CN108227077B - Method and system for estimating fusion loss of ribbon fiber - Google Patents
Method and system for estimating fusion loss of ribbon fiber Download PDFInfo
- Publication number
- CN108227077B CN108227077B CN201711476168.XA CN201711476168A CN108227077B CN 108227077 B CN108227077 B CN 108227077B CN 201711476168 A CN201711476168 A CN 201711476168A CN 108227077 B CN108227077 B CN 108227077B
- Authority
- CN
- China
- Prior art keywords
- curvature
- brightness value
- fiber
- rough
- points
- 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.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G02—OPTICS
- G02B—OPTICAL ELEMENTS, SYSTEMS OR APPARATUS
- G02B6/00—Light guides; Structural details of arrangements comprising light guides and other optical elements, e.g. couplings
- G02B6/24—Coupling light guides
- G02B6/255—Splicing of light guides, e.g. by fusion or bonding
Abstract
The invention provides a method and a system for estimating fusion loss of a ribbon optical fiber, wherein the method comprises the following steps: analyzing the banded optical fiber image after fusion splicing is completed, and calculating the curvature parameter of a certain distance range of an optical fiber fusion splicing point, wherein the curvature parameter comprises: the maximum curvature of the upper edge of the cladding, the maximum curvature of the lower edge of the cladding, the maximum curvature of the upper edge of the fiber core, the maximum curvature of the lower edge of the fiber core and the maximum curvature of the axle center; obtaining bending parameters which can cause the optical fiber to bend in the optical fiber test result; taking the bending parameters and the curvature parameters as initial independent variables, taking actual loss of a test as a dependent variable, obtaining a plurality of test data as training samples, performing linear regression calculation, and screening out optimized independent variables according to the correlation coefficients between the independent variables and the dependent variables; and constructing a multiple linear regression model according to the actual loss and the optimization independent variable, and estimating the loss according to the multiple linear regression model. The accuracy of the loss estimation is improved.
Description
Technical Field
The invention relates to the technical field of optical fiber fusion splicers, in particular to a method and a system for estimating fusion loss of a strip-shaped optical fiber.
Background
The fusion splicer is mainly used for construction and maintenance of optical cables in optical communication, and is applied to optical cable line engineering construction, line maintenance and emergency repair of telecom operators, engineering companies and public institutions, production and test of optical fiber devices, research and teaching of scientific research institutions and the like.
The estimated loss is an estimation result of the actual fusion loss, which cannot represent the actual fusion loss but can play a certain reference value and is an essential step in the optical fiber fusion process. In the process of whole optical fiber adjustment and optical fiber fusion, monitored optical fiber image information is sent to an algorithm analysis program of a fusion splicer through a camera, estimation loss is calculated through the algorithm program of the fusion splicer, the whole process is convenient and fast, an operator only needs to execute fusion splicing operation, and after fusion splicing is completed, the calculated estimation loss can be displayed on a display screen, so that the operator can know the quality of own fusion splicing quickly.
In the prior art, the calculation of estimated loss only simply considers the contribution of thickness change of the optical fiber near the fusion point to loss, and the consideration factor is too single, so that the phenomenon of under-fitting occurs. Other factors, such as: loss due to bending of the weld points, loss that cannot be observed by image alone, and the like cannot be quantitatively analyzed and calculated. Therefore, the welding loss obtained by the existing calculation method for estimating the loss is low in precision.
Disclosure of Invention
The invention aims to provide a method and a system for estimating fusion loss of a ribbon optical fiber, which can improve the accuracy of loss estimation.
In order to solve the above problems, the present invention provides a method for estimating fusion loss of a ribbon fiber, comprising the following steps:
s1: analyzing the banded optical fiber image after fusion splicing is completed, and calculating the curvature parameter of a certain distance range of an optical fiber fusion splicing point, wherein the curvature parameter comprises: the maximum curvature of the upper edge of the cladding, the maximum curvature of the lower edge of the cladding, the maximum curvature of the upper edge of the fiber core, the maximum curvature of the lower edge of the fiber core and the maximum curvature of the axle center;
s2: obtaining bending parameters which can cause the optical fiber to bend in the optical fiber test result;
s3: taking the bending parameters and the curvature parameters as initial independent variables, taking actual loss of a test as a dependent variable, obtaining a plurality of test data as training samples, performing linear regression calculation, and screening out optimized independent variables according to the correlation coefficients between the independent variables and the dependent variables;
s4: and constructing a multiple linear regression model according to the actual loss and the optimization independent variable, and estimating the loss according to the multiple linear regression model.
According to an embodiment of the present invention, the step S1 includes the steps of:
s11: aiming at each row of pixel points in a certain distance range of the optical fiber fusion point in the ribbon optical fiber image, calculating the brightness value on each row;
s12: traversing from top to bottom or from bottom to top for each row of pixel points, if the brightness values of a plurality of pixel points which continuously appear after sudden change of brightness are smaller than or larger than the shadow brightness value, initially judging as a rough cladding upper edge, a rough cladding lower edge, a rough fiber core upper edge and a rough fiber core lower edge, and calculating a rough axis position according to the pixel point brightness values between the rough fiber core upper edge and the rough fiber core lower edge;
s13: aiming at each row of pixel points, performing sub-pixel level calculation according to the rough cladding upper edge, the rough cladding lower edge, the rough fiber core upper edge, the rough fiber core lower edge and the rough axle center position to obtain the cladding upper edge, the cladding lower edge, the fiber core upper edge, the fiber core lower edge and the axle center position on each row;
s14: drawing a discrete optical fiber information curve according to the positions of the upper edge of the cladding, the lower edge of the cladding, the upper edge of the fiber core, the lower edge of the fiber core and the axis of each column;
s15: and solving the curvature of each discrete point in the optical fiber information curve, and calculating the maximum curvature to obtain the maximum value of the curvature of the upper edge of the cladding, the maximum value of the curvature of the lower edge of the cladding, the maximum value of the curvature of the upper edge of the fiber core, the maximum value of the curvature of the lower edge of the fiber core and the maximum value of the curvature of the axis.
According to an embodiment of the present invention, the shadow brightness value is a brightness average value of a row of pixel points including a complete optical fiber in the ribbon optical fiber image when the ribbon optical fiber image is not welded.
According to an embodiment of the present invention, in the step S12, if the traversal from top to bottom is performed for each column of pixel points, the method includes the following steps:
s121 a: after the brightness value of a certain pixel point is changed from being larger than the brightness value of the shadow to being smaller than or equal to the brightness value of the shadow, the brightness values of a plurality of continuous points are smaller than the brightness value of the shadow, and then the points or one point of the points are used as the upper edge of the rough cladding;
s122 a: after the brightness value of a certain pixel point is changed from being smaller than the brightness value of the shadow to being larger than or equal to the brightness value of the shadow, the brightness values of a plurality of continuous points are larger than the brightness value of the shadow, and then the points or one point of the points are used as the upper edge of the rough fiber core;
s123 a: after the brightness value of a certain pixel point is changed from being larger than the brightness value of the shadow to being smaller than or equal to the brightness value of the shadow, the brightness values of a plurality of continuous points are smaller than the brightness value of the shadow, and then the points or one point of the points are used as the lower edge of the rough fiber core;
s124 a: calculating the rough axis position by using a centroid method for the brightness value data of the pixel points between the upper edge of the rough fiber core and the lower edge of the rough fiber core;
s125 a: after the brightness value of a certain pixel point is changed from being smaller than the brightness value of the shadow to being larger than or equal to the brightness value of the shadow, the brightness values of a plurality of continuous points are larger than the brightness value of the shadow, and then the points or one point of the points are used as the lower edge of the rough cladding;
wherein the steps S124a and S125a are interchangeable;
if the traversal is performed from bottom to top for each row of pixel points, the method comprises the following steps:
s121 b: after the brightness value of a certain pixel point is changed from being larger than the brightness value of the shadow to being smaller than or equal to the brightness value of the shadow, the brightness values of a plurality of continuous points are smaller than the brightness value of the shadow, and then the points or one point of the points are used as the lower edge of the rough cladding;
s122 b: after the brightness value of a certain pixel point is changed from being smaller than the brightness value of the shadow to being larger than or equal to the brightness value of the shadow, the brightness values of a plurality of continuous points are larger than the brightness value of the shadow, and then the points or one point of the points are used as the lower edge of the rough fiber core;
s123 b: after the brightness value of a certain pixel point is changed from being larger than the brightness value of the shadow to being smaller than or equal to the brightness value of the shadow, the brightness values of a plurality of continuous points are smaller than the brightness value of the shadow, and then the points or one point of the points are used as the upper edge of the rough fiber core;
s124 b: calculating the rough axis position by using a centroid method for the brightness value data of the pixel points between the upper edge of the rough fiber core and the lower edge of the rough fiber core;
s125 b: after the brightness value of a certain pixel point is changed from being smaller than the brightness value of the shadow to being larger than or equal to the brightness value of the shadow, the brightness values of a plurality of continuous points are larger than the brightness value of the shadow, and then the points or one point of the points are used as the upper edge of the rough cladding;
wherein the steps S124b and S125b are interchangeable.
According to an embodiment of the present invention, the step S13 includes:
s131: respectively carrying out quadratic curve fitting on upper and lower point sets of the rough cladding upper edge, the rough cladding lower edge, the rough fiber core upper edge and the rough fiber core lower edge and corresponding brightness values, substituting the shadow brightness values into each quadratic function, and solving to obtain the cladding upper edge, the cladding lower edge, the fiber core upper edge and the fiber core lower edge on each column;
s132: and performing quadratic curve fitting on the upper and lower point sets of the rough axis position and the corresponding brightness values, solving a symmetry axis of a corresponding unary quadratic function, and taking a pixel point corresponding to the symmetry axis as the axis position.
According to an embodiment of the present invention, in the step S2, the bending parameters include a shaft center offset, a left fiber cutting angle, a right fiber cutting angle, a fiber end face distance, and a fiber pushing overlap amount.
According to an embodiment of the present invention, the step S2 further includes: obtaining discharge parameters in the optical fiber test result, wherein the discharge parameters comprise: pre-melting discharge intensity, pre-melting discharge time, welding discharge intensity and welding discharge time; in step S3, the bending parameters, the discharge parameters, and the curvature parameters are used as preliminary arguments.
According to an embodiment of the present invention, the step S3, the step of selecting the optimized independent variable according to the correlation coefficient between the independent variable and the dependent variable includes: axis offset, left side fiber cutting angle, right side fiber cutting angle, fiber pushing overlap amount, maximum value of fiber core upper edge curvature, maximum value of fiber core lower edge curvature, and maximum value of axis curvature.
According to an embodiment of the present invention, in step S3, a part of the test data is obtained as training samples to perform linear regression calculation, and another part is obtained as test samples to perform regression model test.
According to an embodiment of the present invention, in step S4, the multiple linear regression model is constructed by:
where Loss represents the estimated Loss, x1Indicates the axial offset, x2Denotes the left fiber cut angle, x3Indicates the right fiber cut angle, x4Representing the amount of fibre advancing overlap, x5Representing the maximum value of curvature, x, of the upper edge of the core6Represents the maximum value of curvature, x, of the lower edge of the core7Denotes the maximum value of the curvature of the axis, β0Is a constant term, β1～β7Are regression coefficients.
The present invention also provides a system for estimating fusion loss of a ribbon fiber, comprising:
a curvature parameter calculation module: and analyzing the banded optical fiber image after fusion welding is completed, and calculating the curvature parameter of a certain distance range of the optical fiber fusion welding point, wherein the curvature parameter comprises: the maximum curvature of the upper edge of the cladding, the maximum curvature of the lower edge of the cladding, the maximum curvature of the upper edge of the fiber core, the maximum curvature of the lower edge of the fiber core and the maximum curvature of the axle center;
an influence parameter calculation module: performing bending parameters which can cause the optical fiber to be bent in the optical fiber test result;
an independent variable optimization module: taking the bending parameters and the curvature parameters as preliminary independent variables, taking actual loss of a test as a dependent variable, obtaining a plurality of test data as training samples, performing linear regression calculation, and screening out optimized independent variables according to correlation coefficients between the independent variables and the dependent variables;
a loss estimation module: and constructing a multiple linear regression model according to the actual loss and the optimization independent variable, and estimating the loss according to the multiple linear regression model.
After the technical scheme is adopted, compared with the prior art, the invention has the following beneficial effects:
the method breaks through the loss estimation method of the conventional optical fiber fusion splicer, introduces curvature parameters such as the maximum curvature of the upper edge of the cladding, the maximum curvature of the lower edge of the cladding, the maximum curvature of the upper edge of the fiber core, the maximum curvature of the lower edge of the fiber core and the maximum curvature of the axis, and part of fusion splicing parameters which can cause the bending of the optical fiber, analyzes and estimates the fusion splicing loss in multiple directions, can avoid the under-fitting phenomenon, and improves the accuracy of the loss estimation of the ribbon optical fiber fusion splicer;
the positions of the upper edge of the cladding, the lower edge of the cladding, the upper edge of the fiber core, the lower edge of the fiber core and the axis are roughly positioned and then precisely positioned, so that the positioning accuracy is improved; in the accurate positioning process, the upper and lower point sets of the roughly positioned position are selected to perform quadratic curve fitting, and an accurate position is obtained by solving a unitary quadratic function, so that the positioning is accurate to the sub-pixel level, and the positioning accuracy is improved.
Drawings
FIG. 1 is a flow chart illustrating a method for estimating fusion loss of a ribbon fiber according to an embodiment of the present invention;
FIGS. 2a-2e are schematic diagrams of structural feature types of the weld points;
FIG. 3a is a partial image of a 4-core ribbon fiber and a cut-out first core fiber according to one embodiment of the present invention;
FIG. 3b is a schematic diagram of a luminance waveform corresponding to a dotted line portion in the partial image of FIG. 3 a;
FIGS. 4a-4e are schematic diagrams of fitted curves of the upper and lower edges and the axial center of the cladding and core, in accordance with an embodiment of the present invention;
fig. 5 is a schematic diagram of an optical fiber information curve according to an embodiment of the invention.
Detailed Description
In order to make the aforementioned objects, features and advantages of the present invention comprehensible, embodiments accompanied with figures are described in detail below.
In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present invention. This invention may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein, but rather construed as limited to the embodiments set forth herein.
Referring to fig. 1, in one embodiment, a method for estimating fusion loss of a ribbon fiber includes the steps of:
s1: analyzing the banded optical fiber image after fusion splicing is completed, and calculating the curvature parameter of a certain distance range of an optical fiber fusion splicing point, wherein the curvature parameter comprises: the maximum curvature of the upper edge of the cladding, the maximum curvature of the lower edge of the cladding, the maximum curvature of the upper edge of the fiber core, the maximum curvature of the lower edge of the fiber core and the maximum curvature of the axle center;
s2: obtaining bending parameters which can cause the optical fiber to bend in the optical fiber test result;
s3: taking the bending parameters and the curvature parameters as initial independent variables, taking actual loss of a test as a dependent variable, obtaining a plurality of test data as training samples, performing linear regression calculation, and screening out optimized independent variables according to the correlation coefficients between the independent variables and the dependent variables;
s4: and constructing a multiple linear regression model according to the actual loss and the optimization independent variable, and estimating the loss according to the multiple linear regression model.
The method for estimating fusion loss of a ribbon fiber according to an embodiment of the present invention will be described in more detail below, but should not be construed as being limited thereto.
Referring to fig. 2a-2e, the structural characteristics of the fusion splice point for an optical fiber are generally of the type shown. In which fig. 2a shows a good fusion splice without any bending, fig. 2b shows that the upper edge of the optical fiber is leaned and the lower edge is leaned to make the optical fiber thin, fig. 2c shows that the upper and lower edges of the optical fiber are simultaneously leaned to cause the optical fiber fusion-splicing point to be bent upward, fig. 2d shows that the upper and lower edges of the optical fiber are simultaneously leaned to cause the optical fiber fusion-splicing point to be bent downward, and fig. 2e shows that the upper edge of the optical fiber is leaned and the lower edge is leaned to cause the optical fiber fusion-splicing point to be thick.
In the prior art, loss caused by bending of a welding point, loss which cannot be observed by a single image, and the like cannot be quantitatively analyzed and calculated. The effect of core bending on loss is significant, and thus the present invention incorporates bending factors into the loss estimation. The first problem is how to quantitatively characterize the bending factors, and in the embodiment of the invention, the curvatures of the positions of the upper edge of the cladding, the lower edge of the cladding, the upper edge of the fiber core, the lower edge of the fiber core and the axis near the fusion point of the optical fiber are analyzed and calculated so as to characterize the bending factors.
Step S1 is executed: analyzing the banded optical fiber image after fusion splicing is completed, and calculating the curvature parameter of a certain distance range of an optical fiber fusion splicing point, wherein the curvature parameter comprises: a maximum value of curvature of the upper edge of the cladding, a maximum value of curvature of the lower edge of the cladding, a maximum value of curvature of the upper edge of the core, a maximum value of curvature of the lower edge of the core, and a maximum value of curvature of the axial center. The certain distance range is, for example, 200 rows of pixel points near the optical fiber fusion point in the ribbon optical fiber image, and the optical fiber fusion point is a locatable position point in the ribbon optical fiber image, for example, a position of the tip of the electrode rod, which is not described herein again.
Before calculating curvature information of positions of the upper edge of the cladding, the lower edge of the cladding, the upper edge of the fiber core, the lower edge of the fiber core and the axis, accurate position information of the upper edge of the cladding, the lower edge of the cladding, the upper edge of the fiber core, the lower edge of the fiber core and the axis of each row of pixel points in a certain distance range near the optical fiber fusion point needs to be obtained, and therefore how to obtain the accurate position information is also a problem.
In one embodiment, step S1 may include the following steps S11-S15.
In S11, obtaining brightness values of pixel points on each row for each row within a certain distance range of the fusion point of the optical fiber in the ribbon optical fiber image; for example, 200 rows of pixel points are selected from the left and right of the optical fiber fusion point, and the brightness value of each pixel point on each row is obtained.
Because the ribbon fiber image is composed of pixel points, the positions of the upper and lower edges of the cladding, the upper and lower edges of the fiber core and the axis are a series of discrete data sequences. Fig. 3a shows a 4-core ribbon-like optical fiber image captured by the CCD, in which a brightness waveform corresponding to a dotted line portion in a partial image of the first core optical fiber taken out is shown in fig. 3b, the horizontal axis represents a vertical shift position, the vertical axis represents a brightness value, and the ShadowBright represents a shadow brightness value. The vertical offset position corresponding to the intersection point of the dashed line where the shadow brightness value is located and the brightness waveform can roughly represent the positions of the upper edge of the cladding, the upper edge of the fiber core, the lower edge of the fiber core and the lower edge of the cladding. Of course, the fiber images shown in the figures are merely schematic and the number is not limited to 4 cores.
Preferably, the shadow brightness value is a brightness mean value of a row of pixel points including a complete optical fiber in the ribbon optical fiber image when the ribbon optical fiber image is not welded. For example, the shadow brightness value is the brightness average value of the 100 th row of the optical fiber image when the optical fiber is not welded, because the end surface position of the optical fiber exceeds 100 rows of pixel points after the optical fiber is placed by the ribbon optical fiber welding machine, so that the 100 th row contains intact optical fiber information, and the brightness average value can well distinguish the edge positions of the cladding and the core of each core optical fiber in the ribbon optical fiber image.
In S12, each row of pixels is traversed from top to bottom or from bottom to top, and if the brightness values of a plurality of pixels that continuously appear after the brightness is suddenly changed are smaller or larger than the shadow brightness value, the pixels are initially determined as the upper edge of the rough cladding, the lower edge of the rough cladding, the upper edge of the rough fiber core, and the lower edge of the rough fiber core, and the rough axis position is calculated according to the brightness values of the pixels between the upper edge of the rough fiber core and the lower edge of the rough fiber core. Because the brightness data is stored in association with the pixel points, the corresponding brightness value can be read by traversing each pixel point, and the edge can be distinguished according to the brightness continuous characteristic of the adjacent pixel points.
Preferably, in the step S12, if the traversal is performed from top to bottom for each column of pixel points, the method includes the following steps:
s121 a: after the brightness value of a certain pixel point is changed from being larger than the brightness value of the shadow to being smaller than or equal to the brightness value of the shadow, the brightness values of a plurality of continuous points are smaller than the brightness value of the shadow, and then the points or one point of the points are used as the upper edge of the rough cladding; one of the points may be, for example, an intermediate point, although not limited thereto;
s122 a: after the brightness value of a certain pixel point is changed from being smaller than the brightness value of the shadow to being larger than or equal to the brightness value of the shadow, the brightness values of a plurality of continuous points are larger than the brightness value of the shadow, and then the points or one point of the points are used as the upper edge of the rough fiber core;
s123 a: after the brightness value of a certain pixel point is changed from being larger than the brightness value of the shadow to being smaller than or equal to the brightness value of the shadow, the brightness values of a plurality of continuous points are smaller than the brightness value of the shadow, and then the points or one point of the points are used as the lower edge of the rough fiber core;
s124 a: calculating the rough axis position by using a centroid method for the brightness value data of the pixel points between the upper edge of the rough fiber core and the lower edge of the rough fiber core;
s125 a: after the brightness value of a certain pixel point is changed from being smaller than the brightness value of the shadow to being larger than or equal to the brightness value of the shadow, the brightness values of a plurality of continuous points are larger than the brightness value of the shadow, and then the points or one point of the points are used as the lower edge of the rough cladding;
wherein the steps S124a and S125a are interchangeable;
if the traversal is performed from bottom to top for each row of pixel points, the method comprises the following steps:
s121 b: after the brightness value of a certain pixel point is changed from being larger than the brightness value of the shadow to being smaller than or equal to the brightness value of the shadow, the brightness values of a plurality of continuous points are smaller than the brightness value of the shadow, and then the points or one point of the points are used as the lower edge of the rough cladding;
s122 b: after the brightness value of a certain pixel point is changed from being smaller than the brightness value of the shadow to being larger than or equal to the brightness value of the shadow, the brightness values of a plurality of continuous points are larger than the brightness value of the shadow, and then the points or one point of the points are used as the lower edge of the rough fiber core;
s123 b: after the brightness value of a certain pixel point is changed from being larger than the brightness value of the shadow to being smaller than or equal to the brightness value of the shadow, the brightness values of a plurality of continuous points are smaller than the brightness value of the shadow, and then the points or one point of the points are used as the upper edge of the rough fiber core;
s124 b: calculating the rough axis position by using a centroid method for the brightness value data of the pixel points between the upper edge of the rough fiber core and the lower edge of the rough fiber core;
s125 b: after the brightness value of a certain pixel point is changed from being smaller than the brightness value of the shadow to being larger than or equal to the brightness value of the shadow, the brightness values of a plurality of continuous points are larger than the brightness value of the shadow, and then the points or one point of the points are used as the upper edge of the rough cladding;
wherein the steps S124b and S125b are interchangeable.
The mode of reading and judging in a traversal mode can enable the data reading and writing and processing efficiency to be higher, and the real-time performance of the loss estimation fed back to an operator by the optical fiber fusion splicer is improved.
Specifically, for a fused ribbon fiber image, traversing the brightness value of a certain column in the image to obtain the position information of the cladding and the fiber core of each core fiber in the certain column in the image, the steps are as follows:
1) the brightness value of 5 continuous points is less than the shadow brightness value, the position of the edge on the cladding is recorded, and 2) is executed;
2) the brightness value of 5 continuous points is greater than the shadow brightness value, the position of the upper edge of the fiber core is recorded, and 3) is executed;
3) the brightness value of 5 continuous points is less than the shadow brightness value, the lower edge position of the fiber core is recorded, and 4) is executed;
4) roughly calculating the position of the axis by using a centroid method for data between the upper edge and the lower edge of the fiber core, and executing step 5);
5) the brightness value of 5 continuous points is greater than the shadow brightness value, the position of the lower edge of the cladding is recorded, and 6) is executed;
6) and storing the optical fiber data of the current core, finishing if the current core is the 4 th core optical fiber, otherwise, re-executing 1), and calculating the next optical fiber data.
The above is an example of a 4-core ribbon optical fiber, and the rough calculation of the position of the edge of the ribbon optical fiber is not limited to this.
In order to improve the precision, when the discrete data sequence of the optical fiber fusion point is obtained, the invention also accurately calculates the rough position to the sub-pixel level.
In S13, for each row of pixels, performing sub-pixel level calculation according to the coarse cladding upper edge, the coarse cladding lower edge, the coarse fiber core upper edge, the coarse fiber core lower edge, and the coarse axis position to obtain the cladding upper edge, the cladding lower edge, the fiber core upper edge, the fiber core lower edge, and the axis position on each row.
Preferably, referring to fig. 4a-4e, the step S13 includes:
s131: respectively carrying out quadratic curve fitting on upper and lower point sets of the rough cladding upper edge, the rough cladding lower edge, the rough fiber core upper edge and the rough fiber core lower edge and corresponding brightness values, substituting the shadow brightness values into each quadratic function, and solving to obtain the cladding upper edge, the cladding lower edge, the fiber core upper edge and the fiber core lower edge on each column; the upper and lower point sets of the edge may be a plurality of points selected as the coarse cladding upper edge, the coarse cladding lower edge, the coarse core upper edge, the coarse core lower edge, or one point and its upper and lower consecutive points in the foregoing step S12;
s132: carrying out quadratic curve fitting on the upper and lower point sets of the rough axis position and the corresponding brightness values, solving a symmetry axis of a corresponding unary quadratic function, and taking a pixel point corresponding to the symmetry axis as the axis position; the upper and lower point sets at the axis position can be all pixel points between the upper edge of the fiber core and the lower edge of the fiber core.
The longitudinal gray distribution diagram of the single core of the ribbon fiber is analyzed to find that the distribution of discrete points at the upper edge and the lower edge of the cladding, the upper edge and the lower edge of the fiber core and the axis position is close to a unitary quadratic curve, so that part of the discrete points are respectively selected to carry out quadratic curve fitting, the accurate positions of the upper edge and the lower edge of the cladding, the upper edge and the lower edge of the fiber core and the axis position can be respectively calculated through the respectively fitted quadratic curves, and the positioning at the sub-pixel level is.
The horizontal axis of the 5 graphs shown in fig. 4a-4e represents the vertical offset position in units: and the vertical axis of the pixel (pix) represents the brightness value of the pixel point corresponding to the offset position. Points are discrete points taken from a column in the image that correspond to coarse edge or axis positions, and the curve is a univariate quadratic curve fitted by the discrete points.
Fig. 4a to 4d respectively indicate a point set of upper and lower continuous discrete points of the upper edge of the rough cladding, the lower edge of the rough cladding, the upper edge of the rough core, and the lower edge of the rough core, and a corresponding quadratic fit curve. The calculation of the precise position is to substitute the shadow brightness value into the fitted quadratic function, solve the equation, and select the value closest to the rough edge position in the solution of the two equations as the edge position value precise to the sub-pixel level. Fig. 4e shows several continuous discrete point sets and corresponding fitting curves at the upper and lower ends of the axis position, where the discrete points select all the pixel points between the upper edge of the coarse fiber core and the lower edge of the coarse fiber core, the calculation of the precise axis position is to solve the symmetry axis of the corresponding unitary quadratic function, and the value of the symmetry axis is used as the precise position of the optical fiber axis.
Since unnecessary errors should not be introduced when calculating the precise position, although the precise positions of the upper and lower edges of the fiber core are already calculated, it is still difficult to avoid the errors which may be carried, so that the separate curve fitting and calculation of the axis position is more reliable than the calculation directly by means of the precise values of the upper and lower edges of the fiber core.
The optical fiber information accurate to the sub-pixel level is obtained in the above way. The optical fiber information of 4 optical fibers of 100 rows and 200 rows on the left and right of the optical fiber fusion point can be sequentially obtained for analyzing and calculating the characteristic parameters participating in the estimated loss calculation.
In S14, discrete optical fiber information curves are drawn based on the positions of the upper cladding edge, the lower cladding edge, the upper core edge, the lower core edge, and the axial center on each column.
The precise positions of the upper and lower edges and the axis of the fiber core of 200 rows are calculated and stored, wherein the electrode rod is taken as the center, and the 100 rows are respectively arranged at the left and the right. The accurate optical fiber information of 200 columns near the first core optical fiber fusion splice point of fig. 3a is depicted in fig. 5. Wherein, from top to bottom, five curves respectively show: cladding lower edge, core lower edge, axis, core upper edge, cladding upper edge. The vertical axis represents the vertical offset of the fiber position information, and the horizontal axis represents 0-200 columns.
The discrete data of the 200 groups of cladding, core upper and lower edges and axial center positions may be gaussian smoothed, and then step S15 is performed.
In S15, the curvature of each discrete point in the optical fiber information curve is solved, and the maximum curvature is calculated, so as to obtain the maximum value of the curvature of the upper edge of the cladding, the maximum value of the curvature of the lower edge of the cladding, the maximum value of the curvature of the upper edge of the fiber core, the maximum value of the curvature of the lower edge of the fiber core, and the maximum value of the curvature of the axis. Preferably, the discrete data after the gaussian smoothing process may be solved for the curvature of each discrete point by using a difference method, and a maximum curvature value may be calculated and stored.
Next, step S2 is performed to obtain bending parameters that may cause the optical fiber to bend in the optical fiber test result. Preferably, in step S2, the bending parameters include axial offset, left fiber cutting angle, right fiber cutting angle, fiber end face distance, and fiber pushing overlap amount.
The fiber end face pitch is a distance between the fiber end faces on both sides, and the fiber advancing overlap amount is a distance that the fibers on both sides continue to advance at the time of fusion splicing after coming into contact (the pitch is eliminated). The bending parameters are factors which can cause the bending of the optical fiber fusion splice, and the parameters are not included in the prior art to estimate the loss, but the invention includes the parameters to estimate the influence of the optical fiber bending on the loss.
Step S3 is then executed: in the process of constructing a multiple linear regression model for estimating loss, the bending parameters and the curvature parameters are used as initial independent variables, actual loss of a test is used as a dependent variable, a plurality of test data are obtained to be used as training samples, linear regression calculation is carried out, and optimized independent variables are screened out according to the correlation coefficient between the independent variables and the dependent variables.
Preferably, in step S3, a part of the test data is obtained as a training sample to perform linear regression calculation, and another part of the test data is obtained as a test sample to perform regression model test, so that the multiple linear regression model is more accurate.
Preferably, the step S2 further includes: obtaining discharge parameters in the optical fiber test result, wherein the discharge parameters comprise: pre-melting discharge intensity, pre-melting discharge time, welding discharge intensity and welding discharge time; in step S3, a multiple linear regression model is constructed using the bending parameters, the discharge parameters, and the curvature parameters as preliminary independent variables.
Specifically, 300 groups of test data are obtained through a welding machine test in advance, wherein 250 groups of data are used as training samples to perform linear regression analysis, and the remaining 50 groups of test data are used as test samples to perform regression model test. Through correlation analysis and significance analysis, the finally screened part has larger correlation, and the correlation coefficients between independent variables and dependent variables of the part are shown in the following table (1):
watch (1)
Finally, in step S3, the optimized independent variables screened according to the correlation coefficients between the independent variables and the dependent variables include: axis offset, left side fiber cutting angle, right side fiber cutting angle, fiber pushing overlap amount, maximum value of fiber core upper edge curvature, maximum value of fiber core lower edge curvature, and maximum value of axis curvature.
The pre-fusion discharge intensity and the fusion discharge intensity cannot be quantitatively measured and accurately controlled, and therefore, the discharge parameters are finally removed from the factor domain.
Step S4 is then executed: and constructing a multiple linear regression model according to the actual loss and the optimization independent variable, and estimating the loss according to the multiple linear regression model.
In one embodiment, in step S4, the multiple linear regression model is constructed by:
where Loss represents the estimated Loss, x1Indicates the axial offset, x2Denotes the left fiber cut angle, x3Indicates the right fiber cut angle, x4Representing the amount of fibre advancing overlap, x5Representing the maximum value of curvature, x, of the upper edge of the core6Represents the maximum value of curvature, x, of the lower edge of the core7Denotes the maximum value of the curvature of the axis, β0Is a constant term, β1～β7Specifically, β1Is the parameter x of a multiple linear regression model1β2Is the parameter x of a multiple linear regression model2β3Is the parameter x of a multiple linear regression model3β4Is the parameter x of a multiple linear regression model4β5Is the parameter x of a multiple linear regression model5β6Is the parameter x of a multiple linear regression model6β7Is the parameter x of a multiple linear regression model7The regression coefficient of (2).
In order to simplify the calculation, all independent variables and dependent variables in the model can be normalized, and all variables are divided by their standard deviation after subtracting their mean value, so that a normalized regression model is obtained, wherein the regression coefficient is the normalized regression coefficient. For the real regression model of the estimated loss, the normalization process is as follows:
let t1＝x1，
And carrying out standardized conversion on the multiple linear regression model, wherein,
and
expressed as the variables Loss and t in the sample, respectivelykMean values of (1), SLoss and StkThen the variables Loss and t in the sample are represented respectivelykStandard deviation of (1), t1-t7As intermediate variables, Loss*Is a dependent variable, t, of Loss after normalized conversionk *Is tkNormalizing the transformed independent variables. Then, the regression model becomes:
Loss*＝β1 *t1 *+β2 *t2 *+β3 *t3 *+β4 *t4 *+β5 *t5 *+β6 *t6 *+β7 *t7 *
wherein, β1 *-β7 *Are normalized regression coefficients.
The results obtained by stepwise regression on 250 sets of regression sample data were as follows:
Loss*＝0.0877t1 *-0.0424t2 *+0.272t3 *+0.212t4 *+0.66t5 *-0.027t6 *+0.112t7 *
testing 50 groups of new sample data, referring to table (2), comparing the actual loss and the original predicted loss of the actually measured 50-core data with the predicted loss calculated by the regression model, wherein the sum of squares of residuals of the predicted loss and the actually measured loss of the 50 groups of test data in the embodiment of the invention is 0.0041; the sum of the square residuals SSE of the original predicted loss and the measured loss of the 50 sets of test data is 0.6141, which shows that the predicted loss of the embodiment of the present invention is better than the original predicted loss.
Watch (2)
The present invention also provides a system for estimating fusion loss of a ribbon fiber, comprising:
a curvature parameter calculation module: and analyzing the banded optical fiber image after fusion welding is completed, and calculating the curvature parameter of a certain distance range of the optical fiber fusion welding point, wherein the curvature parameter comprises: the maximum curvature of the upper edge of the cladding, the maximum curvature of the lower edge of the cladding, the maximum curvature of the upper edge of the fiber core, the maximum curvature of the lower edge of the fiber core and the maximum curvature of the axle center;
an influence parameter calculation module: performing bending parameters which can cause the optical fiber to be bent in the optical fiber test result;
an independent variable optimization module: taking the bending parameters and the curvature parameters as preliminary independent variables, taking actual loss of a test as a dependent variable, obtaining a plurality of test data as training samples, performing linear regression calculation, and screening out optimized independent variables according to correlation coefficients between the independent variables and the dependent variables;
a loss estimation module: and constructing a multiple linear regression model according to the actual loss and the optimization independent variable, and estimating the loss according to the multiple linear regression model.
For the details of the system for estimating fusion loss of a ribbon fiber according to the present invention, reference may be made to the description of the method for estimating fusion loss of a ribbon fiber in the foregoing embodiments, and the details are not repeated herein.
Although the present invention has been described with reference to the preferred embodiments, it is not intended to limit the scope of the claims, and those skilled in the art can make various changes and modifications without departing from the spirit and scope of the invention.
Claims (10)
1. A method for estimating fusion loss of a ribbon fiber, comprising the steps of:
s1: analyzing the banded optical fiber image after fusion splicing is completed, and calculating the curvature parameter of a certain distance range of an optical fiber fusion splicing point, wherein the curvature parameter comprises: the maximum curvature of the upper edge of the cladding, the maximum curvature of the lower edge of the cladding, the maximum curvature of the upper edge of the fiber core, the maximum curvature of the lower edge of the fiber core and the maximum curvature of the axle center;
s2: obtaining bending parameters which can cause the optical fiber to bend in the optical fiber test result;
s3: taking the bending parameters and the curvature parameters as initial independent variables, taking actual loss of a test as a dependent variable, obtaining a plurality of test data as training samples, performing linear regression calculation, and screening out optimized independent variables according to the correlation coefficients between the independent variables and the dependent variables;
s4: and constructing a multiple linear regression model according to the actual loss and the optimization independent variable, and estimating the loss according to the multiple linear regression model.
2. The fusion loss estimation method of a ribbon fiber according to claim 1, wherein the step S1 includes the steps of:
s11: aiming at each row of pixel points in a certain distance range of the optical fiber fusion point in the ribbon optical fiber image, calculating the brightness value on each row;
s12: traversing from top to bottom or from bottom to top for each row of pixel points, if the brightness values of a plurality of pixel points which continuously appear after sudden change of brightness are smaller than or larger than the shadow brightness value, initially judging as a rough cladding upper edge, a rough cladding lower edge, a rough fiber core upper edge and a rough fiber core lower edge, and calculating a rough axis position according to the pixel point brightness values between the rough fiber core upper edge and the rough fiber core lower edge; the shadow brightness value is the brightness mean value of a row of pixel points containing complete optical fibers in the ribbon optical fiber image when the ribbon optical fibers are not welded;
s13: aiming at each row of pixel points, performing sub-pixel level calculation according to the rough cladding upper edge, the rough cladding lower edge, the rough fiber core upper edge, the rough fiber core lower edge and the rough axle center position to obtain the cladding upper edge, the cladding lower edge, the fiber core upper edge, the fiber core lower edge and the axle center position on each row;
s14: drawing a discrete optical fiber information curve according to the positions of the upper edge of the cladding, the lower edge of the cladding, the upper edge of the fiber core, the lower edge of the fiber core and the axis of each column;
s15: and solving the curvature of each discrete point in the optical fiber information curve, and calculating the maximum curvature to obtain the maximum value of the curvature of the upper edge of the cladding, the maximum value of the curvature of the lower edge of the cladding, the maximum value of the curvature of the upper edge of the fiber core, the maximum value of the curvature of the lower edge of the fiber core and the maximum value of the curvature of the axis.
3. The method for estimating fusion loss of a ribbon fiber according to claim 2, wherein in step S12, if the traversal from top to bottom is performed for each row of pixels, the method comprises the following steps:
s121 a: after the brightness value of a certain pixel point is changed from being larger than the brightness value of the shadow to being smaller than or equal to the brightness value of the shadow, the brightness values of a plurality of continuous points are smaller than the brightness value of the shadow, and then the points or one point of the points are used as the upper edge of the rough cladding;
s122 a: after the brightness value of a certain pixel point is changed from being smaller than the brightness value of the shadow to being larger than or equal to the brightness value of the shadow, the brightness values of a plurality of continuous points are larger than the brightness value of the shadow, and then the points or one point of the points are used as the upper edge of the rough fiber core;
s123 a: after the brightness value of a certain pixel point is changed from being larger than the brightness value of the shadow to being smaller than or equal to the brightness value of the shadow, the brightness values of a plurality of continuous points are smaller than the brightness value of the shadow, and then the points or one point of the points are used as the lower edge of the rough fiber core;
s124 a: calculating the rough axis position by using a centroid method for the brightness value data of the pixel points between the upper edge of the rough fiber core and the lower edge of the rough fiber core;
s125 a: after the brightness value of a certain pixel point is changed from being smaller than the brightness value of the shadow to being larger than or equal to the brightness value of the shadow, the brightness values of a plurality of continuous points are larger than the brightness value of the shadow, and then the points or one point of the points are used as the lower edge of the rough cladding;
wherein the steps S124a and S125a are interchangeable;
if the traversal is performed from bottom to top for each row of pixel points, the method comprises the following steps:
s121 b: after the brightness value of a certain pixel point is changed from being larger than the brightness value of the shadow to being smaller than or equal to the brightness value of the shadow, the brightness values of a plurality of continuous points are smaller than the brightness value of the shadow, and then the points or one point of the points are used as the lower edge of the rough cladding;
s122 b: after the brightness value of a certain pixel point is changed from being smaller than the brightness value of the shadow to being larger than or equal to the brightness value of the shadow, the brightness values of a plurality of continuous points are larger than the brightness value of the shadow, and then the points or one point of the points are used as the lower edge of the rough fiber core;
s123 b: after the brightness value of a certain pixel point is changed from being larger than the brightness value of the shadow to being smaller than or equal to the brightness value of the shadow, the brightness values of a plurality of continuous points are smaller than the brightness value of the shadow, and then the points or one point of the points are used as the upper edge of the rough fiber core;
s124 b: calculating the rough axis position by using a centroid method for the brightness value data of the pixel points between the upper edge of the rough fiber core and the lower edge of the rough fiber core;
s125 b: after the brightness value of a certain pixel point is changed from being smaller than the brightness value of the shadow to being larger than or equal to the brightness value of the shadow, the brightness values of a plurality of continuous points are larger than the brightness value of the shadow, and then the points or one point of the points are used as the upper edge of the rough cladding;
wherein the steps S124b and S125b are interchangeable.
4. The fusion loss estimation method of a ribbon fiber according to claim 2, wherein the step S13 includes:
s131: respectively carrying out quadratic curve fitting on upper and lower point sets of the rough cladding upper edge, the rough cladding lower edge, the rough fiber core upper edge and the rough fiber core lower edge and corresponding brightness values, substituting the shadow brightness values into each quadratic function, and solving to obtain the cladding upper edge, the cladding lower edge, the fiber core upper edge and the fiber core lower edge on each column;
s132: and performing quadratic curve fitting on the upper and lower point sets of the rough axis position and the corresponding brightness values, solving a symmetry axis of a corresponding unary quadratic function, and taking a pixel point corresponding to the symmetry axis as the axis position.
5. The method for estimating fusion loss of a ribbon fiber according to claim 1, wherein in step S2, the bending parameters include axial center offset, left fiber cut angle, right fiber cut angle, fiber end face spacing, and fiber advance overlap.
6. The method for estimating fusion loss of a ribbon fiber according to claim 5, wherein the step S2 further comprises: obtaining discharge parameters in the optical fiber test result, wherein the discharge parameters comprise: pre-melting discharge intensity, pre-melting discharge time, welding discharge intensity and welding discharge time; in step S3, the bending parameters, the discharge parameters, and the curvature parameters are used as preliminary arguments.
7. The fusion loss estimation method for optical fiber ribbon according to claim 5 or 6, wherein the step S3 for optimizing the independent variable selected according to the correlation coefficient between the independent variable and the dependent variable includes: axis offset, left side fiber cutting angle, right side fiber cutting angle, fiber pushing overlap amount, maximum value of fiber core upper edge curvature, maximum value of fiber core lower edge curvature, and maximum value of axis curvature.
8. The method for estimating fusion loss of a ribbon fiber according to claim 7, wherein in step S3, a part of the test data is obtained as training samples and linear regression is performed, and another part is obtained as test samples and regression model is performed.
9. The method for estimating fusion loss of a ribbon fiber according to claim 7, wherein in step S4, the multivariate linear regression model is constructed by:
where Loss represents the estimated Loss, x1Indicates the axial offset, x2Denotes the left fiber cut angle, x3Indicates the right fiber cut angle, x4Representing the amount of fibre advancing overlap, x5Representing the maximum value of curvature, x, of the upper edge of the core6Represents the maximum value of curvature, x, of the lower edge of the core7Denotes the maximum value of the curvature of the axis, β0Is a constant term, β1～β7Are regression coefficients.
10. A ribbon fiber fusion splice loss estimation system, comprising:
a curvature parameter calculation module: and analyzing the banded optical fiber image after fusion welding is completed, and calculating the curvature parameter of a certain distance range of the optical fiber fusion welding point, wherein the curvature parameter comprises: the maximum curvature of the upper edge of the cladding, the maximum curvature of the lower edge of the cladding, the maximum curvature of the upper edge of the fiber core, the maximum curvature of the lower edge of the fiber core and the maximum curvature of the axle center;
an influence parameter calculation module: performing bending parameters which can cause the optical fiber to be bent in the optical fiber test result;
an independent variable optimization module: taking the bending parameters and the curvature parameters as preliminary independent variables, taking actual loss of a test as a dependent variable, obtaining a plurality of test data as training samples, performing linear regression calculation, and screening out optimized independent variables according to correlation coefficients between the independent variables and the dependent variables;
a loss estimation module: and constructing a multiple linear regression model according to the actual loss and the optimization independent variable, and estimating the loss according to the multiple linear regression model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711476168.XA CN108227077B (en) | 2017-12-29 | 2017-12-29 | Method and system for estimating fusion loss of ribbon fiber |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711476168.XA CN108227077B (en) | 2017-12-29 | 2017-12-29 | Method and system for estimating fusion loss of ribbon fiber |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108227077A CN108227077A (en) | 2018-06-29 |
| CN108227077B true CN108227077B (en) | 2020-03-24 |
Family
ID=62646146
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711476168.XA Active CN108227077B (en) | 2017-12-29 | 2017-12-29 | Method and system for estimating fusion loss of ribbon fiber |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108227077B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110738581A (en) * | 2019-07-25 | 2020-01-31 | 天津大学 | Campus comprehensive building power consumption rating method based on multiple linear regression |
| CN114459735B (en) * | 2020-11-10 | 2024-03-15 | 许继集团有限公司 | All-fiber current transformer optical path loss testing system and testing method |
| CN116912201B (en) * | 2023-07-13 | 2024-03-08 | 上海频准激光科技有限公司 | Optical fiber fusion quality prediction system |
| CN116912205B (en) * | 2023-07-13 | 2024-03-08 | 上海频准激光科技有限公司 | Optical fiber fusion quality prediction method based on neural network model |
| CN116912204B (en) * | 2023-07-13 | 2024-01-26 | 上海频准激光科技有限公司 | Treatment method for fusion splicing of optical fibers |
Family Cites Families (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| SE514476C2 (en) * | 1998-12-18 | 2001-02-26 | Ericsson Telefon Ab L M | Method for estimating loss in a joint made between two optical fibers |
| DE19927583A1 (en) * | 1999-06-16 | 2000-12-21 | Siemens Ag | Method for determining the attenuation of a splice connecting two optical waveguides |
| JP3664985B2 (en) * | 2001-02-16 | 2005-06-29 | 三菱電機株式会社 | Optical fiber fusion splicing device, heating amount adjusting method of optical fiber fusion splicing device, and optical fiber fusion splicing method |
| SE520076C2 (en) * | 2001-04-06 | 2003-05-20 | Ericsson Telefon Ab L M | Apparatus and method for automatically optimizing a splicing loss estimator of an optical fiber splicing machine and optical fiber splicing system and computer software |
| JP4108683B2 (en) * | 2005-01-28 | 2008-06-25 | 三菱電機株式会社 | Optical fiber fusion splice loss estimation apparatus and optical fiber fusion splice loss estimation method |
| JP2006300755A (en) * | 2005-04-21 | 2006-11-02 | Fujikura Ltd | Device and method for estimating connection loss in optical fiber |
| JP4702846B2 (en) * | 2006-05-09 | 2011-06-15 | 日本電信電話株式会社 | Connection loss judgment method at optical fiber connection point |
| JP5863190B2 (en) * | 2012-11-27 | 2016-02-16 | 富士通株式会社 | Image analysis apparatus and optical fiber fusion splicing system |
| CN104504244B (en) * | 2014-12-03 | 2017-06-06 | 中国第一汽车股份有限公司无锡油泵油嘴研究所 | Fuel injector sprays amount estimation method |
-
2017
- 2017-12-29 CN CN201711476168.XA patent/CN108227077B/en active Active
Non-Patent Citations (2)
| Title |
|---|
| 《影响光纤熔接损耗的因素及解决方法》;程蕊岚，刘继芳，马琳;《电子科技》;20090315;第22卷(第5期);50-53页 * |
| 单模光纤熔接损耗分析及自动对准与熔接系统;黄尚廉，石文江，李平，文玉梅;《仪器仪表学报》;19940531;第15卷(第2期);第124-129页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108227077A (en) | 2018-06-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108227077B (en) | Method and system for estimating fusion loss of ribbon fiber | |
| US9582872B2 (en) | Optical film defect detection method and system thereof | |
| CN111054782B (en) | Wide and thick plate shape detection device and method | |
| EP2048472B1 (en) | Lens shape measurement device and method, method of producing eyeglass lens, and method of producing eyeglasses | |
| EP2477469A2 (en) | Suitability determination method for determination standard value and method for specifying optimum value thereof, inspection system for substrate on which components are mounted, simulation method at production site, and simulation system | |
| US20120194729A1 (en) | Fast Auto-Focus in Imaging | |
| SE523806C2 (en) | Method and apparatus for aligning the polarization shafts of fiber ends in two optical polarization preserving fibers with each other | |
| CN108414094B (en) | Labview-based laser beam analysis system and method | |
| JP2004506255A (en) | How to process, analyze and display market information | |
| CN107770520A (en) | A kind of line-scan digital camera auxilary focusing method and device | |
| CN113888472A (en) | Detection method and equipment for consumer electronics defects | |
| CN109059789A (en) | Cable pitch online test method based on machine vision | |
| CN116518868B (en) | Deformation measurement method, device, equipment and storage medium based on artificial intelligence | |
| CN113705298A (en) | Image acquisition method and device, computer equipment and storage medium | |
| CN112434600A (en) | Unmanned aerial vehicle inspection method and system | |
| EP1860397B1 (en) | Method for measuring non-circularity at core portion of optical fiber parent material | |
| CN111899297B (en) | Method for extracting center of light stripe of line structure | |
| US8998511B2 (en) | Cladding alignment for fusion splicing optical fibers | |
| CN112816187B (en) | Quality judgment method for laser spots | |
| CN113066063B (en) | Optical fiber to-be-aligned end image processing method and optical fiber self-adaptive alignment method | |
| US20220270235A1 (en) | Fracture surface analysis apparatus and fracture surface analysis method | |
| US20210279655A1 (en) | Optimization of human-centric processes | |
| JP2001305372A (en) | Optical fiber core measuring device, fusion splicer, focal position setting method for splicer, and optical fiber identifying method | |
| Filipenko et al. | Improving of photonic crystal fibers connection quality using positioning by the autoconvolution method | |
| US20140258068A1 (en) | System and method for sequential count visual indicator |
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 | ||
| PE01 | Entry into force of the registration of the contract for pledge of patent right | ||
| PE01 | Entry into force of the registration of the contract for pledge of patent right |
Denomination of invention: Method and system for estimating fusion loss of ribbon optical fiber Effective date of registration: 20210810 Granted publication date: 20200324 Pledgee: Weihai Branch of Shanghai Pudong Development Bank Co.,Ltd. Pledgor: INNO INSTRUMENT (CHINA) Inc. Registration number: Y2021980007538 |