CN107578420A - A kind of adaptive striation carrying out image threshold segmentation method - Google Patents
A kind of adaptive striation carrying out image threshold segmentation method Download PDFInfo
- Publication number
- CN107578420A CN107578420A CN201710715098.2A CN201710715098A CN107578420A CN 107578420 A CN107578420 A CN 107578420A CN 201710715098 A CN201710715098 A CN 201710715098A CN 107578420 A CN107578420 A CN 107578420A
- Authority
- CN
- China
- Prior art keywords
- mrow
- striation
- image
- section
- optical strip
- 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
Landscapes
- Image Analysis (AREA)
Abstract
A kind of adaptive striation carrying out image threshold segmentation method of the present invention belongs to binocular vision technology field, is related to a kind of adaptive striation carrying out image threshold segmentation method.This method splits initial striation region by traditional fixed threshold image partition method, obtains the row coordinate of striation cross section right boundary;Then gradation of image Evaluation on distribution coefficient is established, according to initial threshold segmentation result, calculates the striation cross section energy intensity of often row striation cross section;According to striation distribution characteristics, the gray scale branch for calculating desired light sliver transvers section energy intensity is horizontal;Resettle with the positively related optical strip image adaptive threshold fuzziness correlation model of optical strip image intensity profile coefficient, to determine the self-adaptive projection method threshold value of optical strip image, striation region is precisely separating out from background.The extraction accuracy of random surface large aerospace component surface striation is the method increase, avoid local overexposure or local striation secretly causes the problem of striation extraction is difficult, and striation extraction accuracy is not high excessively.
Description
Technical field
The invention belongs to binocular vision technology field, is related to a kind of adaptive striation carrying out image threshold segmentation method.
Background technology
During vision measurement, accurate Light stripes center extraction is the key for realizing high precision three-dimensional measurement.However, it is directed to
Large aerospace component is the situation of measurement object, and because its surface is usually free form surface, the striation of projection is through workpiece for measurement table
It is free curve striation that face, which is modulated at distortion in the image of shooting,;And live photoenvironment is complicated, exist it is local high reflective and
The uneven influence of global illumination, cause the optical strip image intensity profile of shooting seriously uneven, while regional area striation be present
Overexposure and the excessively dark phenomenon of local area light bar, so as to have a strong impact on the integrality and precision of striation extraction.In current vision
In measurement process, unique binary-state threshold generally is chosen to shooting gained image, striation contours extract is carried out, at the scene under environment
Easily there is the problem of local fracture and wide local overexposure striation in the striation of extraction.
By literature search, Long Jianwu Master's thesis《Carrying out image threshold segmentation key technology research》, 2014, Jilin was big
Learn, in this paper, three-dimensional popularization is carried out to two-dimentional minimum error method, and combine stereogram reconstruction and carried with dimensionality reduction thought
A kind of Minimum error threshold algorithm of robust is gone out.This method efficiently solves Small object image under the conditions of inhomogeneous illumination
Segmentation problem.But for the situation of large aerospace component, because part and visual field are bigger, party's rule can compare nothing
Power.
The content of the invention
The present invention is directed to the problem of large aerospace component surface optical strip image intensity profile is uneven, has invented a kind of adaptive
Answer optical strip image threshold segmentation method.This method splits initial striation area by traditional fixed threshold image partition method
Domain, obtain the row coordinate of striation region right boundary;Then gradation of image Evaluation on distribution coefficient is established, is split according to initial threshold
As a result, striation cross section average intensity level is defined as to the cross section energy intensity of striation;It is adaptive to resettle optical strip image
Threshold segmentation correlation model, to determine the self-adaptive projection method threshold value of optical strip image, striation area is precisely separating out from background
Domain, the extraction accuracy of random surface large aerospace component surface striation is substantially increased, is avoided due to local overexposure or office
Portion's striation is excessively dark and causes the problem of striation extraction is difficult, and striation extraction accuracy is not high.
The technical solution adopted by the present invention is a kind of adaptive striation carrying out image threshold segmentation method, it is characterized in that, this method
Split initial striation region by traditional fixed threshold image partition method, the row for obtaining striation cross section right boundary are sat
Mark;Then gradation of image Evaluation on distribution coefficient is established, according to initial threshold segmentation result, by striation cross section average intensity level
It is defined as the cross section energy intensity of striation;Calculate the striation cross section energy intensity of often row striation cross section;According to striation point
Cloth feature, the gray scale branch for calculating desired light sliver transvers section energy intensity are horizontal;Obtain optical strip image intensity profile coefficient;Build again
The vertical and positively related optical strip image adaptive threshold fuzziness correlation model of optical strip image intensity profile coefficient, to determine optical strip image
Self-adaptive projection method threshold value, striation region is precisely separating out from background.Method comprises the following steps that:
The first step splits initial striation region
According to line laser fringe gray level Characteristics of Distribution, the intensity profile uniformity of optical strip image is by striation cross section ash
Degree is horizontal and is determined along the shade of gray change on striation direction is comprehensive;Traditional fixed threshold image partition method is used first
Optical strip image is handled, if the original optical strip image of input is f, output image g, f (u, v) represent that input picture is being schemed
As the gray value at pixel (u, v) place, g (u, v) represents gray value of the output image at image pixel (u, v) place, then traditional two
Value segmentation can represent as follows:
In formula, T is image segmentation threshold;
Then left and right boundary coordinate p (v) and the q (v) in striation region are calculated according to bianry image;P (v), q (v) points
Not Biao Shi in image v row striations cross section left and right border column coordinate;
Second step establishes gradation of image Evaluation on distribution coefficient
It is horizontal to calculate actual optical strip image section intensity profile;According to initial threshold segmentation result, striation cross section is put down
Equal grey level is defined as the cross section energy intensity of striation, is described with formula (2):
Wherein, EICS (v) is the striation section energy intensity of v rows in optical strip image;Striation section energy intensity EICS
The main intensity profile for characterizing striation is horizontal;EICS numerical value is bigger, then shows that the sectional position striation brightness is bigger, it is easier from
Separated in background;On the contrary, EICS numerical value is smaller, then the section go out striation may be excessively dark, cause in overall extraction process
Striation characteristic information may be lost here;
Then, define and calculate preferable striation image cross section intensity profile level;In theory, according to the life of line laser striped
Into mechanism, its striation cross section intensity profile Gaussian distributed model of the preferable optical strip image of video camera shooting acquisition, therefore
Image peak gray value saturation, obedience ideal Gaussian distribution striation section average intensity level is defined as into preferable striation to cut
Face energy intensity, is expressed as:
Wherein, iEICS (v) is the striation section energy intensity of v rows in ideal situation hypograph;W (v) is actual striation
The striation width of v rows in image, w (v)=q (v)-p (v);A is desired light sliver transvers section gray scale peak value, and defining its peak value is
Image saturation gray value, therefore A=255;In addition, 3 σ based on Gaussian Profile are theoretical, i.e., striation energy 99.74% concentrates on height
In the range of ± 3 σ of this distribution average, therefore, according to striation overwhelming majority energy has been concentrated in striation width range, w (v) is defined
=6 σw, calculate the standard deviation sigma that standard gaussian corresponding to often being gone in image is distributedw;
The energy intensity EICS (v) in actual striation section and the ratio of preferable striation section energy intensity iEICS (v) are made
For optical strip image intensity profile coefficient ηCGDLS:
3rd step optical strip image adaptive threshold fuzziness correlation model
Utilize the optical strip image intensity profile coefficient η calculatedCGDLS, establish and extracted with it into positively related adaptive threshold
Mathematical modeling:
λ (v)=f (ηCGDLS(v)) in (5) formula, λ (v) is the figure of the v rows cross section grey level of adaptive optical strip image
As segmentation threshold;
Using linear regression method, the self-adaptive projection method threshold value for determining optical strip image is
Wherein, threshupAnd threshdownCorresponding optical strip image intensity profile coefficient maximum is represented respectivelyAnd minimum
ValueThe optical strip image segmentation threshold at place, so that striation region can be precisely separating out from background as standard, pass through priori
Knowledge Acquirement;So, the striation binary-state threshold that often row should use will pass through adaptive mode and choose completion.
The beneficial effects of the invention are as follows by establishing gradation of image Evaluation on distribution coefficient, according to initial threshold segmentation result,
Calculate the striation cross section energy intensity of often row striation cross section;According to striation distribution characteristics, desired light sliver transvers section energy is calculated
The gray scale branch for measuring intensity is horizontal.The extraction accuracy of large aerospace component surface striation is substantially increased, is avoided due to local mistake
Expose or local striation secretly causes the problem of striation extraction is difficult, and striation extraction accuracy is not high excessively.
Brief description of the drawings
Fig. 1 is method flow diagram.
Fig. 2 is optical strip image along striation direction intensity profile index variation profiles.Wherein, abscissa represents v pixel columns,
Ordinate represents the intensity profile coefficient of the pixel column;2- overexposures region, 3,4- cross dark areas.
Fig. 3 is the striation section energy intensity situation of change schematic diagram along striation direction.Abscissa represents v pixel columns,
Ordinate represents the striation section energy intensity value of the pixel column.
Fig. 4 is local overexposure striation schematic diagram.Wherein, 1- parts overexposure region.
Fig. 5 is the result figure after adaptive threshold fuzziness.
Embodiment
Describe the embodiment of the present invention in detail with technical scheme below in conjunction with the accompanying drawings.
In the present embodiment, testee is t800 composite panels, and wavelength 460nm royal purple line lasers are projected into multiple material plate
On.
The present invention is using the video camera shooting optical strip image for configuring wide-angle lens.Video camera model view works VC-
The video cameras of 12MC-M/C 65, resolution ratio:4096 × 3072, imaging sensor:CMOS, frame per second:Silent frame, highest 64.3fps,
Weight:420g.Wide-angle lens model EF 16-35mm f/2.8L II USM, parameter is as follows, lens focus:F=16-
35mm, APS focal length:25.5-52.5 aperture:F2.8, Lens:82×106.Shooting condition is as follows:Picture pixels are 4096
× 3072, lens focus 25mm, object distance 750mm, visual field are about 850mm × 450mm.
The flow chart of method is as shown in figure 1, comprise the following steps that:
The first step, split initial striation region.According to line laser fringe gray level Characteristics of Distribution, the gray scale of optical strip image
Distributing homogeneity determines by striation cross section grey level and along the shade of gray change synthesis on striation direction.Therefore, first
Optical strip image is handled using traditional fixed threshold image partition method, if the original optical strip image of input is f, output
Image is g, and f (u, v) represents gray value of the input picture at image pixel (u, v) place, and g (u, v) represents output image in image
The gray value at pixel (u, v) place, then traditional binary segmentation can be expressed as shown in formula (1).
Then left and right boundary coordinate p (v) and the q (v) in striation region are calculated according to bianry image;P (v), q (v) points
Not Biao Shi in image v row striations cross section left and right border column coordinate.
Second step, establish gradation of image Evaluation on distribution coefficient.It is horizontal to calculate actual optical strip image section intensity profile;According to
Initial threshold segmentation result, striation cross section average intensity level is defined as to the cross section energy intensity of striation, (2) can be used
Formula is described.
Wherein, EICS (v) is the striation section energy intensity of v rows in optical strip image.Striation section energy intensity EICS
The main intensity profile for characterizing striation is horizontal.EICS numerical value is bigger, then shows that the sectional position striation brightness is bigger, it is easier from
Separated in background;On the contrary, EICS numerical value is smaller, then the section go out striation may be excessively dark, as shown in Fig. 2 causing in entirety
Striation characteristic information may be lost in extraction process here.
Preferable striation image cross section intensity profile is horizontal, and in theory, according to the formation mechanism of line laser striped, video camera is clapped
Take the photograph the preferable optical strip image of acquisition its striation cross section intensity profile Gaussian distributed model, therefore by image peak gray value
Saturation, obedience ideal Gaussian distribution striation section average intensity level is defined as preferable striation section energy intensity, states
For formula (3).Wherein, in iEICS (v) ideal situations hypograph v rows striation section energy intensity;W (v) is actual striation
The striation width of v rows in image, w (v)=q (v)-p (v);A is desired light sliver transvers section gray scale peak value, and defining its peak value is
Image saturation gray value, therefore A=255;In addition, 3 σ based on Gaussian Profile are theoretical, i.e., striation energy 99.74% concentrates on height
In the range of ± 3 σ of this distribution average, therefore, according to striation overwhelming majority energy has been concentrated in striation width range, w (v) is defined
=6 σw, it is possible thereby to calculate in image often go corresponding to standard gaussian distribution standard deviation sigmaw。
The energy intensity EICS (v) in actual striation section and the ratio of preferable striation section energy intensity iEICS (v) are made
For optical strip image intensity profile coefficient ηCGDLS, calculated and obtained using formula (4), as shown in Figure 3.
3rd step, establish optical strip image adaptive threshold fuzziness correlation model.Utilize the optical strip image gray scale point calculated
Cloth coefficient ηCGDLS, optical strip image intensity profile coefficient ηCGDLSVariation tendency result.
Established using formula (5) and extract mathematical modeling into positively related adaptive threshold with it.Using linear regression method,
By formula (6), the self-adaptive projection method threshold value of optical strip image is determined.Wherein, threshupAnd threshdownRepresent respectively
Corresponding optical strip image intensity profile coefficient maximumAnd minimum valueThe optical strip image segmentation threshold at place, with can
It is standard that striation region is precisely separating out from background, is obtained by priori.So, the striation binaryzation that often row should use
Threshold value will pass through adaptive mode and choose completion, can obtain uniform optical strip image as shown in Figure 5, avoid shown in Fig. 4
Local overexposure optical strip image.
This method improves the extraction accuracy of random surface large aerospace component surface striation, avoid local overexposure or
Local striation is excessively dark and causes the problem of striation extraction is difficult, and striation extraction accuracy is not high.
Claims (1)
1. a kind of adaptive striation carrying out image threshold segmentation method, it is characterized in that, this method passes through traditional fixed threshold image point
Segmentation method splits initial striation region, obtains the row coordinate of striation cross section right boundary;Then gradation of image distribution is established
Evaluation coefficient, according to initial threshold segmentation result, striation cross section average intensity level is defined as to the cross section energy of striation
Intensity;Calculate the striation cross section energy intensity of often row striation cross section;According to striation distribution characteristics, it is transversal to calculate preferable striation
The gray scale branch of face energy intensity is horizontal;Obtain optical strip image intensity profile coefficient;Resettle and optical strip image intensity profile system
The positively related optical strip image adaptive threshold fuzziness correlation model of number, to determine the self-adaptive projection method threshold value of optical strip image,
Striation region is precisely separating out from background;Method comprises the following steps that:
The first step splits initial striation region
According to line laser fringe gray level Characteristics of Distribution, the intensity profile uniformity of optical strip image is by striation cross section gray scale water
It is gentle to be determined along the shade of gray change on striation direction is comprehensive;
Optical strip image is handled using traditional fixed threshold image partition method first, if the original optical strip image of input
Gray value of the input picture at image pixel (u, v) place is represented for f, output image g, f (u, v), g (u, v) represents output figure
As image pixel (u, v) place gray value, then traditional binary segmentation can represent as follows:
<mrow>
<mi>g</mi>
<mrow>
<mo>(</mo>
<mi>u</mi>
<mo>,</mo>
<mi>v</mi>
<mo>)</mo>
</mrow>
<mo>=</mo>
<mfenced open = "{" close = "">
<mtable>
<mtr>
<mtd>
<mrow>
<mn>0</mn>
<mo>,</mo>
</mrow>
</mtd>
<mtd>
<mrow>
<mi>f</mi>
<mrow>
<mo>(</mo>
<mi>u</mi>
<mo>,</mo>
<mi>v</mi>
<mo>)</mo>
</mrow>
<mo>&le;</mo>
<mi>T</mi>
</mrow>
</mtd>
</mtr>
<mtr>
<mtd>
<mrow>
<mn>1</mn>
<mo>,</mo>
</mrow>
</mtd>
<mtd>
<mrow>
<mi>f</mi>
<mrow>
<mo>(</mo>
<mi>u</mi>
<mo>,</mo>
<mi>v</mi>
<mo>)</mo>
</mrow>
<mo>></mo>
<mi>T</mi>
</mrow>
</mtd>
</mtr>
</mtable>
</mfenced>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>1</mn>
<mo>)</mo>
</mrow>
</mrow>
In formula, T is image segmentation threshold;
Then left and right boundary coordinate p (v) and the q (v) in striation region are calculated according to bianry image;P (v), q (v) difference table
The left and right border column coordinate (in theory, q (v) > p (v)) of v row striations cross section in diagram picture;
Second step establishes gradation of image Evaluation on distribution coefficient
It is horizontal to calculate actual optical strip image section intensity profile;According to initial threshold segmentation result, by the average ash in striation cross section
Degree level is defined as the cross section energy intensity of striation, is described with formula (2):
<mrow>
<mi>E</mi>
<mi>I</mi>
<mi>C</mi>
<mi>S</mi>
<mrow>
<mo>(</mo>
<mi>v</mi>
<mo>)</mo>
</mrow>
<mo>=</mo>
<mfrac>
<mrow>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>u</mi>
<mo>=</mo>
<mi>p</mi>
<mrow>
<mo>(</mo>
<mi>v</mi>
<mo>)</mo>
</mrow>
</mrow>
<mrow>
<mi>q</mi>
<mrow>
<mo>(</mo>
<mi>v</mi>
<mo>)</mo>
</mrow>
</mrow>
</munderover>
<mi>f</mi>
<mrow>
<mo>(</mo>
<mi>u</mi>
<mo>,</mo>
<mi>v</mi>
<mo>)</mo>
</mrow>
</mrow>
<mrow>
<mi>q</mi>
<mrow>
<mo>(</mo>
<mi>v</mi>
<mo>)</mo>
</mrow>
<mo>-</mo>
<mi>p</mi>
<mrow>
<mo>(</mo>
<mi>v</mi>
<mo>)</mo>
</mrow>
</mrow>
</mfrac>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>2</mn>
<mo>)</mo>
</mrow>
</mrow>
Wherein, EICS (v) is the striation section energy intensity of v rows in optical strip image;
Then, define and calculate preferable striation image cross section intensity profile level;In theory, according to the generation machine of line laser striped
Reason, its striation cross section intensity profile Gaussian distributed model of the preferable optical strip image of video camera shooting acquisition, therefore will figure
As peak gray value saturation, obedience ideal Gaussian distribution striation section average intensity level is defined as preferable striation section energy
Intensity is measured, is expressed as:
<mrow>
<mi>i</mi>
<mi>E</mi>
<mi>I</mi>
<mi>C</mi>
<mi>S</mi>
<mrow>
<mo>(</mo>
<mi>v</mi>
<mo>)</mo>
</mrow>
<mo>=</mo>
<mfrac>
<mrow>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>k</mi>
<mo>=</mo>
<mo>-</mo>
<mi>w</mi>
<mrow>
<mo>(</mo>
<mi>v</mi>
<mo>)</mo>
</mrow>
<mo>/</mo>
<mn>2</mn>
</mrow>
<mrow>
<mi>w</mi>
<mrow>
<mo>(</mo>
<mi>v</mi>
<mo>)</mo>
</mrow>
<mo>/</mo>
<mn>2</mn>
</mrow>
</munderover>
<mi>A</mi>
<mo>&CenterDot;</mo>
<msup>
<mi>e</mi>
<mrow>
<mo>-</mo>
<mfrac>
<msup>
<mi>k</mi>
<mn>2</mn>
</msup>
<mrow>
<mn>2</mn>
<msubsup>
<mi>&sigma;</mi>
<mi>w</mi>
<mn>2</mn>
</msubsup>
</mrow>
</mfrac>
</mrow>
</msup>
</mrow>
<mrow>
<mi>w</mi>
<mrow>
<mo>(</mo>
<mi>v</mi>
<mo>)</mo>
</mrow>
</mrow>
</mfrac>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>3</mn>
<mo>)</mo>
</mrow>
</mrow>
Wherein, in iEICS (v) ideal situations hypograph v rows striation section energy intensity;W (v) is in actual optical strip image
The striation width of v rows, w (v)=q (v)-p (v);A is desired light sliver transvers section gray scale peak value, and its peak value is image in definition
Saturation gray value, therefore A=255;In addition, 3 σ based on Gaussian Profile are theoretical, i.e., striation energy 99.74% concentrates on Gauss point
In the range of ± 3 σ of cloth average, therefore, according to striation overwhelming majority energy has been concentrated in striation width range, w (v)=6 is defined
σw, it is possible thereby to calculate in image often go corresponding to standard gaussian distribution standard deviation sigmaw;
Using the energy intensity EICS (v) in actual striation section and the ratio of preferable striation section energy intensity iEICS (v) as light
Bar gradation of image breadth coefficient ηCGDLS:
<mrow>
<msub>
<mi>&eta;</mi>
<mrow>
<mi>C</mi>
<mi>G</mi>
<mi>D</mi>
<mi>L</mi>
<mi>S</mi>
</mrow>
</msub>
<mrow>
<mo>(</mo>
<mi>v</mi>
<mo>)</mo>
</mrow>
<mo>=</mo>
<mfrac>
<mrow>
<mi>E</mi>
<mi>I</mi>
<mi>C</mi>
<mi>S</mi>
<mrow>
<mo>(</mo>
<mi>v</mi>
<mo>)</mo>
</mrow>
</mrow>
<mrow>
<mi>i</mi>
<mi>E</mi>
<mi>I</mi>
<mi>C</mi>
<mi>S</mi>
<mrow>
<mo>(</mo>
<mi>v</mi>
<mo>)</mo>
</mrow>
</mrow>
</mfrac>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>4</mn>
<mo>)</mo>
</mrow>
</mrow>
3rd step optical strip image adaptive threshold fuzziness correlation model
Utilize the optical strip image intensity profile coefficient η calculatedCGDLS, establish and extract mathematics into positively related adaptive threshold with it
Model:
λ (v)=f (ηCGDLS(v)) (5)
In formula, λ (v) is the image segmentation threshold of the v rows cross section grey level of adaptive optical strip image;
Using linear regression method, the self-adaptive projection method threshold value for determining optical strip image is
Wherein, threshupAnd threshdownCorresponding optical strip image intensity profile coefficient maximum is represented respectivelyAnd minimum
ValueThe optical strip image segmentation threshold at place;So, the striation binary-state threshold that often row should use, it will pass through adaptive side
Formula, which is chosen, to be completed.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710715098.2A CN107578420B (en) | 2017-08-21 | 2017-08-21 | A kind of adaptive striation carrying out image threshold segmentation method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710715098.2A CN107578420B (en) | 2017-08-21 | 2017-08-21 | A kind of adaptive striation carrying out image threshold segmentation method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN107578420A true CN107578420A (en) | 2018-01-12 |
CN107578420B CN107578420B (en) | 2019-11-19 |
Family
ID=61034593
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710715098.2A Expired - Fee Related CN107578420B (en) | 2017-08-21 | 2017-08-21 | A kind of adaptive striation carrying out image threshold segmentation method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107578420B (en) |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108550144A (en) * | 2018-04-09 | 2018-09-18 | 大连理工大学 | Laser striation sequence image quality evaluating method based on gray scale reliability |
CN108629790A (en) * | 2018-04-26 | 2018-10-09 | 大连理工大学 | A kind of optical strip image threshold segmentation method based on depth residual error network |
CN110210437A (en) * | 2019-06-10 | 2019-09-06 | 上海联影医疗科技有限公司 | The determination method and system of human region in a kind of image |
CN110232709A (en) * | 2019-04-19 | 2019-09-13 | 武汉大学 | A kind of line-structured light Light stripes center extraction method becoming Threshold segmentation |
CN112710250A (en) * | 2020-11-23 | 2021-04-27 | 武汉光谷卓越科技股份有限公司 | Three-dimensional measurement method based on line structured light and sensor |
CN117437247A (en) * | 2023-12-18 | 2024-01-23 | 津泰(天津)医疗器械有限公司 | Lesion region extraction and segmentation method based on natural cavity image |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20040001195A1 (en) * | 2002-06-28 | 2004-01-01 | Fuji Photo Optical Co., Ltd. | Method of extracting circular region from fringe image |
CN104616325A (en) * | 2015-01-21 | 2015-05-13 | 大连理工大学 | Rapid and high-precision method for extracting light strip center on large surface |
CN105300316A (en) * | 2015-09-22 | 2016-02-03 | 大连理工大学 | Light stripe center rapid extraction method based on gray centroid method |
CN105335988A (en) * | 2015-09-25 | 2016-02-17 | 大连理工大学 | Hierarchical processing based sub-pixel center extraction method |
-
2017
- 2017-08-21 CN CN201710715098.2A patent/CN107578420B/en not_active Expired - Fee Related
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20040001195A1 (en) * | 2002-06-28 | 2004-01-01 | Fuji Photo Optical Co., Ltd. | Method of extracting circular region from fringe image |
CN104616325A (en) * | 2015-01-21 | 2015-05-13 | 大连理工大学 | Rapid and high-precision method for extracting light strip center on large surface |
CN105300316A (en) * | 2015-09-22 | 2016-02-03 | 大连理工大学 | Light stripe center rapid extraction method based on gray centroid method |
CN105335988A (en) * | 2015-09-25 | 2016-02-17 | 大连理工大学 | Hierarchical processing based sub-pixel center extraction method |
Cited By (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108550144A (en) * | 2018-04-09 | 2018-09-18 | 大连理工大学 | Laser striation sequence image quality evaluating method based on gray scale reliability |
CN108550144B (en) * | 2018-04-09 | 2020-04-07 | 大连理工大学 | Laser light bar sequence image quality evaluation method based on gray scale reliability |
CN108629790A (en) * | 2018-04-26 | 2018-10-09 | 大连理工大学 | A kind of optical strip image threshold segmentation method based on depth residual error network |
CN108629790B (en) * | 2018-04-26 | 2020-08-14 | 大连理工大学 | Light bar image threshold segmentation method based on depth residual error network |
CN110232709A (en) * | 2019-04-19 | 2019-09-13 | 武汉大学 | A kind of line-structured light Light stripes center extraction method becoming Threshold segmentation |
CN110232709B (en) * | 2019-04-19 | 2022-07-29 | 武汉大学 | Method for extracting line structured light strip center by variable threshold segmentation |
CN110210437A (en) * | 2019-06-10 | 2019-09-06 | 上海联影医疗科技有限公司 | The determination method and system of human region in a kind of image |
CN112710250A (en) * | 2020-11-23 | 2021-04-27 | 武汉光谷卓越科技股份有限公司 | Three-dimensional measurement method based on line structured light and sensor |
CN117437247A (en) * | 2023-12-18 | 2024-01-23 | 津泰(天津)医疗器械有限公司 | Lesion region extraction and segmentation method based on natural cavity image |
CN117437247B (en) * | 2023-12-18 | 2024-03-05 | 津泰(天津)医疗器械有限公司 | Lesion region extraction and segmentation method based on natural cavity image |
Also Published As
Publication number | Publication date |
---|---|
CN107578420B (en) | 2019-11-19 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107578420A (en) | A kind of adaptive striation carrying out image threshold segmentation method | |
CN104680496B (en) | A kind of Kinect depth map restorative procedures based on color images | |
US20180091798A1 (en) | System and Method for Generating a Depth Map Using Differential Patterns | |
CN103530880A (en) | Camera calibration method based on projected Gaussian grid pattern | |
CN108613637B (en) | Structured light system dephasing method and system based on reference image | |
CN110009693B (en) | Rapid blind calibration method of light field camera | |
JP5633058B1 (en) | 3D measuring apparatus and 3D measuring method | |
CN105100771A (en) | Single-viewpoint video depth obtaining method based on scene classification and geometric dimension | |
CN111563952B (en) | Method and system for realizing stereo matching based on phase information and spatial texture characteristics | |
CN105335968A (en) | Depth map extraction method based on confidence coefficient propagation algorithm and device | |
CN108550160A (en) | Non-homogeneous striation characteristic area extracting method based on light intensity template | |
CN101900536A (en) | Method for measuring object surface appearance based on digital picture method | |
CN103993548A (en) | Multi-camera stereoscopic shooting based pavement damage crack detection system and method | |
CN104111038A (en) | Method for using phase fusion algorithm to repair phase error caused by saturation | |
CN108051183B (en) | Focus type light-field camera parameter calibration method based on first-order theory | |
CN107680152A (en) | Target surface topography measurement method and apparatus based on image procossing | |
JP6009206B2 (en) | 3D measuring device | |
CN108010075A (en) | A kind of sectional perspective matching process based on multiple features combining | |
CN103925889A (en) | Method for fast recovering surface phase of high-light object based on least square method | |
JP2006058091A (en) | Three-dimensional image measuring device and method | |
CN106447718A (en) | 2D-to-3D depth estimation method | |
CN109003308A (en) | A kind of special areas imaging camera calibration system and method based on phase code | |
CN111951339A (en) | Image processing method for performing parallax calculation by using heterogeneous binocular cameras | |
CN115330684A (en) | Underwater structure apparent defect detection method based on binocular vision and line structured light | |
CN109218706B (en) | Method for generating stereoscopic vision image from single 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 | ||
GR01 | Patent grant | ||
GR01 | Patent grant | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20191119 Termination date: 20210821 |
|
CF01 | Termination of patent right due to non-payment of annual fee |