CN103033134A - Detection system and detection method aiming at damaged optical fiber geometry parameters - Google Patents

Abstract

The invention discloses a detection system and a detection method aiming at damaged optical fiber geometry parameters. The detection method comprises the following steps: reading an optical fiber image; threshold value selection step: statistically calculating a grey level histogram and selecting an optimal threshold value; image binarization step: carrying out binarization on the image; edge point selection step: accurately selecting an outer edge point and an inner edge point in the binary image; ellipse fitting and parameter obtaining step: obtaining parameters of an ellipse of an outer edge through the selected outer edge point and fitting parameters of an ellipse of an inner edge through the inner edge point; and obtaining the geometry parameters of an optical fiber picture after the ellipses of the inner edge and the outer edge are fitted. The detection system and the detection method aiming at the damaged optical fiber geometry parameters disclosed by the invention have the beneficial effects that by using a novel method for extracting a coordinate point, the points which are not on the ellipses can be rejected; and further, the ellipse fitting precision is improved.

Description

Detection system and method for damaged optical fiber geometric parameter

Technical field

The invention belongs to the detection technique field, relate to a kind of detection system, relate in particular to a kind of detection system for damaged optical fiber geometric parameter; Simultaneously, the invention still further relates to a kind of detection method for damaged optical fiber geometric parameter.

Background technology

Usually qualified fibre cladding and core diameter can be thought two centers in same point, and close to round ellipse. traditional method detects the geometric parameter that can detect very accurately qualified optical fiber.But certain breakage is arranged when optical fiber, perhaps during deformation, flaw will appear in traditional method.

In view of this, nowadays in the urgent need to designing a kind of new detection method, in order to further improve the precision of ellipse fitting.

Summary of the invention

Technical matters to be solved by this invention is: a kind of detection system for damaged optical fiber geometric parameter is provided, can rejects the not point on ellipse, further improve the precision of ellipse fitting.

In addition, the present invention also provides a kind of detection method for damaged optical fiber geometric parameter, can reject the not point on ellipse, further improves the precision of ellipse fitting.

For solving the problems of the technologies described above, the present invention adopts following technical scheme:

A kind of detection system for damaged optical fiber geometric parameter, described system comprises:

Image fetching unit is in order to read optical fiber image;

Threshold value is chosen the unit, in order to add up grey level histogram, chooses optimal threshold;

The image binaryzation processing unit is in order to carry out binary conversion treatment to image;

The marginal point selected cell is in order to accurate outward flange and the inward flange point selected in binary image;

Ellipse fitting and parameter acquiring unit are obtained outer peripheral elliptic parameter in order to the outward flange point of choosing by all, simulate the parameter of the ellipse of inward flange by the inward flange point; Obtain the geometric parameter of this optical fiber picture behind the ellipse by the match outer edge.

As a preferred embodiment of the present invention, described threshold value is chosen the threshold value of unit selection and is the gray-scale value of trough between first crest and the secondary peak; Wherein, the gray-scale value of first crest is V1, and the gray-scale value of secondary peak is V2; This threshold value is a gray-scale value in gray-scale value V1 and V2 interval, the no more than number of pixels that should other gray-scale values of interval of the number of pixels of this gray-scale value.

As a preferred embodiment of the present invention, establishing the threshold value of choosing is V, and described image binaryzation unit is made as 0 with gray-scale value greater than the pixel of V, and gray-scale value is made as 255 less than or equal to the pixel of V.

As a preferred embodiment of the present invention, every some pixel of image represents that with two-dimensional array described marginal point selected cell is lined by line scan from both direction;

Direction be since the 0th row by column increase write down pixel coordinate (x, the y of this point until first non-vanishing pixel in this row ₁), increase by column until be zero place again, write down this pixel (x, y ₂), add up non-vanishing number of pixels d ₁=y ₂-y ₁, stop the scanning of this direction;

Another direction is to reduce by column since last row, when occurring at first non-vanishing, writes down coordinate (x, the y of this pixel ₃), when again to occur one be zero pixel, write down coordinate (x, the y of this pixel ₄), and statistics (x, y ₄) and (x, y ₃) between number of pixels d ₂=y ₃-y ₄If, y in this row ₄＜=y ₂Then reject the data of changing one's profession; Only has the y of working as ₄Y ₂, and d ₁/ d ₂Just think that at interval [0.9,1.1] Shi Caihui these points meet the condition of fitted ellipse, wherein, (x, y ₁), (x, y ₃) be the coordinate points on the outward flange, (x, y ₂), (x, y ₄) be the point on the inward flange.

As a preferred embodiment of the present invention, A, B are established and are bianry image with the marginal point of the left edge regions of delegation in two zones about described marginal point selected cell is divided into picture, and C, D are that bianry image is with the marginal point of the right edge regions of delegation; Because the center of light cladding diameter and fibre core can regard concentric as, if such as the ratio d1/d2 of the length d 1 of fruit dot A, B point-to-point transmission and the length d 2 between C, D point between the setting district [0.9,1.1], then be judged as the both sides consistency of thickness, namely these marginal points meet the condition of fitted ellipse, and coordinate points A, D are the point on the ellipse of outside, coordinate B, C are the point on the interior ellipse, if d1/d2 in [0.9,1.1], just can not choose these coordinate points not between the setting district; Damaged place will be arranged like this to getting rid of;

Described marginal point selected cell is divided into up and down two halves zone with image, by with about the zone select the identical principle of marginal point to extract coordinate points;

All coordinate points that inspection is extracted are if having the coordinate points of repetition then delete repeating part.

As a preferred embodiment of the present invention, the ellipse fitting of described ellipse fitting and parameter acquiring unit and parameter acquiring process comprise:

The general equation of quafric curve is expressed as x ²+ α xy+ β y ²Quafric curve just can be obtained as long as obtain parameter, if satisfy α in+γ x+ ζ y+ η=0 ²-4 β＜0 this quafric curve represents an ellipse;

All outward flange coordinate (x that choose at described marginal point selected cell ₁, y ₁), (x ₂, y ₂) ..., (x _n, y _n) middle order $F(\mathrm{\&alpha;},\mathrm{\&beta;},\mathrm{\&gamma;},\mathrm{\&zeta;},\mathrm{\&eta;})=\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}(x_i^2+\mathrm{\&alpha;}{x}_i{y}_i+{\mathrm{\&beta;<figure-callout\; id="2"\; label="y"\; filenames="HDA00002710960000011.png"\; state="\{\{state\}\}">y</figure-callout>}}^2+\mathrm{\&gamma;x}+\mathrm{\&zeta;y}+\mathrm{\&eta;});$

According to the elliptic parameter of match, obtain the major axis d of outer ellipse _Macr, minor axis d _Min, center (x _e, y _e); Major axis S with interior ellipse _Macr, minor axis S _Min, center (x _In, y _In); Can obtain the elliptic geometry parameter thus:

Cladding diameter d=(d _Macr+ d _Min)/2;

Non-circularity of cladding

Core diameter S=(S _Macr+ S _Min)/2;

Fiber core non circularity $\mathrm{\&epsiv;}=(S_{\mathrm{macr}}-S_{\min })/S;$

Fibre core-covering nonconcentricity $\mathrm{\&Delta;}=\sqrt{{(x_e-x_{\mathrm{in}})}^2+{(y_e-y_{\mathrm{in}})}^2};$

When F (α, beta, gamma, ζ, η) hour, be the exterior elliptical of best-fit, obtain the exterior elliptical parameter alpha ₁β ₁, γ ₁, ζ ₁, η ₁

In like manner, the coordinate points of all internal edges of extracting is pressed $F(\mathrm{\&alpha;},\mathrm{\&beta;},\mathrm{\&gamma;},\mathrm{\&zeta;},\mathrm{\&eta;})=\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}(x_i^2+\mathrm{\&alpha;}{x}_i{y}_i+{\mathrm{\&beta;<figure-callout\; id="2"\; label="y"\; filenames="HDA00002710960000011.png"\; state="\{\{state\}\}">y</figure-callout>}}^2+\mathrm{\&gamma;x}+\mathrm{\&zeta;y}+\mathrm{\&eta;});$ When F (α, beta, gamma, ζ, η) hour, be that the inside of best-fit is oval, obtain inner elliptic parameter α ₂, β ₂, γ ₂, ζ ₂, η _2.

A kind of detection method for damaged optical fiber geometric parameter, described method comprises:

Read optical fiber image;

The threshold value selecting step, the statistics grey level histogram is chosen optimal threshold;

The image binaryzation treatment step carries out binary conversion treatment to image;

Marginal point is selected step, accurate outward flange and the inward flange point selected in binary image;

Ellipse fitting and parameter acquiring step are obtained outer peripheral elliptic parameter by all outward flange points of choosing, simulate the parameter of the ellipse of inward flange by the inward flange point; Obtain the geometric parameter of this optical fiber picture behind the ellipse by the match outer edge.

As a preferred embodiment of the present invention, the threshold value of choosing in the described threshold value selecting step is the gray-scale value of trough between first crest and the secondary peak; Wherein, the gray-scale value of first crest is V1, and the gray-scale value of secondary peak is V2;

This threshold value is a gray-scale value in gray-scale value V1 and V2 interval, the no more than number of pixels that should other gray-scale values of interval of the number of pixels of this gray-scale value.

If the threshold value of choosing is V, described image binaryzation unit is made as 0 with gray-scale value greater than the pixel of V, and gray-scale value is made as 255 less than or equal to the pixel of V.

As a preferred embodiment of the present invention, described marginal point is selected to line by line scan from both direction in the step;

Direction be since the 0th row by column increase write down pixel coordinate (x, the y of this point until first non-vanishing pixel in this row ₁), increase by column until be zero place again, write down this pixel (x, y ₂), add up non-vanishing number of pixels d ₁=y ₂-y ₁, stop the scanning of this direction;

Another direction is to reduce by column from last row beginning, when occurring at first non-vanishing, writes down coordinate (x, the y of this pixel ₃), when again to occur one be zero pixel, write down coordinate (x, the y of this pixel ₄), and statistics (x, y ₄) and (x, y ₃) between number of pixels d ₂=y ₃-y ₄If, y in this row ₄＜=y ₂Then reject the data of changing one's profession; Only has the y of working as ₄Y ₂, and d ₁/ d ₂Just think that at interval [0.9,1.1] Shi Caihui these points meet the condition of fitted ellipse, wherein, (x, y ₁), (x, y ₃) be the coordinate points on the outward flange, (x, y ₂), (x, y ₄) be the point on the inward flange.

As a preferred embodiment of the present invention, described ellipse fitting and parameter acquiring step comprise:

The general equation of quafric curve is expressed as x ²+ α xy+ β y ²Quafric curve just can be obtained as long as obtain parameter, if satisfy α in+γ x+ ζ y+ η=0 ²-4 β＜0 this quafric curve represents an ellipse;

All outward flange coordinate (x that choose at described marginal point selected cell ₁, y ₁), (x ₂, y ₂) ..., (x _n, y _n) middle order $F(\mathrm{\&alpha;},\mathrm{\&beta;},\mathrm{\&gamma;},\mathrm{\&zeta;},\mathrm{\&eta;})=\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}(x_i^2+\mathrm{\&alpha;}{x}_i{y}_i+{\mathrm{\&beta;<figure-callout\; id="2"\; label="y"\; filenames="HDA00002710960000011.png"\; state="\{\{state\}\}">y</figure-callout>}}^2+\mathrm{\&gamma;x}+\mathrm{\&zeta;y}+\mathrm{\&eta;});$

When F (α, beta, gamma, ζ, η) hour, be the exterior elliptical of best-fit, obtain the exterior elliptical parameter alpha ₁, β ₁, γ ₁, ζ ₁, η ₁

In like manner, the coordinate points of all internal edges of extracting is pressed $F(\mathrm{\&alpha;},\mathrm{\&beta;},\mathrm{\&gamma;},\mathrm{\&zeta;},\mathrm{\&eta;})=\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}(x_i^2+\mathrm{\&alpha;}{x}_i{y}_i+{\mathrm{\&beta;<figure-callout\; id="2"\; label="y"\; filenames="HDA00002710960000011.png"\; state="\{\{state\}\}">y</figure-callout>}}^2+\mathrm{\&gamma;x}+\mathrm{\&zeta;y}+\mathrm{\&eta;});$ When F (α, beta, gamma, ζ, η) hour, be that the inside of best-fit is oval, obtain inner elliptic parameter α ₂, β ₂, γ ₂, ζ ₂, η ₂

Beneficial effect of the present invention is: compare with traditional detection method, the method that detection system and the method for damaged optical fiber geometric parameter that the present invention proposes used novel extraction coordinate points, can reject the not point on ellipse, further improve the precision of ellipse fitting.

The present invention processes the detection that the optical fiber picture accurately calculates the geometric parameter (comprising: cladding diameter, non-circularity of cladding, core diameter, fiber core non circularity, fibre core-covering nonconcentricity) of optical fiber with the method that image is processed, and especially can detect exactly the geometric parameter of damaged optical fiber.

Description of drawings

Fig. 1 is the composition synoptic diagram of detection system of the present invention.

Fig. 2 is the process flow diagram of detection method of the present invention.

Fig. 3 is the optical fiber picture synoptic diagram that collects.

Fig. 4 is the synoptic diagram that image binaryzation is processed.

Fig. 5 is the synoptic diagram that marginal point is selected.

Fig. 6 is the exterior elliptical of best-fit, inner oval synoptic diagram.

Embodiment

Describe the preferred embodiments of the present invention in detail below in conjunction with accompanying drawing.

Embodiment one

See also Fig. 1, the present invention has disclosed a kind of detection system for damaged optical fiber geometric parameter, and described system comprises: image fetching unit 1, threshold value are chosen unit 2, image binaryzation processing unit 3, marginal point selected cell 4, ellipse fitting and parameter acquiring unit 5.

Image fetching unit 1 is in order to read optical fiber image.

Threshold value is chosen unit 2 in order to add up grey level histogram, chooses optimal threshold.Particularly, the described threshold value threshold value of choosing unit selection is the gray-scale value of trough between first crest and the secondary peak; Wherein, the gray-scale value of first crest is V1, and the gray-scale value of secondary peak is V2; This threshold value is a gray-scale value in gray-scale value V1 and V2 interval, the no more than number of pixels that should other gray-scale values of interval of the number of pixels of this gray-scale value.If the threshold value of choosing is V, described image binaryzation unit is made as 0 with gray-scale value greater than the pixel of V, and gray-scale value is made as 255 less than or equal to the pixel of V.

Image binaryzation processing unit 3 is in order to carry out binary conversion treatment to image.

Marginal point selected cell 4 is in order to accurate outward flange and the inward flange point selected in binary image.

Every some pixel of image represents with two-dimensional array, and described marginal point selected cell 4 is lined by line scan from both direction;

Direction be since the 0th row by column increase write down pixel coordinate (x, the y of this point until first non-vanishing pixel in this row ₁), increase by column until be zero place again, write down this pixel (x, y ₂), add up non-vanishing number of pixels d ₁=y ₂-y ₁, stop the scanning of this direction;

Another direction is to reduce by column from last row beginning, when occurring at first non-vanishing, writes down coordinate (x, the y of this pixel ₃), when again to occur one be zero pixel, write down coordinate (x, the y of this pixel ₄), and statistics (x, y ₄) and (x, y ₃) between number of pixels d ₂=y ₃-y ₄If, y in this row ₄＜=y ₂Then reject the data of changing one's profession; Only has the y of working as ₄Y ₂, and d ₁/ d ₂Just think that at interval [0.9,1.1] Shi Caihui these points meet the condition of fitted ellipse, wherein, (x, y ₁), (x, y ₃) be the coordinate points on the outward flange, (x, y ₂), (x, y ₄) be the point on the inward flange.

A, B are established and are bianry image with the marginal point of the left edge regions of delegation in two zones about described marginal point selected cell 4 is divided into picture, and C, D are that bianry image is with the marginal point of the right edge regions of delegation; Because the center of light cladding diameter and fibre core can regard concentric as, if such as the ratio d1/d2 of the length d 1 of fruit dot A, B point-to-point transmission and the length d 2 between C, D point between the setting district [0.9,1.1], then be judged as the both sides consistency of thickness, namely these marginal points meet the condition of fitted ellipse, and coordinate points A, D are the point on the ellipse of outside, coordinate B, C are the point on the interior ellipse, if d1/d2 in [0.9,1.1], just can not choose these coordinate points not between the setting district; Damaged place will be arranged like this to getting rid of.Similarly, described marginal point selected cell is divided into up and down two halves zone with image, by with about the zone select the identical principle of marginal point to extract coordinate points.All coordinate points that inspection is extracted are if having the coordinate points of repetition then delete repeating part.

Ellipse fitting and parameter acquiring unit 5 are obtained outer peripheral elliptic parameter in order to the outward flange point of choosing by all, simulate the parameter of the ellipse of inward flange by the inward flange point; Obtain the geometric parameter of this optical fiber picture behind the ellipse by the match outer edge.

The ellipse fitting of described ellipse fitting and parameter acquiring unit 5 and parameter acquiring process comprise:

The general equation of quafric curve is expressed as x ²+ α xy+ β y ²Quafric curve just can be obtained as long as obtain parameter, if satisfy α in+γ x+ ζ y+ η=0 ²-4 β＜0 this quafric curve represents an ellipse;

All outward flange coordinate (x that choose at described marginal point selected cell ₁, y ₁), (x ₂, y ₂) ..., (x _n, y _n) middle order $F(\mathrm{\&alpha;},\mathrm{\&beta;},\mathrm{\&gamma;},\mathrm{\&zeta;},\mathrm{\&eta;})=\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}({\mathrm{<figure-callout\; id="2"\; label="x\; i"\; filenames="HDA00002710960000011.png"\; state="\{\{state\}\}">x</figure-callout>}}_i^{}+\mathrm{\&alpha;}{x}_i{y}_i+{\mathrm{\&beta;<figure-callout\; id="2"\; label="y"\; filenames="HDA00002710960000011.png"\; state="\{\{state\}\}">y</figure-callout>}}^2+\mathrm{\&gamma;x}+\mathrm{\&zeta;y}+\mathrm{\&eta;});$

According to the elliptic parameter of match, obtain the major axis d of outer ellipse _Macr, minor axis d _Min, center (x _e, y _e); Major axis S with interior ellipse _Macr, minor axis S _Min, center (x _In, y _In); Can obtain the elliptic geometry parameter thus:

Cladding diameter d=(d _Macr+ d _Min)/2;

Non-circularity of cladding

Core diameter S=(S _Macr+ S _Min)/2;

Fiber core non circularity $\mathrm{\&epsiv;}=(S_{\mathrm{macr}}-S_{\min })/S;$

Fibre core-covering nonconcentricity $\mathrm{\&Delta;}=\sqrt{{(x_e-x_{\mathrm{in}})}^2+{(y_e-y_{\mathrm{in}})}^2};$

When F (α, beta, gamma, ζ, η) hour, be the exterior elliptical of best-fit, obtain the exterior elliptical parameter alpha ₁, β ₁, γ ₁, ζ ₁, η ₁

In like manner, the coordinate points of all internal edges of extracting is pressed $F(\mathrm{\&alpha;},\mathrm{\&beta;},\mathrm{\&gamma;},\mathrm{\&zeta;},\mathrm{\&eta;})=\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}({\mathrm{<figure-callout\; id="2"\; label="x\; i"\; filenames="HDA00002710960000011.png"\; state="\{\{state\}\}">x</figure-callout>}}_i^{}+\mathrm{\&alpha;}{x}_i{y}_i+{\mathrm{\&beta;<figure-callout\; id="2"\; label="y"\; filenames="HDA00002710960000011.png"\; state="\{\{state\}\}">y</figure-callout>}}^2+\mathrm{\&gamma;x}+\mathrm{\&zeta;y}+\mathrm{\&eta;});$ When F (α, beta, gamma, ζ, η) hour, be that the inside of best-fit is oval, obtain inner elliptic parameter α ₂, β ₂, γ ₂, ζ ₂, η ₂

More than introduced the detection system that the present invention is directed to damaged optical fiber geometric parameter, the present invention also discloses a kind of detection method for damaged optical fiber geometric parameter when disclosing said system; See also Fig. 2, described method comprises the steps:

[step S1] reads optical fiber image;

[step S2] threshold value selecting step, the statistics grey level histogram is chosen optimal threshold;

The threshold value of choosing in the described threshold value selecting step is the gray-scale value of trough between first crest and the secondary peak; Wherein, the gray-scale value of first crest is V1, and the gray-scale value of secondary peak is V2; This threshold value is a gray-scale value in gray-scale value V1 and V2 interval, the no more than number of pixels that should other gray-scale values of interval of the number of pixels of this gray-scale value; If the threshold value of choosing is V, gray-scale value is made as 0 greater than the pixel of V, gray-scale value is made as 255 less than or equal to the pixel of V.

[step S3] image binaryzation treatment step carries out binary conversion treatment to image;

[step S4] marginal point is selected step, accurate outward flange and the inward flange point selected in binary image;

Every some pixel of image represents that with two-dimensional array described marginal point is selected to line by line scan from both direction in the step;

Direction be since the 0th row by column increase write down pixel coordinate (x, the y of this point until first non-vanishing pixel in this row ₁), increase by column until be zero place again, write down this pixel (x, y ₂), add up non-vanishing number of pixels d ₁=y ₂-y ₁, stop the scanning of this direction;

Another direction is to reduce by column from last row beginning, when occurring at first non-vanishing, writes down coordinate (x, the y of this pixel ₃), when again to occur one be zero pixel, write down coordinate (x, the y of this pixel ₄), and statistics (x, y ₄) and (x, y ₃) between number of pixels d ₂=y ₃-y ₄If, y in this row ₄＜=y ₂Then reject the data of changing one's profession; Only has the y of working as ₄Y ₂, and d ₁/ d ₂Just think that at interval [0.9,1.1] Shi Caihui these points meet the condition of fitted ellipse, wherein, (x, y ₁), (x, y ₃) be the coordinate points on the outward flange, (x, y ₂), (x, y ₄) be the point on the inward flange.

[step S5] ellipse fitting and parameter acquiring step are obtained outer peripheral elliptic parameter by all outward flange points of choosing, simulate the parameter of the ellipse of inward flange by the inward flange point; Obtain the geometric parameter of this optical fiber picture behind the ellipse by the match outer edge.

Particularly, described ellipse fitting and parameter acquiring step comprise:

The general equation of quafric curve is expressed as x ²+ α xy+ β y ²Quafric curve just can be obtained as long as obtain parameter, if satisfy α in+γ x+ ζ y+ η=0 ²-4 β＜0 this quafric curve represents an ellipse;

All outward flange coordinate (x that choose at described marginal point selected cell ₁, y ₁), (x ₂, y ₂) ..., (x _n, y _n) middle order $F(\mathrm{\&alpha;},\mathrm{\&beta;},\mathrm{\&gamma;},\mathrm{\&zeta;},\mathrm{\&eta;})=\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}({\mathrm{<figure-callout\; id="2"\; label="x\; i"\; filenames="HDA00002710960000011.png"\; state="\{\{state\}\}">x</figure-callout>}}_i^{}+\mathrm{\&alpha;}{x}_i{y}_i+{\mathrm{\&beta;<figure-callout\; id="2"\; label="y"\; filenames="HDA00002710960000011.png"\; state="\{\{state\}\}">y</figure-callout>}}^2+\mathrm{\&gamma;x}+\mathrm{\&zeta;y}+\mathrm{\&eta;});$

When F (α, beta, gamma, ζ, η) hour, be the exterior elliptical of best-fit, obtain the exterior elliptical parameter alpha ₁, β ₁, γ ₁, ζ ₁, η ₁

In like manner, the coordinate points of all internal edges of extracting is pressed $F(\mathrm{\&alpha;},\mathrm{\&beta;},\mathrm{\&gamma;},\mathrm{\&zeta;},\mathrm{\&eta;})=\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}({\mathrm{<figure-callout\; id="2"\; label="x\; i"\; filenames="HDA00002710960000011.png"\; state="\{\{state\}\}">x</figure-callout>}}_i^{}+\mathrm{\&alpha;}{x}_i{y}_i+{\mathrm{\&beta;<figure-callout\; id="2"\; label="y"\; filenames="HDA00002710960000011.png"\; state="\{\{state\}\}">y</figure-callout>}}^2+\mathrm{\&gamma;x}+\mathrm{\&zeta;y}+\mathrm{\&eta;});$ When F (α, beta, gamma, ζ, η) hour, be that the inside of best-fit is oval, obtain inner elliptic parameter α ₂, β ₂, γ ₂, ζ ₂, η ₂

Embodiment two

Detection method of the present invention and system divide the two large divisions.First: the obtaining and explaining of image geometry parameter; Second portion: the workflow of detection system.

[1] the image geometry parameter obtaining and explaining

This paper has proposed a kind of method that can detect damaged optical fiber, and key is the accuracy of fitting parameter, and how this gets a little accurately if just being related to.Fig. 5 is the optical fiber picture that collects under actual application environment, and this picture obviously has damaged, next is with method of the present invention this image to be processed.

Specifically be expressed as follows:

1. add up grey level histogram, choose binary-state threshold:

The threshold value of choosing should be the gray-scale value (trough) between first crest (gray-scale value is V1) and the secondary peak (gray-scale value is V2), this threshold value is a gray-scale value in gray-scale value V1 and V2 interval, the no more than number of pixels that should other gray-scale values of interval of the number of pixels of this gray-scale value.

2. image binaryzation (such as Fig. 6):

If the threshold value of choosing is V, gray-scale value is made as 0 greater than the pixel of V, gray-scale value is made as 255 less than or equal to the pixel of V.

3. select marginal point:

The choosing method of marginal point is very large on the impact of the degree of accuracy of the parameter of match, and we are divided into picture first left and right sides two halves zone, and wherein A, B, C, D are with the marginal point of the left right two halves of delegation in the bianry image.Because the center of light cladding diameter and fibre core can regard concentric as, if can think that the length d 1 of an A, B point-to-point transmission and the ratio d1/d2 of the length d 2 between C, D point (are selecting between interval 0.9 to 1.1) about 1 here, be the both sides consistency of thickness, just think that these marginal points meet the condition of fitted ellipse, coordinate points A, D are the point on the ellipse of outside, coordinate B, C are the point on the interior ellipse, if d1/d2 not in this scope, just can not choose these points.So just can will damaged place be arranged to removing point (just will be removed such as zone 1, zone 2).

Image is divided into up and down two halves zone, extracts coordinate points with same principle.

All coordinate points that inspection is extracted are if having the coordinate points of repetition then delete repeating part.

4. ellipse fitting:

The general equation of quafric curve can be expressed as x ²+ α xy+ β y ²Quafric curve just can be obtained as long as obtain parameter, if satisfy α in+γ x+ ζ y+ η=0 ²-4 β＜0 this quafric curve represents an ellipse.

At above all outward flange coordinate (x that choose ₁, y ₁), (x ₂, y ₂) ..., (x _n, y _n) middle order $F(\mathrm{\&alpha;},\mathrm{\&beta;},\mathrm{\&gamma;},\mathrm{\&zeta;},\mathrm{\&eta;})=\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}({\mathrm{<figure-callout\; id="2"\; label="x\; i"\; filenames="HDA00002710960000011.png"\; state="\{\{state\}\}">x</figure-callout>}}_i^{}+\mathrm{\&alpha;}{x}_i{y}_i+{\mathrm{\&beta;<figure-callout\; id="2"\; label="y"\; filenames="HDA00002710960000011.png"\; state="\{\{state\}\}">y</figure-callout>}}^2+\mathrm{\&gamma;x}+\mathrm{\&zeta;y}+\mathrm{\&eta;})---\left(1.1\right)$

Then when F (α, beta, gamma, ζ, η) hour, be the exterior elliptical (such as Fig. 6) of best-fit, can obtain the exterior elliptical parameter alpha ₁, β ₁, γ ₁, ζ ₁, η ₁

In like manner, to the coordinate knock type (1.1) of all internal edges of extracting when F (α, beta, gamma, ζ, η) hour, be that the inside of best-fit is oval, can obtain inner elliptic parameter α ₂, β ₂, γ ₂, ζ ₂, η ₂

5. find the solution optical fiber geometric parameter:

According to the elliptic parameter of match, can obtain the major axis d of outer ellipse _Macr, minor axis d _Min, center (x _e, y _e); Major axis S with interior ellipse _Macr, minor axis S _Min, center (x _In, y _In).Can obtain the elliptic geometry parameter thus:

Cladding diameter d=(d _Macr+ d _Min)/2

Non-circularity of cladding

Core diameter S=(S _Macr+ S _Min)/2

Fiber core non circularity $\mathrm{\&epsiv;}=(S_{\mathrm{macr}}-S_{\min })/S;$

Fibre core-covering nonconcentricity $\mathrm{\&Delta;}=\sqrt{{(x_e-x_{\mathrm{in}})}^2+{(y_e-y_{\mathrm{in}})}^2};$

2.[2] the detection system workflow

1. reading images.

2. add up grey level histogram, choose optimal threshold, and image is carried out binary conversion treatment.

3. in binary image, select accurately outward flange and inward flange point, and carry out ellipse fitting.

Every some pixel supposing image represents with two-dimensional array, line by line scan from both direction, direction be since the 0th row by column increase write down pixel coordinate (x, the y of this point until first non-vanishing pixel in this row ₁), increase by column until be zero place again, write down this pixel (x, y ₂), add up non-vanishing number of pixels d ₁=y ₂-y ₁, stop the scanning of this direction; Another direction is to reduce by column from last row beginning, when occurring at first non-vanishing, writes down coordinate (x, the y of this pixel ₃), when again to occur one be zero pixel, write down coordinate (x, the y of this pixel ₄), and statistics (x, y ₄) and (x, y ₃) between number of pixels d ₂=y ₃-y ₄If, y in this row ₄＜=y ₂Then reject the data of changing one's profession.Only has the y of working as ₄Y ₂, and d ₁/ d ₂(in interval (0.9,1.1)) can just think that these points meet the condition of fitted ellipse, wherein (x, y near 1 ₁), (x, y ₃) be the coordinate points on the outward flange, (x, y ₂), (x, y ₄) be the point on the inward flange.

4. ellipse fitting and obtain geometric parameter

Obtain with the used outward flange point of choosing

Obtain α, beta, gamma, ζ, the parameters such as η.In like manner can simulate by the inward flange point parameter of the ellipse of inward flange.Just can obtain the geometric parameter of this optical fiber picture by the oval rear of match outer edge.

In sum, the method that detection system and the method for damaged optical fiber geometric parameter that the present invention proposes used novel extraction coordinate points can be rejected the not point on ellipse, further improves the precision of ellipse fitting.

Here description of the invention and application is illustrative, is not to want with scope restriction of the present invention in the above-described embodiments.Here the distortion of disclosed embodiment and change is possible, and the various parts of the replacement of embodiment and equivalence are known for those those of ordinary skill in the art.Those skilled in the art are noted that in the situation that does not break away from spirit of the present invention or essential characteristic, and the present invention can be with other form, structure, layout, ratio, and realize with other assembly, material and parts.In the situation that does not break away from the scope of the invention and spirit, can carry out other distortion and change to disclosed embodiment here.

Claims (10)

1. the detection system for damaged optical fiber geometric parameter is characterized in that, described system comprises:

Image fetching unit is in order to read optical fiber image;

Threshold value is chosen the unit, in order to add up grey level histogram, chooses optimal threshold;

The image binaryzation processing unit is in order to carry out binary conversion treatment to image;

The marginal point selected cell is in order to accurate outward flange and the inward flange point selected in binary image;

Ellipse fitting and parameter acquiring unit are obtained outer peripheral elliptic parameter in order to the outward flange point of choosing by all, simulate the parameter of the ellipse of inward flange by the inward flange point; Obtain the geometric parameter of this optical fiber picture behind the ellipse by the match outer edge.

2. the detection system for damaged optical fiber geometric parameter according to claim 1 is characterized in that:

Described threshold value is chosen the threshold value of unit selection and is the gray-scale value of trough between first crest and the secondary peak; Wherein, the gray-scale value of first crest is V1, and the gray-scale value of secondary peak is V2;

This threshold value is a gray-scale value in gray-scale value V1 and V2 interval, the no more than number of pixels that should other gray-scale values of interval of the number of pixels of this gray-scale value.

3. the detection system for damaged optical fiber geometric parameter according to claim 2 is characterized in that:

If the threshold value of choosing is V, described image binaryzation unit is made as 0 with gray-scale value greater than the pixel of V, and gray-scale value is made as 255 less than or equal to the pixel of V.

4. the detection system for damaged optical fiber geometric parameter according to claim 1 is characterized in that:

Described marginal point selected cell is lined by line scan from both direction;

Direction be since the 0th row by column increase write down pixel coordinate (x, the y of this point until first non-vanishing pixel in this row ₁), increase by column until be zero place again, write down this pixel (x, y ₂), add up non-vanishing number of pixels d ₁=y ₂-y ₁, stop the scanning of this direction;

Another direction is to reduce by column since last row, when occurring at first non-vanishing, writes down coordinate (x, the y of this pixel ₃), when again to occur one be zero pixel, write down coordinate (x, the y of this pixel ₄), and statistics (x, y ₄) and (x, y ₃) between number of pixels d ₂=y ₃-y ₄If, y in this row ₄＜=y ₂Then reject the data of changing one's profession; Only has the y of working as ₄Y ₂, and d ₁/ d ₂Just think that at interval [0.9,1.1] Shi Caihui these points meet the condition of fitted ellipse, wherein, (x, y ₁), (x, y ₃) be the coordinate points on the outward flange, (x, y ₂), (x, y ₄) be the point on the inward flange.

5. the detection system for damaged optical fiber geometric parameter according to claim 4 is characterized in that:

A, B are established and are bianry image with the marginal point of the left edge regions of delegation in two zones about described marginal point selected cell is divided into picture, and C, D are that bianry image is with the marginal point of the right edge regions of delegation; Because the center of light cladding diameter and fibre core can regard concentric as, if such as the ratio d1/d2 of the length d 1 of fruit dot A, B point-to-point transmission and the length d 2 between C, D point between the setting district [0.9,1.1], then be judged as the both sides consistency of thickness, namely these marginal points meet the condition of fitted ellipse, and coordinate points A, D are the point on the ellipse of outside, coordinate B, C are the point on the interior ellipse, if d1/d2 in [0.9,1.1], just can not choose these coordinate points not between the setting district; Damaged place will be arranged like this to getting rid of;

Described marginal point selected cell is divided into up and down two halves zone with image, by with about the zone select the identical principle of marginal point to extract coordinate points.

6. the detection system for damaged optical fiber geometric parameter according to claim 4 is characterized in that:

The ellipse fitting of described ellipse fitting and parameter acquiring unit and parameter acquiring process comprise:

The general equation of quafric curve is expressed as x ²+ α xy+ β y ²Quafric curve just can be obtained as long as obtain parameter, if satisfy α in+γ x+ ζ y+ η=0 ²-4 β＜0 this quafric curve represents an ellipse;

All outward flange coordinate (x that choose at described marginal point selected cell ₁, y ₁), (x ₂, y ₂) ..., (x _n, y _n) middle order

According to the elliptic parameter of match, obtain the major axis d of outer ellipse _Macr, minor axis d _Min, center (x _e, y _e); Major axis S with interior ellipse _Macr, minor axis S _Min, center (x _In, y _In); Can obtain the elliptic geometry parameter thus:

Cladding diameter d=(d _Macr+ d _Min)/2;

Non-circularity of cladding

Core diameter S=(S _Macr+ S _Min)/2;

Fiber core non circularity

Fibre core-covering nonconcentricity

When F (α, beta, gamma, ζ, η) hour, be the exterior elliptical of best-fit, obtain the exterior elliptical parameter alpha ₁, β ₁, γ ₁, ζ ₁, η ₁In like manner, the coordinate points of all internal edges of extracting is pressed

When F (α, beta, gamma, ζ, η) hour, be that the inside of best-fit is oval, obtain inner elliptic parameter α ₂, β ₂, γ ₂, ζ ₂, η ₂

7. the detection method for damaged optical fiber geometric parameter is characterized in that, described method comprises:

Read optical fiber image;

The threshold value selecting step, the statistics grey level histogram is chosen optimal threshold;

The image binaryzation treatment step carries out binary conversion treatment to image;

Marginal point is selected step, accurate outward flange and the inward flange point selected in binary image;

Ellipse fitting and parameter acquiring step are obtained outer peripheral elliptic parameter by all outward flange points of choosing, simulate the parameter of the ellipse of inward flange by the inward flange point; Obtain the geometric parameter of this optical fiber picture behind the ellipse by the match outer edge.

8. the detection method for damaged optical fiber geometric parameter according to claim 7 is characterized in that:

The threshold value of choosing in the described threshold value selecting step is the gray-scale value of trough between first crest and the secondary peak; Wherein, the gray-scale value of first crest is V1, and the gray-scale value of secondary peak is V2;

The number of pixels of the gray-scale value of choosing is lacked than the number of pixels of any one gray-scale value except this gray-scale value in V1 and V2 interval;

If the threshold value of choosing is V, described image binaryzation unit is made as 0 with gray-scale value greater than the pixel of V, and gray-scale value is made as 255 less than or equal to the pixel of V.

9. the detection method for damaged optical fiber geometric parameter according to claim 7 is characterized in that:

Every some pixel of image represents that with two-dimensional array described marginal point is selected to line by line scan from both direction in the step;

Direction be since the 0th row by column increase write down pixel coordinate (x, the y of this point until first non-vanishing pixel in this row ₁), increase by column until be zero place again, write down this pixel (x, y ₂), add up non-vanishing number of pixels d ₁=y ₂-y ₁, stop the scanning of this direction;

Another direction is to reduce by column since last row, when occurring at first non-vanishing, writes down coordinate (x, the y of this pixel ₃), when again to occur one be zero pixel, write down coordinate (x, the y of this pixel ₄), and statistics (x, y ₄) and (x, y ₃) between number of pixels d ₂=y ₃-y ₄If, y in this row ₄＜=y ₂Then reject the data of changing one's profession; Only has the y of working as ₄Y ₂, and d ₁/ d ₂Just think that at interval [0.9,1.1] Shi Caihui these points meet the condition of fitted ellipse, wherein, (x, y ₁), (x, y ₃) be the coordinate points on the outward flange, (x, y ₂), (x, y ₄) be the point on the inward flange.

10. the detection method for damaged optical fiber geometric parameter according to claim 9 is characterized in that:

Described ellipse fitting and parameter acquiring step comprise:

The general equation of quafric curve is expressed as x ²+ α xy+ β y ²Quafric curve just can be obtained as long as obtain parameter, if satisfy α in+γ x+ ζ y+ η=0 ²-4 β＜0 this quafric curve represents an ellipse;

All outward flange coordinate (x that choose at described marginal point selected cell ₁, y ₁), (x ₂, y ₂) ..., (x _n, y _n) middle order

When F (α, beta, gamma, ζ, η) hour, be the exterior elliptical of best-fit, obtain the exterior elliptical parameter alpha ₁, β ₁, γ ₁, ζ ₁, η ₁

In like manner, the coordinate points of all internal edges of extracting is pressed

When F (α, beta, gamma, ζ, η) hour, be that the inside of best-fit is oval, obtain inner elliptic parameter α ₂, β ₂, γ ₂, ζ ₂, η ₂

Images