device for ophthalmologists.
Conventional slit lamp inspection, doctors can only observe the eye condition of a patient through the eyepiece of a slit lamp microscope, cannot comprehensively and systematically record inspection results, cannot deeply analyze the inspection results, and therefore, the inspection results are gradually replaced by a slit lamp analysis system based on digitalization. The slit lamp analysis system based on digitalization provides richer and comprehensive functions for clinical slit lamp inspection by integrating slit lamp hardware and software.
However, most digital slit lamp analysis systems cannot automatically identify the eye types of the detected person due to the limitation of the current technology, and often a doctor needs to manually fill in whether the eye types of the current detection are left eye or right eye after shooting the eye images of the detected person, so that time and labor are wasted on one hand, and the efficiency of clinical slit lamp detection is reduced; on the other hand, medical accidents caused by manual filling errors are easy to occur, and the medical service level is reduced.
In view of the foregoing, there is a great need to design a new slit lamp analysis system that overcomes at least some of the above-described deficiencies of existing slit lamp analysis systems.
Disclosure of Invention
The invention provides a left-right eye identification system, a method, electronic equipment and a storage medium applied to a slit lamp, which can improve the efficiency of clinical slit lamp inspection and avoid human errors.
In order to solve the technical problems, according to one aspect of the present invention, the following technical scheme is adopted:
a left and right eye identification system for a slit lamp, the left and right eye identification system comprising:
the human eye image region to be detected identification module is used for receiving and dividing the human eye region to be detected;
The image preprocessing module is used for preprocessing the human eye image to be detected, which is received by the human eye image region to be detected identification module, and sending the preprocessed image to the feature extraction module;
the feature extraction module is used for extracting features of the human eye image to be detected after being preprocessed by the image preprocessing module; and
The feature recognition module is used for recognizing the human eye features extracted by the feature extraction module so as to judge and output the human eye type to be detected, wherein the human eye type is left human eye or right human eye.
As one implementation mode of the invention, the human eye image region to be detected identification module is used for receiving human eye images acquired by the slit lamp system, processing the images and dividing left eye and right eye regions to be detected from the images to be detected;
The image preprocessing module is used for performing operations such as image filtering, image normalization and the like on the image of the human eye area to be detected so as to meet the input requirements of the feature extraction module;
The image preprocessing module performs image preprocessing operation on a region to be detected of human eyes, wherein the image preprocessing operation comprises image filtering and image normalization operation;
The method comprises the steps of eliminating the influence of noise on a human eye image area to be detected during image acquisition through image filtering; eliminating the influence of unit and scale differences among the features of the human eye region to be detected through image normalization so as to treat each dimension feature equally;
the normalization method is to perform linear transformation on original data, map human eye data to be detected between [0,1], and adopt normalization functions as follows:
Wherein max (x) is the maximum value of the sample data of the human eye region to be detected, and min (x) is the minimum value of the sample data of the human eye region to be detected.
As an embodiment of the present invention, the feature extraction module is configured to extract feature points of an eye region to be detected after image preprocessing, including extracting feature points of regions such as pupils, corners of eyes, and eyelashes;
Marking pupils, corners of eyes, eyelashes and the like in sampling areas of the left eye and the right eye respectively, and determining position coordinates; extracting the characteristics of the left eye and the right eye to-be-detected areas after image pretreatment, and adopting Harr characteristics as the characteristics of the left eye and the right eye;
Classifying the image subareas by calculating the difference value of adjacent rectangular pixels in the detection window, wherein Harr features are of three types of edge features, linear features, central features and diagonal features and can be combined into a feature template; the characteristic template is internally provided with two rectangles of white and black, the characteristic value of the template is defined as white rectangular characteristic pixels and subtracting black rectangular characteristic pixels, and the characteristics suitable for expressing the corners, pupils and eyelashes of the left eye and the right eye are selected on the Harr characteristic template to distinguish the categories of the left eye and the right eye.
As an implementation mode of the invention, the feature recognition module is used for carrying out human eye type feature recognition and output on the extracted human eye feature points of the region to be detected, and outputting the recognition result to the slit lamp image processing system so as to automatically display the eye type of the detected person;
The characteristic extraction module acquires Harr characteristics of corners, pupils and eyelashes of the left eye and the right eye, identifies the Harr characteristics of the left eye and the right eye through an AdaBoost algorithm, combines a plurality of weak classifiers into a new strong classifier through cascading the AdaBoost algorithm, and outputs a left eye and right eye discrimination result;
Training individual weak classifiers by taking the characteristics of the canthus, pupil, eyelashes and the like of the left eye and the right eye as samples respectively; the training samples comprise positive and negative samples, the positive samples are left and right eye characteristics to be detected, the negative samples refer to other samples except the positive samples, all the samples to be detected are normalized to the same image size in an image preprocessing stage, the image size is not fixed, and the size can be determined according to specific requirements;
The AdaBoost algorithm is adopted to distribute corresponding weight values for different features at the initial stage based on Harr features of the left eye and the right eye, weak classifier weights of different samples are calculated and updated after multiple iterations, finally when the classification error of the combined strong classifier is 0, the iteration is ended, and the trained model is used for detecting the left eye region and the right eye region to be detected.
According to another aspect of the invention, the following technical scheme is adopted: a left-right eye identification method applied to a slit lamp, the left-right eye identification method comprising:
Step S1, identifying a region to be detected of a human eye image: receiving and dividing a human eye area to be detected;
S2, an image preprocessing step: performing image preprocessing on the human eye image to be detected received in the human eye image region to be detected identification step;
Step S3, a feature extraction step: extracting features of the human eye image to be detected after the pretreatment of the image pretreatment step;
Step S4, a feature recognition step: and identifying the human eye characteristics extracted in the characteristic extraction step to judge and output the human eye type to be detected, wherein the human eye type is left human eye or right human eye.
In step S1, a human eye image collected by a slit lamp system is received, the image is processed, and a left eye area and a right eye area to be detected are divided from the image to be detected;
In the step S2, operations such as image filtering, image normalization, etc. are performed on the image of the human eye region to be detected;
The method comprises the steps of eliminating the influence of noise on a human eye image area to be detected during image acquisition through image filtering; eliminating the influence of unit and scale differences among the features of the human eye region to be detected through image normalization so as to treat each dimension feature equally;
the normalization method is to perform linear transformation on original data, map human eye data to be detected between [0,1], and adopt normalization functions as follows:
Wherein max (x) is the maximum value of the sample data of the human eye region to be detected, and min (x) is the minimum value of the sample data of the human eye region to be detected.
In step S3, feature points of the eye region to be detected after the image preprocessing are extracted, including extraction of feature points of the pupil, the canthus, the eyelashes and other regions;
Marking pupils, corners of eyes, eyelashes and the like in sampling areas of the left eye and the right eye respectively, and determining position coordinates; extracting the characteristics of the left eye and the right eye to-be-detected areas after image pretreatment, and adopting Harr characteristics as the characteristics of the left eye and the right eye;
Classifying the image subareas by calculating the difference value of adjacent rectangular pixels in the detection window, wherein Harr features are of three types of edge features, linear features, central features and diagonal features and can be combined into a feature template; the characteristic template is internally provided with two rectangles of white and black, the characteristic value of the template is defined as white rectangular characteristic pixels and subtracting black rectangular characteristic pixels, and the characteristics suitable for expressing the corners, pupils and eyelashes of the left eye and the right eye are selected on the Harr characteristic template to distinguish the categories of the left eye and the right eye.
In the step S4, the human eye type feature of the extracted region to be detected is identified and output, and the identification result is output to the slit lamp image processing system to automatically display the eye type of the detected person;
Identifying Harr characteristics of left and right eyes through an AdaBoost algorithm according to the Harr characteristics of the corners, pupils and eyelashes of the left and right eyes, which are acquired in the characteristic extraction step, wherein the AdaBoost algorithm combines a plurality of weak classifiers into a new strong classifier through cascading, and outputs a left and right eye discrimination result;
Training individual weak classifiers by taking the characteristics of the canthus, pupil, eyelashes and the like of the left eye and the right eye as samples respectively; the training samples comprise positive and negative samples, the positive samples are left and right eye characteristics to be detected, the negative samples refer to other samples except the positive samples, all the samples to be detected are normalized to the same image size in an image preprocessing stage, the image size is not fixed, and the size can be determined according to specific requirements;
The AdaBoost algorithm is adopted to distribute corresponding weight values for different features at the initial stage based on Harr features of the left eye and the right eye, weak classifier weights of different samples are calculated and updated after multiple iterations, finally when the classification error of the combined strong classifier is 0, the iteration is ended, and the trained model is used for detecting the left eye region and the right eye region to be detected.
According to a further aspect of the invention, the following technical scheme is adopted: an electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the steps of the above method when the computer program is executed.
According to a further aspect of the invention, the following technical scheme is adopted: a storage medium having stored thereon computer program instructions which, when executed by a processor, perform the steps of the above method.
The invention has the beneficial effects that: the left and right eye identification system, the method, the electronic equipment and the storage medium applied to the slit lamp can improve the efficiency of clinical slit lamp inspection and avoid human errors. Meanwhile, the invention provides an auxiliary digital slit lamp analysis system for accurate analysis and diagnosis based on left and right eyes.
The invention combines the left and right eye identification device and method with the digital slit lamp analysis system, can solve the problem that the existing slit lamp system cannot automatically identify the type of eyes of a detected person to a great extent, and assists the digital slit lamp analysis system to accurately analyze and diagnose based on the left and right eyes, thereby improving the efficiency and the accuracy of clinical slit lamp inspection.
Drawings
Fig. 1 is a schematic diagram of a left-right eye identification system applied to a slit lamp according to an embodiment of the invention.
Fig. 2 is a flowchart of a method for identifying left and right eyes applied to a slit lamp according to an embodiment of the invention.
Fig. 3 is a schematic diagram of sampling left and right eye feature points according to an embodiment of the present invention.
Fig. 4 is a schematic structural diagram of a left-right eye identification device applied to a slit lamp according to an embodiment of the invention.
Fig. 5 is a flowchart illustrating a left-right eye identification method applied to a slit lamp according to an embodiment of the invention.
Fig. 6 is a schematic diagram of an electronic device according to an embodiment of the invention.
Detailed Description
Preferred embodiments of the present invention will be described in detail below with reference to the accompanying drawings.
For a further understanding of the present invention, preferred embodiments of the invention are described below in conjunction with the examples, but it should be understood that these descriptions are merely intended to illustrate further features and advantages of the invention, and are not limiting of the claims of the invention.
The description of this section is intended to be illustrative of only a few exemplary embodiments and the invention is not to be limited in scope by the description of the embodiments. It is also within the scope of the description and claims of the invention to interchange some of the technical features of the embodiments with other technical features of the same or similar prior art.
The description of the steps in the various embodiments in the specification is merely for convenience of description, and the implementation of the present application is not limited by the order in which the steps are implemented.
"Connected" in the specification includes both direct and indirect connections.
The invention discloses a left and right eye identification system applied to a slit lamp, and FIG. 1 is a schematic diagram of the components of the left and right eye identification system applied to the slit lamp in an embodiment of the invention; referring to fig. 1, the left-right eye recognition system includes: the device comprises a human eye image region to be detected identification module 1, an image preprocessing module 2, a characteristic extraction module 3 and a characteristic identification module 4.
The human eye image region to be detected identification module 1 is used for receiving and dividing the human eye region to be detected.
The image preprocessing module 2 is configured to perform image preprocessing on the human eye image to be detected received by the human eye image region to be detected identification module, and send the preprocessed image to the feature extraction module 3.
The feature extraction module 3 is used for extracting features of the human eye image to be detected after being preprocessed by the image preprocessing module 2;
the feature recognition module 4 is configured to recognize the human eye feature extracted by the feature extraction module 3, so as to determine and output a human eye type to be detected, where the human eye type is a left human eye or a right human eye.
In an embodiment of the present invention, the human eye image to-be-detected area identifying module 1 is configured to receive a human eye image collected by the slit lamp system, process the image, and divide a left eye to-be-detected area and a right eye to-be-detected area from the to-be-detected image.
The human eye image region to be detected identification module 1 divides left eye and right eye regions to be detected from the received image to be detected. Wherein the human eye image acquired by the acquisition device can be received by common video acquisition devices such as monocular cameras, binocular cameras and the like. The signals of the acquisition equipment can support the following transmission forms, namely, the signals can support human eye images (including, but not limited to HTTP, HTTPS, RTSP, RTMP and the like) transmitted back through a network camera in remote consultation; the camera can be connected with a computer host through a USB interface, and the human eye picture shot by the camera can be directly accessed and acquired; the camera can also be a video interface such as HDMI or SDI, and the human eye picture shot by the camera can be obtained by connecting with the image acquisition card of the host computer.
The image preprocessing module 2 is used for performing image preprocessing operations on the human eye region image to be detected, including operations such as image filtering and image normalization, so as to meet the input requirements of the feature extraction module.
The method comprises the steps of eliminating the influence of noise on a human eye image area to be detected during image acquisition through image filtering; and eliminating the influence of unit and scale differences among the features of the human eye region to be detected through image normalization so as to treat each dimension feature equally.
The normalization method is to perform linear transformation on original data, map human eye data to be detected between [0,1], and adopt normalization functions as follows:
Wherein max (x) is the maximum value of the sample data of the human eye region to be detected, and min (x) is the minimum value of the sample data of the human eye region to be detected.
FIG. 3 is a schematic diagram illustrating sampling of left and right eye feature points according to an embodiment of the present invention; referring to fig. 3, in an embodiment of the present invention, the feature extraction module 3 is configured to extract feature points of a region of the human eye to be detected after the image preprocessing, including extracting feature points of regions such as pupils, corners of eyes, and lashes. Marking pupils, corners of eyes, eyelashes and the like in sampling areas of the left eye and the right eye respectively, and determining position coordinates; and extracting the characteristics of the left eye and the right eye to-be-detected areas after image pretreatment, and adopting Harr characteristics as the characteristics of the left eye and the right eye.
Classifying the image subareas by calculating the difference value of adjacent rectangular pixels in the detection window, wherein Harr features are of three types of edge features, linear features, central features and diagonal features and can be combined into a feature template; the characteristic template is internally provided with two rectangles of white and black, the characteristic value of the template is defined as white rectangular characteristic pixels and subtracting black rectangular characteristic pixels, and the characteristics suitable for expressing the corners, pupils and eyelashes of the left eye and the right eye are selected on the Harr characteristic template to distinguish the categories of the left eye and the right eye.
In an embodiment of the present invention, the feature recognition module 4 is configured to perform human eye type feature recognition and output on the extracted human eye feature points of the region to be detected, and output the recognition result to the slit lamp image processing system to automatically display the eye type of the detected person.
The characteristic extraction module 3 acquires Harr characteristics of corners, pupils and eyelashes of the left eye and the right eye, identifies the Harr characteristics of the left eye and the right eye through an AdaBoost algorithm, and outputs a left eye and right eye discrimination result; the AdaBoost algorithm combines a new strong classifier by cascading multiple weak classifiers.
Training individual weak classifiers by taking the characteristics of the canthus, pupil, eyelashes and the like of the left eye and the right eye as samples respectively; the training samples comprise positive and negative samples, the positive samples are left and right eye characteristics to be detected, the negative samples refer to other samples except the positive samples, all the samples to be detected are normalized to the same image size in an image preprocessing stage, the image size is not fixed, and the size can be determined according to specific requirements;
The AdaBoost algorithm is adopted to distribute corresponding weight values for different features at the initial stage based on Harr features of the left eye and the right eye, weak classifier weights of different samples are calculated and updated after multiple iterations, finally when the classification error of the combined strong classifier is 0, the iteration is ended, and the trained model is used for detecting the left eye region and the right eye region to be detected. In one embodiment, the classification result is output as two values based on the usage scenario of the present invention, where the classification result is represented by the letter M for example, and the value of M is 0 or 1; when m=0, the region to be detected is determined to be left eye, and when m=1, the region to be detected is determined to be right eye.
The invention further discloses a left and right eye identification method applied to the slit lamp, which comprises the following steps:
step S1, identifying a region to be detected of a human eye image: and receiving and dividing the human eye area to be detected.
In one embodiment of the invention, human eye images acquired by a slit lamp system are received, the images are processed, and left eye and right eye areas to be detected are divided from the images to be detected.
And dividing left-eye and right-eye areas to be detected from the received image to be detected by the human eye image area identification module to be detected. Wherein an image of a human eye acquired by a common video acquisition device including, but not limited to, a monocular camera, a binocular camera, etc., may be received.
The signals of the acquisition equipment can support the following transmission forms, namely, the signals can support human eye images (including, but not limited to HTTP, HTTPS, RTSP, RTMP and the like) transmitted back through a network camera in remote consultation; the camera can be connected with a computer host through a USB interface, and the human eye picture shot by the camera can be directly accessed and acquired; the camera can also be a video interface such as HDMI or SDI, and the human eye picture shot by the camera can be obtained by connecting with the image acquisition card of the host computer.
An image preprocessing step: and carrying out image preprocessing on the human eye image to be detected received in the human eye image region to be detected identification step.
In an embodiment of the invention, the operations of image filtering, image normalization and the like are performed on the image of the human eye region to be detected;
The method comprises the steps of eliminating the influence of noise on a human eye image area to be detected during image acquisition through image filtering; eliminating the influence of unit and scale differences among the features of the human eye region to be detected through image normalization so as to treat each dimension feature equally;
the normalization method is to perform linear transformation on original data, map human eye data to be detected between [0,1], and adopt normalization functions as follows:
Wherein max (x) is the maximum value of the sample data of the human eye region to be detected, and min (x) is the minimum value of the sample data of the human eye region to be detected.
Feature extraction step (step S3): and carrying out feature extraction on the human eye image to be detected which is preprocessed in the image preprocessing step.
In an embodiment of the present invention, feature points of the eye area to be detected after image preprocessing are extracted, including extraction of feature points of areas such as pupils, corners of eyes, eyelashes, etc.;
Marking pupils, corners of eyes, eyelashes and the like in sampling areas of the left eye and the right eye respectively, and determining position coordinates; extracting the characteristics of the left eye and the right eye to-be-detected areas after image pretreatment, and adopting Harr characteristics as the characteristics of the left eye and the right eye;
Classifying the image subareas by calculating the difference value of adjacent rectangular pixels in the detection window, wherein Harr features are of three types of edge features, linear features, central features and diagonal features and can be combined into a feature template; the characteristic template is internally provided with two rectangles of white and black, the characteristic value of the template is defined as white rectangular characteristic pixels and subtracting black rectangular characteristic pixels, and the characteristics suitable for expressing the corners, pupils and eyelashes of the left eye and the right eye are selected on the Harr characteristic template to distinguish the categories of the left eye and the right eye.
Feature recognition step (step S4): and identifying the human eye characteristics extracted in the characteristic extraction step to judge and output the human eye type to be detected, wherein the human eye type is left human eye or right human eye.
In an embodiment of the invention, human eye type feature recognition and output are performed on the extracted human eye feature points of the region to be detected, and recognition results are output to a slit lamp image processing system so as to automatically display the eye types of the detected person.
Acquiring Harr characteristics of corners, pupils and eyelashes of the left eye and the right eye according to the characteristic extraction step, identifying the Harr characteristics of the left eye and the right eye by an AdaBoost algorithm, and combining a plurality of weak classifiers into a new strong classifier by the AdaBoost algorithm in a cascading way, and outputting a discrimination result of the left eye and the right eye;
Training individual weak classifiers by taking the characteristics of the canthus, pupil, eyelashes and the like of the left eye and the right eye as samples respectively; the training samples comprise positive and negative samples, the positive samples are left and right eye characteristics to be detected, the negative samples refer to other samples except the positive samples, all the samples to be detected are normalized to the same image size in an image preprocessing stage, the image size is not fixed, and the size can be determined according to specific requirements;
The AdaBoost algorithm is adopted to distribute corresponding weight values for different features at the initial stage based on Harr features of the left eye and the right eye, weak classifier weights of different samples are calculated and updated after multiple iterations, finally when the classification error of the combined strong classifier is 0, the iteration is ended, and the trained model is used for detecting the left eye region and the right eye region to be detected. In one embodiment, the classification result is output as two values based on the usage scenario of the present invention, where the classification result is represented by the letter M for example, and the value of M is 0 or 1; when m=0, the region to be detected is determined to be left eye, and when m=1, the region to be detected is determined to be right eye.
FIG. 4 is a schematic diagram of a left and right eye identification device applied to a slit lamp according to an embodiment of the invention; referring to fig. 4, in an embodiment of the present invention, the left and right eye identification system of the present invention can be applied to left and right eye identification of a slit lamp.
FIG. 5 is a flowchart of a left-right eye identification method applied to a slit lamp according to an embodiment of the invention; referring to fig. 5, in an embodiment of the invention, the left-right eye identification method of the invention can be applied to the left-right eye identification method of the slit lamp.
The invention also discloses an electronic device, and FIG. 6 is a schematic diagram of the composition of the electronic device in an embodiment of the invention; referring to fig. 6, the electronic device includes a memory, a processor, and at least one network interface at a hardware level; the processor may be a microprocessor, and the memory may include a memory, for example, a random access memory (Random Access Memory, RAM), a non-volatile memory (non-volatile memory), and so on. Of course, the electronic device may also be provided with other hardware as desired.
The processor, network interface, and memory may be interconnected by an internal bus, which may be an ISA (industry standard architecture) bus, a PCI (peripheral component interconnect standard) bus, or an EISA (extended industry standard architecture) bus, etc.; the buses may include address buses, data buses, control buses, and the like. The memory is used for storing programs (which can comprise operating system programs and application programs); the program may include program code that may include computer operating instructions. The memory may include memory and non-volatile storage and provide instructions and data to the processor.
In one embodiment, the processor may read the corresponding program from the nonvolatile memory to the memory and then run the program; the processor is capable of executing the program stored in the memory and is specifically configured to perform the following operations (as shown in fig. 2):
Step S1, identifying a region to be detected of a human eye image: receiving and dividing a human eye area to be detected;
an image preprocessing step: performing image preprocessing on the human eye image to be detected received in the human eye image region to be detected identification step;
feature extraction step (step S3): extracting features of the human eye image to be detected after the pretreatment of the image pretreatment step; and
Feature recognition step (step S4): and identifying the human eye characteristics extracted in the characteristic extraction step to judge and output the human eye type to be detected, wherein the human eye type is left human eye or right human eye.
The invention further discloses a storage medium having stored thereon computer program instructions which when executed by a processor perform the following steps of the method of the invention (as shown in fig. 2):
Step S1, identifying a region to be detected of a human eye image: receiving and dividing a human eye area to be detected;
an image preprocessing step: performing image preprocessing on the human eye image to be detected received in the human eye image region to be detected identification step;
feature extraction step (step S3): extracting features of the human eye image to be detected after the pretreatment of the image pretreatment step; and
Feature recognition step (step S4): and identifying the human eye characteristics extracted in the characteristic extraction step to judge and output the human eye type to be detected, wherein the human eye type is left human eye or right human eye.
In summary, the left and right eye identification system, the method, the electronic device and the storage medium applied to the slit lamp provided by the invention can improve the efficiency of clinical slit lamp inspection and avoid human errors. Meanwhile, the invention provides an auxiliary digital slit lamp analysis system for accurate analysis and diagnosis based on left and right eyes.
The invention combines the left and right eye identification device and method with the digital slit lamp analysis system, can solve the problem that the existing slit lamp system cannot automatically identify the type of eyes of a detected person to a great extent, and assists the digital slit lamp analysis system to accurately analyze and diagnose based on the left and right eyes, thereby improving the efficiency and the accuracy of clinical slit lamp inspection.
It should be noted that the present application may be implemented in software and/or a combination of software and hardware; for example, an Application Specific Integrated Circuit (ASIC), a general purpose computer, or any other similar hardware device may be employed. In some embodiments, the software program of the present application may be executed by a processor to implement the above steps or functions. Likewise, the software program of the present application (including the related data structures) may be stored in a computer-readable recording medium; such as RAM memory, magnetic or optical drives or diskettes, and the like. In addition, some steps or functions of the present application may be implemented in hardware; for example, as circuitry that cooperates with the processor to perform various steps or functions.
The technical features of the above-described embodiments may be arbitrarily combined, and all possible combinations of the technical features in the above-described embodiments are not described for brevity of description, however, as long as there is no contradiction between the combinations of the technical features, they should be considered as the scope of the description.
The description and applications of the present invention herein are illustrative and are not intended to limit the scope of the invention to the embodiments described above. Effects or advantages referred to in the embodiments may not be embodied in the embodiments due to interference of various factors, and description of the effects or advantages is not intended to limit the embodiments. Variations and modifications of the embodiments disclosed herein are possible, and alternatives and equivalents of the various components of the embodiments are known to those of ordinary skill in the art. It will be clear to those skilled in the art that the present invention may be embodied in other forms, structures, arrangements, proportions, and with other assemblies, materials, and components, without departing from the spirit or essential characteristics thereof. Other variations and modifications of the embodiments disclosed herein may be made without departing from the scope and spirit of the invention.
Claims (10)
1. A left and right eye identification system for a slit lamp, the left and right eye identification system comprising:
the human eye image region to be detected identification module is used for receiving and dividing the human eye region to be detected;
The image preprocessing module is used for preprocessing the human eye image to be detected, which is received by the human eye image region to be detected identification module, and sending the preprocessed image to the feature extraction module;
the feature extraction module is used for extracting features of the human eye image to be detected after being preprocessed by the image preprocessing module; and
The feature recognition module is used for recognizing the human eye features extracted by the feature extraction module so as to judge and output the human eye type to be detected, wherein the human eye type is left human eye or right human eye.
2. The left and right eye identification system for a slit lamp according to claim 1, wherein:
the human eye image to-be-detected area identification module is used for receiving human eye images acquired by the slit lamp system, processing the images and dividing left eye and right eye to-be-detected areas from the to-be-detected images;
The image preprocessing module is used for carrying out image filtering and image normalization operation on the image of the human eye area to be detected so as to meet the input requirement of the feature extraction module;
The image preprocessing module performs image preprocessing operation on a region to be detected of human eyes, wherein the image preprocessing operation comprises image filtering and image normalization operation;
The method comprises the steps of eliminating the influence of noise on a human eye image area to be detected during image acquisition through image filtering; eliminating the influence of unit and scale differences among the features of the human eye region to be detected through image normalization so as to treat each dimension feature equally;
the normalization method is to perform linear transformation on original data, map human eye data to be detected between [0,1], and adopt normalization functions as follows:
Wherein max (x) is the maximum value of the sample data of the human eye region to be detected, and min (x) is the minimum value of the sample data of the human eye region to be detected.
3. The left and right eye identification system for a slit lamp according to claim 1, wherein:
The feature extraction module is used for extracting feature points of the human eye region to be detected after the image preprocessing, and comprises the steps of extracting feature points of pupils, corners of eyes and eyelashes;
Marking pupils, corners of eyes, eyelashes and the like in sampling areas of the left eye and the right eye respectively, and determining position coordinates; extracting the characteristics of the left eye and the right eye to-be-detected areas after image pretreatment, and adopting Harr characteristics as the characteristics of the left eye and the right eye;
Classifying the image subareas by calculating the difference value of adjacent rectangular pixels in the detection window, wherein Harr features are of three types of edge features, linear features, central features and diagonal features and can be combined into a feature template; the characteristic template is internally provided with two rectangles of white and black, the characteristic value of the template is defined as white rectangular characteristic pixels and subtracting black rectangular characteristic pixels, and the characteristics suitable for expressing the corners, pupils and eyelashes of the left eye and the right eye are selected on the Harr characteristic template to distinguish the categories of the left eye and the right eye.
4. The left and right eye identification system for a slit lamp according to claim 1, wherein:
The feature recognition module is used for carrying out human eye type feature recognition and output on the extracted human eye feature points of the region to be detected, and outputting a recognition result to the slit lamp image processing system so as to automatically display the eye type of the detected person;
The characteristic extraction module acquires Harr characteristics of corners, pupils and eyelashes of the left eye and the right eye, identifies the Harr characteristics of the left eye and the right eye through an AdaBoost algorithm, combines a plurality of weak classifiers into a new strong classifier through cascading the AdaBoost algorithm, and outputs a left eye and right eye discrimination result;
Training individual weak classifiers by taking canthus, pupil and eyelash characteristics of left and right eyes as samples respectively; the training samples comprise positive and negative samples, the positive samples are left and right eye characteristics to be detected, the negative samples refer to other samples except the positive samples, all the samples to be detected are normalized to the same image size in an image preprocessing stage, the image size is not fixed, and the size can be determined according to specific requirements;
The AdaBoost algorithm is adopted to distribute corresponding weight values for different features at the initial stage based on Harr features of the left eye and the right eye, weak classifier weights of different samples are calculated and updated after multiple iterations, finally when the classification error of the combined strong classifier is 0, the iteration is ended, and the trained model is used for detecting the left eye region and the right eye region to be detected.
5. A left-right eye identification method applied to a slit lamp, characterized in that the left-right eye identification method comprises the following steps:
Step S1, identifying a region to be detected of a human eye image: receiving and dividing a human eye area to be detected;
S2, an image preprocessing step: performing image preprocessing on the human eye image to be detected received in the human eye image region to be detected identification step;
Step S3, a feature extraction step: extracting features of the human eye image to be detected after the pretreatment of the image pretreatment step;
Step S4, a feature recognition step: and identifying the human eye characteristics extracted in the characteristic extraction step to judge and output the human eye type to be detected, wherein the human eye type is left human eye or right human eye.
6. The left-right eye identification method applied to a slit lamp according to claim 5, wherein:
In the step S1, receiving a human eye image acquired by a slit lamp system, processing the image, and dividing left eye and right eye areas to be detected from the image to be detected;
In the step S2, image filtering and image normalization are performed on the image of the human eye region to be detected;
The method comprises the steps of eliminating the influence of noise on a human eye image area to be detected during image acquisition through image filtering; eliminating the influence of unit and scale differences among the features of the human eye region to be detected through image normalization so as to treat each dimension feature equally;
the normalization method is to perform linear transformation on original data, map human eye data to be detected between [0,1], and adopt normalization functions as follows:
Wherein max (x) is the maximum value of the sample data of the human eye region to be detected, and min (x) is the minimum value of the sample data of the human eye region to be detected.
7. The left-right eye identification method applied to a slit lamp according to claim 5, wherein:
in the step S3, feature points of the eye region to be detected after the image preprocessing are extracted, including extracting feature points of pupil, canthus and eyelash regions;
Marking pupils, corners of eyes, eyelashes and the like in sampling areas of the left eye and the right eye respectively, and determining position coordinates; extracting the characteristics of the left eye and the right eye to-be-detected areas after image pretreatment, and adopting Harr characteristics as the characteristics of the left eye and the right eye;
Classifying the image subareas by calculating the difference value of adjacent rectangular pixels in the detection window, wherein Harr features are of three types of edge features, linear features, central features and diagonal features and can be combined into a feature template; the characteristic template is internally provided with two rectangles of white and black, the characteristic value of the template is defined as white rectangular characteristic pixels and subtracting black rectangular characteristic pixels, and the characteristics suitable for expressing the corners, pupils and eyelashes of the left eye and the right eye are selected on the Harr characteristic template to distinguish the categories of the left eye and the right eye.
8. The left-right eye identification method applied to a slit lamp according to claim 5, wherein:
in the step S4, human eye type feature recognition and output are performed on the extracted human eye feature points of the region to be detected, and the recognition result is output to the slit lamp image processing system so as to automatically display the eye type of the detected person;
Identifying Harr characteristics of left and right eyes through an AdaBoost algorithm according to the Harr characteristics of the corners, pupils and eyelashes of the left and right eyes, which are acquired in the characteristic extraction step, wherein the AdaBoost algorithm combines a plurality of weak classifiers into a new strong classifier through cascading, and outputs a left and right eye discrimination result;
Training individual weak classifiers by taking canthus, pupil and eyelash characteristics of left and right eyes as samples respectively; the training samples comprise positive and negative samples, the positive samples are left and right eye characteristics to be detected, the negative samples refer to other samples except the positive samples, all the samples to be detected are normalized to the same image size in an image preprocessing stage, the image size is not fixed, and the size can be determined according to specific requirements;
The AdaBoost algorithm is adopted to distribute corresponding weight values for different features at the initial stage based on Harr features of the left eye and the right eye, weak classifier weights of different samples are calculated and updated after multiple iterations, finally when the classification error of the combined strong classifier is 0, the iteration is ended, and the trained model is used for detecting the left eye region and the right eye region to be detected.
9. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the steps of the method of any of claims 5 to 8 when the computer program is executed by the processor.
10. A storage medium having stored thereon computer program instructions which, when executed by a processor, implement the steps of the method of any of claims 5 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410109920.0A CN118097764A (en) | 2024-01-26 | 2024-01-26 | Left-right eye identification system and method applied to slit lamp, electronic equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410109920.0A CN118097764A (en) | 2024-01-26 | 2024-01-26 | Left-right eye identification system and method applied to slit lamp, electronic equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN118097764A true CN118097764A (en) | 2024-05-28 |
Family
ID=91146656
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410109920.0A Pending CN118097764A (en) | 2024-01-26 | 2024-01-26 | Left-right eye identification system and method applied to slit lamp, electronic equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118097764A (en) |
-
2024
- 2024-01-26 CN CN202410109920.0A patent/CN118097764A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107451998B (en) | Fundus image quality control method | |
| US7474775B2 (en) | Automatic detection of red lesions in digital color fundus photographs | |
| KR102155381B1 (en) | Method, apparatus and software program for cervical cancer decision using image analysis of artificial intelligence based technology | |
| US10395091B2 (en) | Image processing apparatus, image processing method, and storage medium identifying cell candidate area | |
| CN106056064A (en) | Face recognition method and face recognition device | |
| JPH07504056A (en) | Methods for identifying objects using data processing techniques | |
| US10952604B2 (en) | Diagnostic tool for eye disease detection using smartphone | |
| US20210090248A1 (en) | Cervical cancer diagnosis method and apparatus using artificial intelligence-based medical image analysis and software program therefor | |
| KR101854991B1 (en) | System and method for correcting color of digital image based on the human sclera and pupil | |
| CN116309584B (en) | Image processing system for cataract area identification | |
| CN114187223A (en) | New coronary pneumonia disease patient risk screening deep learning system based on ocular surface characteristics | |
| CN114937024A (en) | Image evaluation method and device and computer equipment | |
| KR20210124024A (en) | Method, apparatus and computer program for calculating quality score of fundus image data using artificial intelligence | |
| KR102034648B1 (en) | Medical Image Management System, Method and Computer Readable Recording Medium | |
| KR20210033902A (en) | Method, apparatus and software program for cervical cancer diagnosis using image analysis of artificial intelligence based technology | |
| Giancardo et al. | Quality assessment of retinal fundus images using elliptical local vessel density | |
| CN118097764A (en) | Left-right eye identification system and method applied to slit lamp, electronic equipment and storage medium | |
| CN115359548B (en) | Handheld intelligent pupil detection device and detection method | |
| KR20130063430A (en) | Analysis device of user cognition and method for analysis of user cognition | |
| US20220245811A1 (en) | Analysis of retinal imaging using video | |
| CN115439920A (en) | Consciousness state detection system and equipment based on emotional audio-visual stimulation and facial expression | |
| CN113287298B (en) | Image processing device, image processing method, and storage medium | |
| CN111612755B (en) | Lung focus analysis method, device, electronic equipment and storage medium | |
| CN115206494A (en) | Film reading system and method based on fundus image classification | |
| CN112580462A (en) | Feature point selection method, terminal and storage medium |
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 |