CN108852280A - A kind of Image Acquisition of vision drop and analysis method, system and equipment - Google Patents
A kind of Image Acquisition of vision drop and analysis method, system and equipment Download PDFInfo
- Publication number
- CN108852280A CN108852280A CN201810774153.XA CN201810774153A CN108852280A CN 108852280 A CN108852280 A CN 108852280A CN 201810774153 A CN201810774153 A CN 201810774153A CN 108852280 A CN108852280 A CN 108852280A
- Authority
- CN
- China
- Prior art keywords
- investigated
- those
- face portion
- iris
- portion image
- 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
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
- A61B3/12—Objective types, i.e. instruments for examining the eyes independent of the patients' perceptions or reactions for looking at the eye fundus, e.g. ophthalmoscopes
-
- 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
- A61B3/103—Objective types, i.e. instruments for examining the eyes independent of the patients' perceptions or reactions for determining refraction, e.g. refractometers, skiascopes
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/13—Edge detection
-
- 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
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Physics & Mathematics (AREA)
- Medical Informatics (AREA)
- General Health & Medical Sciences (AREA)
- Surgery (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Biomedical Technology (AREA)
- Molecular Biology (AREA)
- Ophthalmology & Optometry (AREA)
- Animal Behavior & Ethology (AREA)
- Biophysics (AREA)
- Public Health (AREA)
- Veterinary Medicine (AREA)
- Heart & Thoracic Surgery (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Radiology & Medical Imaging (AREA)
- Quality & Reliability (AREA)
- Eye Examination Apparatus (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
Abstract
The present invention relates to a kind of Image Acquisition of vision drop and analysis method, system and equipment.Method includes the following steps：A1, when the pupil of those who are investigated is in magnifying state, open be set to its eye immediately ahead of point light source its eye ground is illuminated, and obtain a those who are investigated in described illuminate the first face portion image；B1, the iris range boundary for identifying those who are investigated in the first face portion image, the ametropia reflective pixel in eyeground is calculated using the characteristic to red, green, blue triple channel within the scope of iris and calculates the statistical property of the transverse and longitudinal coordinate distribution of the pixel, and the ametropic degree of those who are investigated is calculated by the statistical property that the transverse and longitudinal coordinate of the pixel is distributed.Technical solution of the present invention can be used under including the scene in family by general masses, and process for using is simple and fast.Technical solution of the present invention completes a variety of eyesight detection risks such as ametropia, strabismus in the inspection for being once as short as the several seconds.
Description
Technical field
The present invention relates to eye examination or vision drop fields, and in particular to a kind of Image Acquisition and analysis of vision drop
Method, system and equipment.
Background technique
A variety of ophthalmology diseases and with eye it is improper can all cause with ametropia (myopia/farsightedness) and strabismus be representative
Affect vision function damage, influences the quality of life of patient throughout one's life.Wherein, due to infant to the cognitive ability of vision impairment and
Communication capability is insufficient, often can just cause guardian's note that risk is especially prominent and harm is larger after its severe exacerbation.
Although existing detection methods multiplicity, generally requires subject to interact, focal attention force-fitting inspection, and
This is difficult to carry out infant；Automation inspection equipment cost is higher, and there is still a need for those who are investigated to a certain extent
Cooperation.Specifically, there are following one or more defects in commercially available and other published designs：
1. needing those who are investigated's interoperation, specific objective is persistently such as stared, block simple eye or eyes or operates dedicated set
It is standby, it is not suitable for checking infant；
2. the equipment of artificial screening is simple and cheap, but operating process is more complex, is not suitable for layman and uses；
3. special equipment design is complicated, at high cost, be not suitable for a wide range of universal；
4. may be used when traditional screening and arrive mydriasis drug, it is harmful to the human body；
5. special equipment is not suitable for selling to ordinary consumer, it is unfavorable for improving screening frequency to disclosure risks as early as possible.
Summary of the invention
The technical problem to be solved by the present invention is to：It proposes a kind of based on real by the intelligent camera apparatus of representative of smart phone
The Image Acquisition and analysis method, system and equipment of existing vision drop.
The technical solution (one) that the present invention proposes to solve above-mentioned technical problem is：A kind of Image Acquisition of vision drop with
Analysis method includes the following steps：
A1, when the pupil of those who are investigated is in magnifying state, open be set to its eye immediately ahead of point light source to its eye
Bottom retina is illuminated, and obtains the first face portion image of a those who are investigated in described illuminate；
B1, the iris range boundary for identifying those who are investigated in the first face portion image, utilize within the scope of iris
The ametropia reflective pixel in eyeground is calculated to the characteristic of red, green, blue triple channel and calculates the transverse and longitudinal of the pixel
The statistical property of coordinate distribution calculates those who are investigated's dioptric not by the statistical property that the transverse and longitudinal coordinate of the pixel is distributed
Positive degree.
It further, further include a2, when the pupil of those who are investigated is in contraction state, by being set to its drawing axis
Point light source carries out blast to its eyeball position, and obtains the second face portion image of those who are investigated this moment；
B2, the iris range boundary for identifying two eyes of those who are investigated in the second face portion image, are searched for respectively
And the brightness of red, green, blue triple channel reaches the pixel coordinate of maximum value in the iris range boundary of two eyes of record, counts respectively
The difference between its coordinate mean value and corresponding eye iris circle center point is calculated, the absolute difference of this difference by calculating two eyes
It can calculate the degree of strabismus.
Further, when carrying out a1 and a2 step, under the premise of the pupil of those who are investigated is in magnifying state, first lead to
Cross camera and enter the continued recording of video recording mode, then by control flash lamp be spaced the flash of light of high brightness twice and
So that flash lamp is kept being always on for normal brightness between high brightness flash of light twice, then terminates video record；
The acquisition of the first face portion image and the second face portion image refers in every frame image of institute's recorded video,
Brightness is selected after high-intensitive flash of light triggering twice respectively more than threshold value and reach a frame and on time for local maximum for the first time
Between sequence be successively successively determined as the first face portion image and the second face portion image.It is allow program inhibition pre-flash
In system, the photo for acquisition of taking a picture twice is also can be used in the first direct picture and the second direct picture.
Further, when b1 identifies the iris range boundary of those who are investigated in the first face portion image, including
Following sub-step
B11, by the preparatory trained eyes identification model based on deep learning mode, in the first face portion figure
The location and range of eyes are estimated as in；
B12, in the location and range of the eyes by circle recognizer calculate and search in the range of every eye
It is simultaneously determined as iris range boundary by the several picture boundary of the closest circle of rope or circular arc.
The technical solution (two) that the present invention proposes to solve above-mentioned technical problem is：A kind of Image Acquisition of vision drop with
Analysis system, including camera control module, flash of light lamp control module and computing module；
The flash of light lamp control module is suitable for when the pupil of those who are investigated is in magnifying state, and control flash lamp carries out high
The flash of light of brightness is illuminated with the eye ground to the those who are investigated；
The camera control module be suitable for by camera when the eye ground of those who are investigated is illuminated, get by
The first face portion image of examiner；
The computing module is suitable for identifying the iris range boundary of those who are investigated in the first face portion image, in rainbow
It is utilized within the scope of film and the ametropia reflective pixel in eyeground is calculated to the characteristic of red, green, blue triple channel and is calculated described
The statistical property of the transverse and longitudinal coordinate distribution of pixel, by the statistical property that the transverse and longitudinal coordinate of the pixel is distributed calculate by
The ametropic degree of examiner.
Further, the flash of light lamp control module is further adapted for when the pupil of those who are investigated is in contraction state, control
Flash lamp carries out blast to its eyeball position；
The camera control module is further adapted for through camera when the eyeball position of those who are investigated is by blast, obtains tested
The second face portion image for the person of looking into；
The computing module is further adapted for identifying the iris model of two eyes of those who are investigated in the second face portion image
Surrounding edge circle, searches for respectively and records red, green, blue triple channel brightness in the iris range boundary of two eyes and reach maximum value
Pixel coordinate calculates separately the difference between its coordinate mean value and corresponding eye iris circle center point, passes through this of two eyes of calculating
The absolute difference of one difference can calculate the degree of strabismus.
Further, the flash of light lamp control module is spaced in described illuminate with blast by controlling flash lamp
High brightness twice glistens and so that flash lamp is kept being always on for normal brightness between the flash of light of high brightness twice,
The first face portion image of the acquisition and the second face portion image are carried out by camera control module control camera
Continued video record, then extract in video file highest two frame of brightness and chronological be successively successively determined as the
One face portion image and the second face portion image.
Further, in identifying the first face portion image when the iris range boundary of those who are investigated, the meter
It calculates module and is suitable for first passing through the preparatory trained eyes identification model based on deep learning mode, in the first face portion figure
The location and range of eyes are estimated as in, then pass through round recognizer in every eye in the location and range of the eyes
In the range of calculate and search for the several picture boundary of closest circle or circular arc and it is determined as iris range boundary.
The technical solution (three) that the present invention proposes to solve above-mentioned technical problem is：A kind of Image Acquisition of vision drop with
Analytical equipment, the equipment have CPU, flash lamp and camera, and the equipment is suitable for being in amplification shape in the pupil of those who are investigated
When state, opens the flash lamp being set to immediately ahead of its eye and its eye ground is illuminated, and obtain one by camera
The first face portion image of those who are investigated in described illuminate；
The equipment is suitable for identifying the iris range boundary of those who are investigated in the first face portion image, in iris model
Interior utilize is enclosed to calculate the ametropia reflective pixel in eyeground to the characteristic of red, green, blue triple channel and calculate the pixel
The statistical property of the transverse and longitudinal coordinate distribution of point is calculated examined by the statistical property that the transverse and longitudinal coordinate of the pixel is distributed
The ametropic degree of person.
It further, further include that the equipment is suitable for when the pupil of those who are investigated is in contraction state, by being set to it
Flash lamp immediately ahead of eyes carries out blast to its eyeball position, and by camera acquisition one this moment those who are investigated second
Face portion image；
The equipment is suitable for identifying the iris range boundary of two eyes of those who are investigated in the second face portion image,
It searches for respectively and records red, green, blue triple channel brightness in the iris range boundary of two eyes and reach the pixel of maximum value and sit
Mark, calculates separately the difference between its coordinate mean value and corresponding eye iris circle center point, by this difference for calculating two eyes
Absolute difference can calculate the degree of strabismus.
The beneficial effects of the invention are as follows：
Based on every innovative point, technical solution of the present invention provides following gain compared with conventional art：
Without special equipment and special operation process, screening function is realized using the intelligent capture apparatus such as smart phone.By
It is very high in smart phone popularization degree, and widely general user is familiar with its operating method, therefore technical solution of the present invention can be by
General masses use under including the scene in family, not only reduce equipment cost, the technical staff's deployment cost of vision drop,
And user's routine use can be allowed to improve screening frequency, infant is additionally aided at home or in environment known to school etc.
Loosen to cooperate completion screening.
Process for using is simple and fast.Technical solution of the present invention is completed ametropia, oblique in the inspection for being once as short as the several seconds
Depending on etc. a variety of eyesight detection risks, and interacted by sound, light non-conversational mode with subject, effectively prevent baby children
Youngster is difficult to the difficulty for cooperating dialogue to command, so that subject's the range of age substantially extends.
Test result realizes automated analysis.It is analyzed compared to widely used artificial estimation, technical solution of the present invention
Analytic process is not necessarily to manual intervention, not only reduces human cost, and result is quantitative, is conducive to instruct therapeutic intervention and continue
Monitoring.
Detailed description of the invention
The Image Acquisition of vision drop of the invention and analysis method, system and equipment are made with reference to the accompanying drawing further
Explanation.
Fig. 1 be method in the present invention, system and equipment obtain acquire image when by horizontal time axis operational flowchart
Schematic diagram；
Fig. 2 is method in the present invention, system and equipment in the flow chart analyzed acquired video or photo；
Fig. 3 is index path when point light source illuminates normal eyes retina；
Fig. 4 is that normal eyes are illuminated photo collected after its retina；
Fig. 5 is index path when point light source illuminates near-sighted eyes retina；
Fig. 6 is that myopia is illuminated photo collected after its retina；
Fig. 7 is index path when point light source illuminates long sight eyes retina；
Fig. 8 is the photo collected when carrying out strabismus checking；
Fig. 9 is the schematic diagram of iris range and corneal reflection；
Figure 10 is the shown reflective spot position view of non-slanting eye；
Figure 11 is the shown reflective spot position view of horizontal strabismus eye；
Figure 12 is the shown reflective spot position view of vertical squint eye.
Wherein, " shooting control " part in Fig. 1 includes video recording and two kinds of acquisition shapes for selecting a selection of taking pictures simultaneously
Formula.
Specific embodiment
Referring in detail to attached drawing and embodiment is described now.In order to fully understand the present invention, it mentions in the following detailed description
Numerous details.It will be appreciated by those skilled in the art that the present invention may not need these details and realize.?
In other examples, it is not described in detail well known method, process, component and circuit, in order to avoid unnecessarily make embodiment fuzzy.
Embodiment one：
Method can be based on common intelligent filming apparatus, Smartphone device or with flash lamp, camera in the present invention
And the hardware such as separate type/integrated system of operational part are carried out and are realized.
Method of the invention for ease of understanding, in the present embodiment in conjunction with Smartphone device operation for retouched
It states, preceding method includes the following steps：
A1, when the pupil of those who are investigated is in magnifying state, open be set to its eye immediately ahead of point light source to its eye
Bottom retina is illuminated, and obtains the first face portion image of a those who are investigated in described illuminate.
It in the present embodiment, is to be placed in those who are investigated in order to enable the pupil of those who are investigated is at magnifying state
In darkroom (having using human eye can magnify as far as possible pupil to see the characteristic of thing clearly under low light environment), naturally it is also possible to
Light weak night carries out.
Point light source in this example is the flash lamp (can certainly be external flash lamp) that Smartphone device carries, and is shone
The acquisition of piece is then carried out by the camera that Smartphone device carries.Then step can then be showed when actually carrying out in a1
For (one of several scenes)：In checking process, it is desirable that subject is stood firm or settled down in half darkroom.Examiner its just
Face, hand-held intelligent mobile phone or is fixed on bracket, and opens related shooting software, the camera preview interface in shooting software
On the eyes position mark that can preferably be labeled with, subject's eyes snap to aforementioned eyes position mark by examiner, with
By mobile phone to subject eye distance from adjusting to suitable.Those who are investigated's its pupil after adapting to dark surrounds amplifies, and flash lamp is prominent this moment
So strong sudden strain of a muscle and camera absorb be reflected into extraneous imaging again through its crystal after those who are investigated's retina is illuminated simultaneously.
As shown in figure 3, if the normal visual acuity of those who are investigated, the reflective of retina penetrates crystal backtracking, due to camera not position
In on the reflected line of light, therefore as shown in figure 4, camera intake is reflective less than the retina of those who are investigated or be only capable of intake and arrive
The retina of kermesinus is reflective (red-eye effect), and in addition bright spot is the cornea of those who are investigated because of flash of light light irradiation formation in Fig. 4
Luminous point.As shown in figure 5, if those who are investigated have myopia, retina it is reflective, because be biased under the refraction action of lesion crystal into
Light side is penetrated, and before falling in retina because of rear lens focal plane, is formed on the CMOS chip of camera as shown in Figure 6 at this time
Because of ametropia reflective hot spot.If those who are investigated have long sight, as shown in fig. 7, repeating no more.
B1, the iris range boundary for identifying those who are investigated in the first face portion image, utilize within the scope of iris
The ametropia reflective pixel in eyeground is calculated to the characteristic of red, green, blue triple channel and calculates the transverse and longitudinal of the pixel
The statistical property of coordinate distribution calculates those who are investigated's dioptric not by the statistical property that the transverse and longitudinal coordinate of the pixel is distributed
Positive degree.In the present embodiment, aforementioned identification, calculate be using smart phone CPU and its peripheral module realize, and
Finally result is exported on the display screen of smart phone.When calculating, within the scope of iris, by brightness and shape distinguish two kinds with
In the characteristics of image for judging eyesight risk：I, uniform kermesinus ophthalmoscopic image；Ii, the higher local speck of brightness.Wherein, two
The brightness of (i) is inconsistent in eye means lenticular opacities, and the appearance of (ii) represents ametropia, and its shape has with degree
It closes.
It can be used as preferably：The iris range boundary of those who are investigated in the first face portion image is identified in b1
When, including following sub-step
(b11), by the preparatory trained eyes identification model based on deep learning mode, in first face portion
The location and range of eyes are estimated in image；
(b12), calculated in the range of every eye by round recognizer in the location and range of the eyes and
It searches for the several picture boundary of closest circle or circular arc and it is determined as iris range boundary, such as become using such as Hough circle
It changes detection algorithm and is accurately positioned every eye iris center and its radius.
To further realize the screening to strabismus, method further includes in the present invention：
A2, when the pupil of those who are investigated is in contraction state, by be set to its drawing axis point light source to its eyeball
Position carries out blast, and obtains the second face portion image of those who are investigated this moment.
It in this step, is that the flash lamp in a1 exists in order to enable the pupil of those who are investigated is at magnifying state
It is restored to Conventional brightness after strong sudden strain of a muscle and keeps being always on, the pupil of those who are investigated is shunk automatically because light stimulates at this time, makes its pupil
The purpose that hole is shunk is to avoid reflective mix of reflective and cornea of retina to overlap CMOS in the acquisition of subsequent image
On, so that influencing the later period accurately analyzes slant visibility.In this step, the sudden strain of a muscle when flash lamp holding is always on and reaches preset duration
Light lamp dodges by force again, and the camera second face portion image of intake to those who are investigated while strong sudden strain of a muscle again.
B2, the iris range boundary for identifying two eyes of those who are investigated in the second face portion image, are searched for respectively
And it records red, green, blue triple channel brightness in the iris range boundaries of two eyes and reaches maximum value (i.e. flash lamp is on iris
Reflective spot) pixel coordinate, calculate separately the difference between its coordinate mean value and corresponding eye iris circle center point, pass through calculating
The absolute difference of this difference of two eyes can calculate the degree of strabismus, and projectional technique is as follows.
By taking horizontal strabismus as an example, the strabismus degree as unit of pixel is calculated first.If in left eye brightness maxima point with
Iris circle center point coordinate difference is (Δ xL,ΔyL), brightness maxima point is with iris circle center point coordinate difference in right eye
(ΔxR,ΔyR).According to definition, the distance of horizontal strabismus is Sbhrz,px=| Δ xL-ΔxR| unit is pixel.
In view of the average interpupillary distance (i.e. two iris center spacing) of the mankind is about 63 millimeters and error is seldom more than positive and negative
10%, the conversion that being accurately positioned according to the iris in abovementioned steps can calculate in photo between physical length and length in pixels is closed
System, and then calculate horizontal, vertical squint physics off-axis length.For referring to iris diameter calculation, if the interpupillary distance identified
dip,pxPixel, the conversion scale of physical length and length in pixels is in photo：The physical distance of horizontal strabismus is
Sbhrz,mm=μ | Δ xL-ΔxR|.Because the eyeball diameter of the mankind is about 24.5 millimeters and error is minimum, can be according to trigonometric function
Calculate horizontal strabismus angle
Similar, vertical squint angle
And this is the standard of clinical diagnosis strabismus.Since the iris diameter of people is about 10.8mm and error range is minimum,
Above-mentioned reckoning is also thus completed, method is identical, repeats no more.
Identification, search and calculating in b2 are also that the CPU and its peripheral module by utilizing smart phone are realized, no longer
It repeats.
As shown in Figure 1, the acquisition of the first, second face portion image in the a1 and a2 of the present embodiment is existed by camera
Obtained by being taken pictures while flash lamp dodges by force.
It can also be used as preferably by gained after video recording：As also shown in Figure 1, amplification is in the pupil of those who are investigated
Under the premise of state, camera is first entered into the continued recording of video recording mode, then be spaced twice by controlling flash lamp
High brightness flash of light and make flash lamp that normal brightness be kept to be always on (flashlight mode) between the flash of light of high brightness twice so that
This flash lamp lights-maintains-and extinguishing process videoed complete documentation, then terminate video record.And it can be in flash lamp
Flashlight mode when opening, by rhythmical flashing, play the means such as sound those who are investigated attracted to see to flash lamp to keep
Its pupil contraction.As shown in Fig. 2, it is decoded frame by frame and bright using the pixel in each frame in the video processing to recording
Current flash state is estimated in degree distribution.If flash lamp has been likely to be breached maximum brightness, this frame is cached.It is handling
After complete all frames, selection flash lamp reaches or approaches a frame of maximum brightness or multiframe is saved.In the present embodiment, by institute
Brightness is more than threshold value and reaches part for the first time after selecting high-intensitive flash of light triggering twice in every frame image of recorded video respectively
A highest frame simultaneously chronological is successively successively determined as the first face portion image and the second face portion image.Wherein, preceding
" part " stated refers to that brightness continues to exceed the subrange that the multiframe of threshold value is constituted.
Embodiment two：
The present embodiment is the Image Acquisition and analysis system of the vision drop designed on the basis of embodiment one.Embodiment one
On the basis of middle method, which includes camera control module, flash of light lamp control module and computing module.
The flash of light lamp control module is suitable for when the pupil of those who are investigated is in magnifying state, and control flash lamp carries out high
The flash of light of brightness is illuminated with the eye ground to the those who are investigated.
The camera control module be suitable for by camera when the eye ground of those who are investigated is illuminated, get by
The first face portion image of examiner.
The computing module is suitable for identifying the iris range boundary of those who are investigated in the first face portion image, in rainbow
It is utilized within the scope of film and the ametropia reflective pixel in eyeground is calculated to the characteristic of red, green, blue triple channel and is calculated described
The statistical property of the transverse and longitudinal coordinate distribution of pixel, by the statistical property that the transverse and longitudinal coordinate of the pixel is distributed calculate by
The ametropic degree of examiner.
It can be used as preferably：The flash of light lamp control module is further adapted for being in contraction state in the pupil of those who are investigated
When, control flash lamp carries out blast to its eyeball position.The camera control module is further adapted for through camera in those who are investigated
Eyeball position by blast when, obtain the second face portion image of those who are investigated.The computing module is further adapted for identifying described
The iris range boundary of two eyes of those who are investigated, searches for respectively and records the iris model of two eyes in second face portion image
The brightness of red, green, blue triple channel reaches the pixel coordinate of maximum value in surrounding edge circle, calculates separately its coordinate mean value and corresponding eye rainbow
Difference between film circle center point can calculate the degree of strabismus by calculating the absolute difference of this difference of two eyes.
It can be used as preferably：The flash of light lamp control module is in described illuminate with blast, by controlling flash lamp
It is spaced high brightness flash of light twice and so that flash lamp is kept being always on for normal brightness between the flash of light of high brightness twice, it is described
It obtains the first face portion image and the second face portion image is to carry out continued video record by camera control module control camera
It makes, then extracts brightness after high intensity flash of light twice triggers respectively in video file and be more than threshold value and reach local highest for the first time
One frame simultaneously chronological is successively successively determined as the first face portion image and the second face portion image.
It can be used as preferably：The iris range boundary of those who are investigated in identifying the first face portion image
When, the computing module is suitable for first passing through the preparatory trained eyes identification model based on deep learning mode, described the
The location and range of eyes are estimated in one face portion image, then are calculated in the location and range of the eyes by round identification
Method calculates and searches for the several picture boundary of closest circle or circular arc in the range of every eye and it is determined as iris model
Surrounding edge circle.
Embodiment three：
The present embodiment is the Image Acquisition and analytical equipment of the vision drop designed on the basis of embodiment one.Embodiment one
On the basis of middle method, which has CPU, flash lamp and camera, and the equipment, which is suitable for being in the pupil of those who are investigated, puts
When big state, opens the flash lamp being set to immediately ahead of its eye and its eye ground is illuminated, and obtained by camera
The first face portion image of one those who are investigated in described illuminate.
The equipment is suitable for identifying the iris range boundary of those who are investigated in the first face portion image, in iris model
Interior utilize is enclosed to calculate the ametropia reflective pixel in eyeground to the characteristic of red, green, blue triple channel and calculate the pixel
The statistical property of the transverse and longitudinal coordinate distribution of point is calculated examined by the statistical property that the transverse and longitudinal coordinate of the pixel is distributed
The ametropic degree of person.
It can be used as preferably：Further include that the equipment is suitable for when the pupil of those who are investigated is in contraction state, leads to
The flash lamp crossed immediately ahead of its eye carries out blast to its eyeball position, and obtains one by camera and be examined this moment
The second face portion image of person；
The equipment is suitable for identifying the iris range boundary of two eyes of those who are investigated in the second face portion image,
It searches for respectively and records red, green, blue triple channel brightness in the iris range boundary of two eyes and reach the pixel of maximum value and sit
Mark, calculates separately the difference between its coordinate mean value and corresponding eye iris circle center point, by this difference for calculating two eyes
Absolute difference can calculate the degree of strabismus.
Of the invention is not limited to the above embodiment, and the technical solution of above-mentioned each embodiment of the invention can be handed over each other
Fork combination form new technical solution, in addition it is all using equivalent replacement formed technical solution, all fall within the present invention claims guarantor
It protects in range.
Claims (10)
1. the Image Acquisition and analysis method of a kind of vision drop, including following characteristics step：
A1, when the pupil of those who are investigated is in magnifying state, open be set to its eye immediately ahead of point light source to its eyeground regard
Nethike embrane is illuminated, and obtains the first face portion image of a those who are investigated in described illuminate；
B1, the iris range boundary for identifying those who are investigated in the first face portion image, within the scope of iris using to it is red,
The characteristic of green, blue triple channel calculates the ametropia reflective pixel in eyeground and calculates the transverse and longitudinal coordinate point of the pixel
The statistical property of cloth calculates the ametropic degree of those who are investigated by the statistical property that the transverse and longitudinal coordinate of the pixel is distributed
Number.
2. the Image Acquisition and analysis method of vision drop according to claim 1, it is characterised in that：Further include
A2, when the pupil of those who are investigated is in contraction state, by be set to its drawing axis point light source to its eyeball position
Blast is carried out, and obtains the second face portion image of those who are investigated this moment；
B2, the iris range boundary for identifying two eyes of those who are investigated in the second face portion image, search for respectively and remember
The pixel coordinate that red, green, blue triple channel brightness in the iris range boundary of two eyes reaches maximum value is recorded, it is calculated separately
Difference between coordinate mean value and corresponding eye iris circle center point can be pushed away by calculating the absolute difference of this difference of two eyes
Calculate the degree of strabismus.
3. the Image Acquisition and analysis method of vision drop according to claim 2, it is characterised in that：Carrying out a1 and a2 step
When rapid, under the premise of the pupil of those who are investigated is in magnifying state, first passes through camera and enter the continued record of video recording mode
System, then the flash of light of high brightness twice is spaced by control flash lamp and keeps flash lamp between the flash of light of high brightness twice
Being always on for normal brightness, then terminates video record；
The acquisition of the first face portion image and the second face portion image refers in every frame image of institute's recorded video respectively
After choosing high-intensitive flash of light triggering twice, brightness is more than threshold value and the frame and in chronological order for reaching local maximum for the first time
Be successively successively determined as the first face portion image and the second face portion image.
4. the according to claim 1, Image Acquisition and analysis method of 2 or 3 vision drops, it is characterised in that：
When b1 identifies the iris range boundary of those who are investigated in the first face portion image, including following sub-step
B11, by the preparatory trained eyes identification model based on deep learning mode, in the first face portion image
Estimate the location and range of eyes；
B12, it by round recognizer calculates and searches for most in the range of every eye in the location and range of the eyes
To approach the several picture boundary of circle or circular arc and it being determined as iris range boundary.
5. the Image Acquisition and analysis system of a kind of vision drop, it is characterised in that：It is controlled including camera control module, flash lamp
Module and computing module；
The flash of light lamp control module is suitable for when the pupil of those who are investigated is in magnifying state, and control flash lamp carries out high brightness
Flash of light illuminated with the eye ground to the those who are investigated；
The camera control module is suitable for through camera when the eye ground of those who are investigated is illuminated, and gets examined
The first face portion image of person；
The computing module is suitable for identifying the iris range boundary of those who are investigated in the first face portion image, in iris model
Interior utilize is enclosed to calculate the ametropia reflective pixel in eyeground to the characteristic of red, green, blue triple channel and calculate the pixel
The statistical property of the transverse and longitudinal coordinate distribution of point is calculated examined by the statistical property that the transverse and longitudinal coordinate of the pixel is distributed
The ametropic degree of person.
6. the Image Acquisition and analysis system of vision drop according to claim 5, it is characterised in that：
The flash of light lamp control module is further adapted for when the pupil of those who are investigated is in contraction state, controls flash lamp to its eyeball
Position carries out blast；
The camera control module is further adapted for through camera when the eyeball position of those who are investigated is by blast, obtains those who are investigated
The second face portion image；
The computing module is further adapted for identifying the iris range side of two eyes of those who are investigated in the second face portion image
The pixel that red, green, blue triple channel brightness in the iris range boundary of two eyes reaches maximum value is searched for respectively and is recorded on boundary
Coordinate calculates separately the difference between its coordinate mean value and corresponding eye iris circle center point, by calculate two eyes this is poor
The absolute difference of value can calculate the degree of strabismus.
7. the Image Acquisition and analysis system of vision drop according to claim 6, it is characterised in that：
The flash of light lamp control module is spaced the sudden strain of a muscle of high brightness twice in described illuminate with blast, by controlling flash lamp
Light simultaneously makes flash lamp keep being always on for normal brightness between the flash of light of high brightness twice,
The first face portion image of the acquisition and the second face portion image are continued by camera control module control camera progress
Video record, then after extracting in video file high-intensitive flash of light triggering twice respectively, brightness is more than threshold value and reaches for the first time
One frame of local maximum simultaneously chronological is successively successively determined as the first face portion image and the second face portion image.
8. according to claim 5,6 or the Image Acquisition and analysis method of 7 vision drops, it is characterised in that：It is identifying
In the first face portion image when iris range boundary of those who are investigated, the computing module is suitable for first passing through training in advance
The eyes identification model based on deep learning mode, position and the model of eyes are estimated in the first face portion image
It encloses, then calculates and search in the range of every eye by round recognizer in the location and range of the eyes and connect the most
It is simultaneously determined as iris range boundary by the several picture boundary of nearly circle or circular arc.
9. the Image Acquisition and analytical equipment of a kind of vision drop, which has CPU, flash lamp and camera, and feature exists
In：
The equipment is suitable for when the pupil of those who are investigated is in magnifying state, opens the flash lamp pair being set to immediately ahead of its eye
Its eye ground is illuminated, and the first face portion figure of a those who are investigated in described illuminate is obtained by camera
Picture；
The equipment is suitable for identifying the iris range boundary of those who are investigated in the first face portion image, within the scope of iris
The ametropia reflective pixel in eyeground is calculated using the characteristic to red, green, blue triple channel and calculates the pixel
The statistical property of transverse and longitudinal coordinate distribution calculates those who are investigated by the statistical property that the transverse and longitudinal coordinate of the pixel is distributed and bends
The not positive degree of light.
10. the Image Acquisition and analytical equipment of vision drop according to claim 9, it is characterised in that：Further include
The equipment is suitable for when the pupil of those who are investigated is in contraction state, passes through the flash lamp pair being set to immediately ahead of its eye
Its eyeball position carries out blast, and passes through the second face portion image of camera one those who are investigated this moment of acquisition；
The equipment is suitable for identifying the iris range boundary of two eyes of those who are investigated in the second face portion image, respectively
The pixel coordinate that red, green, blue triple channel brightness in the iris range boundary of two eyes reaches maximum value is searched for and records, point
Do not calculate the difference between its coordinate mean value and corresponding eye iris circle center point, by calculate two eyes this difference it is absolute
Difference can calculate the degree of strabismus.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810774153.XA CN108852280A (en) | 2018-07-16 | 2018-07-16 | A kind of Image Acquisition of vision drop and analysis method, system and equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810774153.XA CN108852280A (en) | 2018-07-16 | 2018-07-16 | A kind of Image Acquisition of vision drop and analysis method, system and equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108852280A true CN108852280A (en) | 2018-11-23 |
Family
ID=64301864
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810774153.XA Pending CN108852280A (en) | 2018-07-16 | 2018-07-16 | A kind of Image Acquisition of vision drop and analysis method, system and equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108852280A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109636796A (en) * | 2018-12-19 | 2019-04-16 | 中山大学中山眼科中心 | A kind of artificial intelligence eye picture analyzing method, server and system |
| CN110141188A (en) * | 2019-05-15 | 2019-08-20 | 佛山科学技术学院 | A kind of big visual field fundus camera Uniform Illumination method and system |
| CN110287796A (en) * | 2019-05-24 | 2019-09-27 | 北京爱诺斯科技有限公司 | A kind of dioptric screening method based on mobile phone and external equipment |
| CN113899312A (en) * | 2021-09-30 | 2022-01-07 | 苏州天准科技股份有限公司 | Image measuring equipment and image measuring method |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5011276A (en) * | 1988-06-27 | 1991-04-30 | Ryusyo Industrial Co., Ltd. | Apparatus for measuring refractive power of eye |
| CN102085089A (en) * | 2009-12-02 | 2011-06-08 | 尼德克株式会社 | Eye refractive power measurement apparatus, refractive error detection apparatus and method |
| CN102961117A (en) * | 2012-11-06 | 2013-03-13 | 温州医学院 | Strabismus diagnosis device based on mobile platform |
| CN104661580A (en) * | 2012-07-06 | 2015-05-27 | Iriss医疗科技有限公司 | Strabismus detection |
| CN106037627A (en) * | 2016-05-20 | 2016-10-26 | 上海青研科技有限公司 | Full-automatic visual acuity examination method and device for infants |
| CN106264444A (en) * | 2016-10-20 | 2017-01-04 | 鹄誉医疗科技（上海）有限公司 | A kind of human eye method for detecting diopter of high robust |
| CN107890336A (en) * | 2017-12-05 | 2018-04-10 | 中南大学 | Diopter detecting system based on intelligent handheld device |
-
2018
- 2018-07-16 CN CN201810774153.XA patent/CN108852280A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5011276A (en) * | 1988-06-27 | 1991-04-30 | Ryusyo Industrial Co., Ltd. | Apparatus for measuring refractive power of eye |
| CN102085089A (en) * | 2009-12-02 | 2011-06-08 | 尼德克株式会社 | Eye refractive power measurement apparatus, refractive error detection apparatus and method |
| CN104661580A (en) * | 2012-07-06 | 2015-05-27 | Iriss医疗科技有限公司 | Strabismus detection |
| CN102961117A (en) * | 2012-11-06 | 2013-03-13 | 温州医学院 | Strabismus diagnosis device based on mobile platform |
| CN106037627A (en) * | 2016-05-20 | 2016-10-26 | 上海青研科技有限公司 | Full-automatic visual acuity examination method and device for infants |
| CN106264444A (en) * | 2016-10-20 | 2017-01-04 | 鹄誉医疗科技（上海）有限公司 | A kind of human eye method for detecting diopter of high robust |
| CN107890336A (en) * | 2017-12-05 | 2018-04-10 | 中南大学 | Diopter detecting system based on intelligent handheld device |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109636796A (en) * | 2018-12-19 | 2019-04-16 | 中山大学中山眼科中心 | A kind of artificial intelligence eye picture analyzing method, server and system |
| CN110141188A (en) * | 2019-05-15 | 2019-08-20 | 佛山科学技术学院 | A kind of big visual field fundus camera Uniform Illumination method and system |
| CN110141188B (en) * | 2019-05-15 | 2023-08-08 | 佛山科学技术学院 | Uniform illumination method and system for large-view-field fundus camera |
| CN110287796A (en) * | 2019-05-24 | 2019-09-27 | 北京爱诺斯科技有限公司 | A kind of dioptric screening method based on mobile phone and external equipment |
| CN110287796B (en) * | 2019-05-24 | 2020-06-12 | 北京爱诺斯科技有限公司 | Refractive screening method based on mobile phone and external equipment |
| CN113899312A (en) * | 2021-09-30 | 2022-01-07 | 苏州天准科技股份有限公司 | Image measuring equipment and image measuring method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10426332B2 (en) | System and device for preliminary diagnosis of ocular diseases | |
| CN108852280A (en) | A kind of Image Acquisition of vision drop and analysis method, system and equipment | |
| CN109310314B (en) | Mobile device application for eye skew measurement | |
| CN109431452B (en) | Unmanned eye health screening instrument | |
| CN107184178A (en) | A kind of hand-held vision drop instrument of intelligent portable and optometry method | |
| GB2491941A (en) | Image processing to detect abnormal eye conditions | |
| US20110157550A1 (en) | Adaptive Photoscreening System | |
| US6663242B1 (en) | Simultaneous, wavelength multiplexed vision screener | |
| US6616277B1 (en) | Sequential eye screening method and apparatus | |
| KR20140103900A (en) | System and methods for documenting and recording of the pupillary red reflex test and corneal light reflex screening of the eye in infants and young children | |
| US20230263385A1 (en) | Methods and apparatus for making a determination about an eye | |
| CN110916608B (en) | Diopter detection device | |
| CN110279391A (en) | Portable infrared camera eyesight detection algorithm | |
| KR20190041818A (en) | Method and Apparatus for jaundice diagnosis based on an image | |
| CN210582453U (en) | Portable mydriasis-free eye fundus camera | |
| KR20220024411A (en) | eye test | |
| CN105380590B (en) | A kind of equipment and its implementation with eye position detection function | |
| CN105996986B (en) | A kind of devices and methods therefor based on multispectral detection human eye Meibomian gland model | |
| Zheng et al. | An automatic stimulus and synchronous tracking system for strabismus assessment based on cover test | |
| KR101369565B1 (en) | Pupil measuring system with smart device and method for measuring pupil by using the system | |
| CN113974546A (en) | Pterygium detection method and mobile terminal | |
| US10531794B1 (en) | Photorefractive flash device and system | |
| CN115153425A (en) | Portable eye surface disease analysis and diagnosis system | |
| EP3695775B1 (en) | Smartphone-based handheld optical device and method for capturing non-mydriatic retinal images | |
| CN108230287B (en) | Method and device for detecting crystal region of anterior segment image |
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: 20181123 |
|
| RJ01 | Rejection of invention patent application after publication |