CN103565400A - Auxiliary image recognition method - Google Patents
Auxiliary image recognition method Download PDFInfo
- Publication number
- CN103565400A CN103565400A CN201210281864.6A CN201210281864A CN103565400A CN 103565400 A CN103565400 A CN 103565400A CN 201210281864 A CN201210281864 A CN 201210281864A CN 103565400 A CN103565400 A CN 103565400A
- Authority
- CN
- China
- Prior art keywords
- image
- mentioned
- contrast
- recognition method
- split
- 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.)
- Granted
Links
Images
Abstract
The invention relates to an auxiliary image recognition method, comprising the following steps of generating image information with a split-image screen by using a testing instrument; setting a region of interest (ROI) around the split-image screen of the image information; adding the pixel brightness of the image information of the ROI, so that all image information brightness contrast in the ROI is improved; contrasting and correcting the ROI with improved brightness contrast, so that the contrast of brightness of the split-image screen in the ROI and a peripheral block of the split-image screen is increased. The split-screen image generated by the traditional ophthalmic test instrument is specially optimized, so that the ophthalmoscope image output contrast and sharpness are improved, and an operator can easily recognize two rectangular images in the split-image screen, and therefore the vertical relative position of the two rectangular images is adjusted to achieve optimum focusing operation.
Description
Technical field
The present invention is about a kind of image recognition method that is applied to optical detection focusing, refer to especially a kind of will in detecting instrument, split the auxiliary image recognition method optimized of picture screen image contrast and sharpness, make detecting instrument operator easily the ophthalmofundoscope image output after identification optimization complete focusing work.
Background technology
Mostly tradition fundus camera is to utilize and splits the mode that resembles screen (Split image) and carry out auxiliary focusing, refer to shown in the 1st figure, what is called is split as screen and is referred to and utilize two Pedicellus et Pericarpium Trapae mirrors and a slit at charge coupled cell (Charge Coupled Device, CCD) two rectangle images that are positioned at upper and lower relative position of upper generation, when not focusing, upper and lower two rectangle images present the unjustified state staggering, and work as upper and lower two rectangle images, can synthesize a complete aligned condition, represent to have focused.
Yet, the light intensity can bear due to mankind's eyes is not high, therefore, the light intensity that detecting instrument is squeezed in eyes must limit to some extent, and the physical characteristic that is separately subject to optical element affects, the picture screen image that splits producing in traditional detection instrument is not so clear, as shown in Figure 2, detect operator and only can in the pitch-dark block in splitting picture screen, roughly observe out the rectangle image that central authorities have two very fuzzy glimmers, to such an extent as to operator has the difficulty of identification two rectangle image positions, and then cause accurately the complete para-position of rectangle image being focused to reach best.
Because splitting as the operation of focusing of screen mode, the utilization of existing ophthalmology detecting instrument there is inconvenience and the disappearance in many accurate focusings, therefore, tradition is used the real necessity that has improvement innovation of ophthalmology detection method split picture screen, allow operator can be rapidly and clearly identification split picture screen image.
Summary of the invention
Whence is, main purpose of the present invention, aim to provide a kind of auxiliary image recognition method, what traditional ophthalmology detecting instrument was produced splits picture screen image with special optimization method, improve contrast and the sharpness of ophthalmofundoscope image output, allow operator be easy to identify and split the two rectangle images that picture shields, adjust thus the upper and lower relative position of two rectangle images to reach optimized focusing operation.
For on reaching, take off object, the auxiliary image recognition method of the present invention comprises: use detecting instrument to produce one and have the image information that splits picture screen; The picture screen that splits in above-mentioned image information is set an attention region around; The image information in above-mentioned attention region is carried out to pixel intensity and be added processing, all image information luminance contrast in above-mentioned attention region are improved; And contrast correction to improving the attention region of luminance contrast, make to note that splitting picture screen brightness and splitting the contrast of block around of picture screen in region increases.
Wherein, to be made as a detected object be amphiblestroid ophthalmology detecting instrument to detecting instrument of the present invention; And the attention area size of above-mentioned image information is manually to set the manual value of a required scope by operator, also can be selected from the inner predefined default value of detecting instrument; And above-mentioned pixel intensity is added that to process be to adopt each pixel of above-mentioned attention intra-zone to coordinate the mutual superposition mode of brightness of neighbor, all images in above-mentioned attention region are brightened.
In addition, in a preferred embodiment, it is to adopt gamma correction that above-mentioned contrast is proofreaied and correct, so-called gamma correction is that the relatively high pixel of brightness is improved to pixel intensity again, the relatively low pixel of brightness reduces pixel intensity, so can make to note that splitting as screen brightness and the contrast of block around of block inside increases, make to produce the strong and weak operator of confession of the brightness identification that obviously can distinguish between the two.
Hence one can see that, feature of the present invention is the fuzzy picture screen image that splits that ophthalmology detecting instrument is produced, the pixel intensity of facing mutually of carrying out local set point image is added and gamma contrast correction, the contrast and the sharpness that split the interior rectangle block of picture screen and peripheral region are significantly promoted, improve tradition and split the unclear disappearance of image fog in picture screen, allow operator be easy to identify and split two rectangle images in picture screen, adjust thus the upper and lower relative position of two rectangle images to reach optimized focusing operation.
Accompanying drawing explanation
Fig. 1 is the structural representation that traditional detection instrument splits picture screen;
Fig. 2 adopts the 1st figure to split the schematic diagram of picture screen imaging;
Fig. 3 is the auxiliary image identification block diagram of ophthalmology detecting instrument of the present invention;
Fig. 4 is that the schematic diagram that attention region is set in the imaging of picture screen is split in the present invention;
Fig. 5 is that the present invention notices that region carries out the schematic diagram that pixel intensity be added to be processed;
Fig. 6 is the datagram that the present invention adopts Gamma correcting image; And
Fig. 7 is that the present invention adopts Gamma to proofread and correct the schematic diagram that picture screen image brilliance and contrast are split in increase.
The specific embodiment
Hereby, for ease of further structure of the present invention, use and feature thereof being had to clear and definite, the full and accurate understanding of darker one deck and understanding, preferred embodiment is enumerated in whence, coordinates graphic being described in detail as follows:
Refer to shown in the 3rd figure, in a preferred embodiment, the auxiliary image recognition method of the present invention comprises following steps:
(a) use detecting instrument to produce one and there is the image information that splits picture screen;
(b) in the picture screen that splits of above-mentioned image information, set an attention region (Region of interest, ROI) around;
(c) image information in above-mentioned attention region is carried out to pixel intensity and be added processing, all image information luminance contrast in above-mentioned attention region are improved; And
(d) to improving the attention region of luminance contrast, contrast correction, make to note that splitting picture screen brightness and splitting the contrast of block around of picture screen in region increases.
First, please join shown in the 2nd figure, the detecting instrument that step of the present invention (a) is used is that a detected object is amphiblestroid ophthalmology detecting instrument, the light beam that it mainly sends light source, after spectroscope, produce upper and lower twice light and focus on optical fundus, then imaging on optical detector via fundus reflex, making can observe two rectangle images that are positioned at upper and lower different relative positions in detecting instrument, wherein, above-mentioned optical detector can be charge coupled cell (CCD).
Please join shown in the 4th figure, step (b) is application one image processing system, capture in advance one in above-mentioned image information and note region, wherein, above-mentioned attention region must comprise two pictures that split that take shape in diverse location and shield, as shown in the figure, the present invention irises out a rectangular box with predefined default value in image processing system, and containing in rectangular box the shown fuzzy image going out of step a split picture screen image.If but in above-mentioned image processing system, there is no the attention area size that meets demand, the present invention also can be adopted by operator the manual value of manual setting one required scope.
Please join shown in the 5th figure, step (c) is application one image processing system equally, and the pixel intensity that image information is carried out addition processing, it is mainly that the pixel of utilization own adds adjacent pixel brightness value all around, makes to note that all image information luminance contrast in region improve.
For instance, as the brightness of splitting picture screen is 10, and split the brightness of comparatively dim block around of picture screen, be 5, both are 5 in the luminance difference distance of not yet carrying out before image processing, if but each pixel is by itself and its left side, right, on, under neighbor be added, split brightness as screen itself will be promoted to 10 ()+10 (left side)+10 (right side)+10 own (on)+10 (under)=50, and the brightness of dim block around will be promoted to 5 ()+5 (left side)+5 (right side)+5 own (on)+5 (under)=25, now luminance difference is between the two apart from being 25, relatively the 4th figure and the 5th figure can obviously tell, the splitting picture screen image carrying out after pixel intensity addition processing transfers to high-visible, can allow operator obviously tell two and split the relative position that picture shields.
About step (d), be to utilize image processing system to contrast correction, in a preferred embodiment, the present invention contrasts and proofreaies and correct is to adopt gamma correction (Gamma correction), being called again gamma non-linearization (gamma nonlinearity) or gamma coding (gamma encoding), is mainly for the light briliancy (luminance) in image system or tristimulus value (tristimulus values) carries out nonlinear computing or inverse operation.
And gamma correction of the present invention is that the relatively high pixel of brightness is improved to pixel intensity again, the relatively low pixel of brightness reduces pixel intensity, so can make to note that splitting as screen brightness and the contrast of block around of block inside increases, make to produce the strong and weak operator of confession of the brightness identification that obviously can distinguish between the two.
Please join shown in the 6th figure, the brightness that above-mentioned dim block carries out after pixel intensity addition is processed is 25 (transverse axis), after gamma correction, 5 original (longitudinal axis) fall back in the pixel intensity of above-mentioned dim block, and the above-mentioned brightness of splitting after the addition of picture screen pixel intensity is processed is 50 (transverse axis), after gamma correction, can find, the above-mentioned pixel intensity of splitting picture screen is promoted to 70 (longitudinal axis).Please join again shown in the 7th figure, brightness number after gamma correction of the present invention is respectively 70 and 5, now split picture screen with the luminance difference of dim block around apart from 65, as shown in the figure, splitting picture screen image and can produce sharp contrast and sharp keenization with block around in above-mentioned image information, so operator can present when not yet focusing two rectangle images that stagger be adjusted to easily can be synthetic a complete aligned condition, to complete focusing program.
In sum, the auxiliary image recognition method of the present invention produces the fuzzy gloomy picture that splits by ophthalmology detecting instrument and shields image, the pixel intensity of facing mutually of carrying out local set point image is added and dynamic contrast correction (Gamma Correction), the contrast and the sharpness that split the interior rectangle block of picture screen and peripheral region are significantly promoted, improve tradition and split the disappearance that is difficult for identification as the unclear operator of making of image fog in screen, the present invention can allow operator identify at an easy rate to split two rectangle images in picture screen, adjust thus the upper of two rectangle images, lower relative position is to reach optimized focusing operation.
Above illustrated embodiment, only with the present invention for convenience of description, be not in a limitative way, not from the present invention spirit category, various Simple transformeds and modification that those of ordinary skills do according to the present patent application the scope of the claims and invention description, all must include in following claim.
Claims (7)
1. an auxiliary image recognition method, comprises following steps:
Use detecting instrument to produce one and there is the image information that splits picture screen;
The picture screen that splits in above-mentioned image information is set an attention region around;
The image information in above-mentioned attention region is carried out to pixel intensity and be added processing, all image information luminance contrast in above-mentioned attention region are improved; And
To improving the attention region of luminance contrast, contrast correction, make to note that splitting picture screen brightness and splitting the contrast of block around of picture screen in region increases.
2. auxiliary image recognition method as claimed in claim 1, wherein, above-mentioned detecting instrument is made as an ophthalmology detecting instrument.
3. auxiliary image recognition method as claimed in claim 2, wherein, the detected object of above-mentioned ophthalmology detecting instrument is retina.
4. auxiliary image recognition method as claimed in claim 1, wherein, above-mentioned attention area size is manually to set the manual value of a required scope by operator.
5. auxiliary image recognition method as claimed in claim 1, wherein, above-mentioned attention area size is to be selected from the inner predefined default value of detecting instrument.
6. auxiliary image recognition method as claimed in claim 1, wherein, above-mentioned pixel intensity is added and processes is to notice that all image informations in region adopt the mutual superposition of brightness of neighbors.
7. auxiliary image recognition method as claimed in claim 1, wherein, it is to adopt gamma correction that above-mentioned contrast is proofreaied and correct.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210281864.6A CN103565400B (en) | 2012-08-09 | 2012-08-09 | Auxiliary image recognition method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210281864.6A CN103565400B (en) | 2012-08-09 | 2012-08-09 | Auxiliary image recognition method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103565400A true CN103565400A (en) | 2014-02-12 |
| CN103565400B CN103565400B (en) | 2015-12-09 |
Family
ID=50038742
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201210281864.6A Active CN103565400B (en) | 2012-08-09 | 2012-08-09 | Auxiliary image recognition method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103565400B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1988603A (en) * | 2006-12-30 | 2007-06-27 | 华中科技大学 | Computer auxiliary medical image observing system |
| US20080117384A1 (en) * | 2004-09-15 | 2008-05-22 | Satoru Inakagata | Perimeter |
| CN101520888A (en) * | 2008-02-27 | 2009-09-02 | 中国科学院自动化研究所 | Method for enhancing blood vessels in retinal images based on the directional field |
| US20100149489A1 (en) * | 2007-05-23 | 2010-06-17 | Kabushiki Kaisha Topcon | Fundus oculi observation device and program for controlling the same |
| CN102245081A (en) * | 2008-12-12 | 2011-11-16 | 卡尔蔡司医疗技术股份公司 | High precision contrast ratio display for visual stimulus |
-
2012
- 2012-08-09 CN CN201210281864.6A patent/CN103565400B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080117384A1 (en) * | 2004-09-15 | 2008-05-22 | Satoru Inakagata | Perimeter |
| CN1988603A (en) * | 2006-12-30 | 2007-06-27 | 华中科技大学 | Computer auxiliary medical image observing system |
| US20100149489A1 (en) * | 2007-05-23 | 2010-06-17 | Kabushiki Kaisha Topcon | Fundus oculi observation device and program for controlling the same |
| CN101520888A (en) * | 2008-02-27 | 2009-09-02 | 中国科学院自动化研究所 | Method for enhancing blood vessels in retinal images based on the directional field |
| CN102245081A (en) * | 2008-12-12 | 2011-11-16 | 卡尔蔡司医疗技术股份公司 | High precision contrast ratio display for visual stimulus |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103565400B (en) | 2015-12-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US9224193B2 (en) | Focus stacking image processing apparatus, imaging system, and image processing system | |
| EP3496383A1 (en) | Image processing method, apparatus and device | |
| DE69906403T2 (en) | Method and device for detecting a face-like area | |
| CN105578019B (en) | Image extraction system capable of obtaining depth information and focusing method | |
| US10694101B2 (en) | Contrast-enhanced combined image generation systems and methods | |
| CN105147311B (en) | For the visualization device sub-scanning localization method and system in CT system | |
| US8610803B2 (en) | Image processing apparatus for acquiring an image using a specific image capturing method and acquiring an anomalous pixel using a different image capturing method | |
| US9332247B2 (en) | Image processing device, non-transitory computer readable recording medium, and image processing method | |
| CN104853184B (en) | Detection method for parallax scope influencing comfort level of stereo image | |
| DE102020211408A1 (en) | Welding information providing device | |
| WO2017084391A1 (en) | Method and system for performing fluorescence processing on an image of a projector | |
| EP2635019A2 (en) | Image processing device, image processing method, and program | |
| CN110879131B (en) | Imaging quality testing method and imaging quality testing device for visual optical system, and electronic apparatus | |
| CN108037142A (en) | Mask plate optical defect detection method based on image intensity value | |
| CN105205300A (en) | Automatic adjustment examination paper go-over method and device | |
| CN104000593A (en) | Color card and skin test method and system | |
| JP5362405B2 (en) | Pupil measuring device, pupil measuring method and pupil measuring program | |
| CN106709888B (en) | A kind of high dynamic range images production method based on human vision model | |
| KR20140051082A (en) | Image processing device using difference camera | |
| CN101980299B (en) | Chessboard calibration-based camera mapping method | |
| CN106412436A (en) | Focusing method | |
| CN103565400A (en) | Auxiliary image recognition method | |
| JP2017023474A (en) | Skin pore analysis apparatus, skin pore analysis method and program for use in device | |
| CN111766049B (en) | AR lens testing method, device, equipment and computer readable storage medium | |
| JP2018163106A (en) | Lens meter |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant |