CN109191527B - Alignment method and device based on minimum distance deviation - Google Patents
Alignment method and device based on minimum distance deviation Download PDFInfo
- Publication number
- CN109191527B CN109191527B CN201811358794.3A CN201811358794A CN109191527B CN 109191527 B CN109191527 B CN 109191527B CN 201811358794 A CN201811358794 A CN 201811358794A CN 109191527 B CN109191527 B CN 109191527B
- Authority
- CN
- China
- Prior art keywords
- distance
- platform
- product
- image
- straight line
- 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
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/70—Determining position or orientation of objects or cameras
- G06T7/73—Determining position or orientation of objects or cameras using feature-based methods
- G06T7/74—Determining position or orientation of objects or cameras using feature-based methods involving reference images or patches
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10004—Still image; Photographic image
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
- Length Measuring Devices By Optical Means (AREA)
- Image Processing (AREA)
Abstract
The embodiment of the application shows a contraposition method and a contraposition device based on minimum distance deviation, firstly, the contraposition method based on minimum distance deviation directly starts with the specification of a contraposition product as a target, and a calculated contraposition result is more in line with the evaluation standard of production; secondly, the alignment method disclosed by the embodiment of the application can effectively solve the non-centered alignment problem and is suitable for alignment scenes of irregular-shaped products in actual production.
Description
Technical Field
The invention relates to the technical field of machine vision, in particular to a contraposition method and a contraposition device based on minimum distance deviation.
Background
In the industrial fields of electronic semiconductors, touch screens, solar energy, automobiles, parts and the like, two or more parts are often required to be aligned and mounted. The precision of the alignment installation directly determines the quality of products, and in the process of modern industrial production, a machine vision alignment system is generally used for realizing the alignment of parts.
Fig. 1 is a schematic structural diagram of a machine vision alignment system commonly used at present, which includes a target platform 110, a target platform 120, a target image capturing device 130, and a target image capturing device 140; a target object 150 is placed on a target platform 110, a real-time object 160 is placed on the target platform 120, the target object 150 and the real-time object 160 are both provided with alignment marks 170, in an actual industrial production process, for example, the real-time object 160 may be a mobile phone liquid crystal display, the target object 150 may be a backlight module, and the machine vision alignment system is used for realizing the attachment of the two; the target image capturing device 130 and the object image capturing device 140 are used for capturing the alignment mark 170. The process of counterpoint laminating includes: the object image acquisition device 140 acquires and captures the alignment mark 170 of the real-time object 160, and calculates the object platform coordinates of the alignment mark 170 in the real-time object 160 according to the predetermined mapping relationship between the object image plane and the object platform plane; similarly, the target image capturing device 130 captures the alignment mark 170 of the target object 150, and calculates the target platform coordinates of the alignment mark 170 in the target object 150 according to the predetermined mapping relationship between the target image plane and the target platform plane; the object platform 120 sends the real-time object 160 to the position right above the target platform 110, and adjusts the position of the object platform 120 according to the object platform coordinate and the target platform coordinate of the corresponding alignment mark 170, and the real-time object 160 is also adjusted in position under the driving of the object platform 120, thereby completing the attachment of the real-time object 160 and the target object 150. In the alignment fitting process, the mapping relation between the object image plane and the object platform plane and between the target image plane and the target platform plane is determined in advance as the basis of alignment fitting, which is called calibration; the accuracy of the calibration is the key to the alignment and fitting accuracy.
At present, a calibration method of a commonly used machine vision alignment system is mainly a nine-point circular target calibration method, and comprises the following steps: placing the circular target with 9 determined positions on the target platform 110, and acquiring the object image coordinates of the circle center of the circular target through the object image acquisition device 140; manually operating the object platform 120 to move, enabling the reference point of the object platform 120 to sequentially pass through the circle centers of 9 circular targets, observing and determining that the reference point is coincident with the circle center, and then recording the object platform coordinate of the reference point; and calculating and determining the mapping relation between the object image plane and the object platform plane according to the object image coordinate difference adjacent to the circle center and the object platform coordinate difference of the reference point, and finishing calibration. However, as products diversify, alignment scenarios are also more complex. The traditional centering alignment mode is not suitable for the scene when the products with irregular shapes or materials are aligned and the specific distance after alignment is required to be the standard distance specification.
Disclosure of Invention
The invention aims to provide a contraposition method and a contraposition device based on minimum distance deviation, and aims to solve the problems of poor precision and poor applicability of a calibration method in the prior art.
The first aspect of the embodiments of the present application shows a positioning method based on minimized distance deviation, where the method includes:
the object image acquisition device acquires a reference image on a target platform, establishes a coordinate system on the reference image, and acquires coordinates of a plurality of characteristic quantities of a product on the object platform, wherein the reference image comprises: a reference characteristic amount, and a reference straight line;
calculating a distance array of the feature quantity and a reference straight line according to the coordinates of the feature quantity and the reference straight line, and calculating a reference distance array according to the reference feature quantity and the reference straight line;
controlling the object platform to move for presetting the deviation, acquiring the coordinates of a plurality of characteristic quantities of a product on the object platform again, and calculating the distance array of the characteristic quantities and the reference straight line;
and calculating the distance deviation square sum of the distance array and the reference distance array, and determining a target characteristic quantity according to the distance deviation square sum, wherein the target characteristic quantity is an alignment result.
Optionally, the step of controlling the movement of the object platform by the preset deviation amount comprises:
control the object platform to move a fixed distance along X, Y, the coordinate direction of the object platform, and/or rotate a fixed angle.
Alternatively, the feature quantities include: characteristic points, and/or, straight lines.
Optionally, the step of obtaining coordinates of a plurality of feature quantities of the product on the object platform includes:
acquiring an image of a product on the object platform;
and positioning the characteristic quantity of the product according to the image of the product, wherein the characteristic quantity comprises a centered contraposition point or a centered straight line of the product.
Optionally, the step of calculating a sum of squares of distance deviations of the distance array and the reference distance array, and determining the target feature amount according to the sum of squares of distance deviations includes:
respectively calculating the distance deviation square sum of the distance array and the reference distance array;
and performing reverse iterative calculation and solving through a gradient descent algorithm, wherein if the sum of squares of the distance deviations reaches a threshold value or reaches the maximum iteration times, the characteristic quantity result at the moment is the target characteristic quantity.
Optionally, the step of calculating a sum of squares of distance deviations of the distance array, and determining the target feature quantity according to the sum of squares of distance deviations, includes:
respectively calculating the distance deviation square sum of the distance array and the reference distance array;
the feature quantity that yields the minimum distance deviation squared and the corresponding is determined as the target feature quantity.
Optionally, the step of obtaining the coordinates of the plurality of feature quantities of the product on the object platform specifically includes:
and obtaining the coordinates of the object characteristic quantity through an image positioning algorithm.
A second aspect of the embodiments of the present application shows an alignment apparatus based on minimizing distance deviation, the apparatus including:
the acquisition unit is used for initializing the characteristic quantity, and the object image acquisition device acquires the coordinates of a reference image on the target platform and acquires the coordinates of a plurality of characteristic quantities of a product on the object platform;
a calculation unit, configured to acquire a reference image on a target platform by using an object image acquisition device, establish a coordinate system on the reference image, and acquire coordinates of a plurality of feature quantities of a product on the object platform, where the reference image includes: a reference characteristic amount, and a reference straight line;
an array calculation unit configured to calculate a distance array of the feature amount and a reference straight line based on the coordinates of the feature amount and the reference straight line, and calculate a reference distance array based on the reference feature amount and the reference straight line;
the control unit is used for controlling the object platform to move for presetting the deviation, acquiring the coordinates of a plurality of characteristic quantities of a product on the object platform again, and calculating the distance array of the characteristic quantities and the reference straight line;
and the determining unit is used for calculating the distance deviation square sum of the distance array and the reference distance array, and determining a target characteristic quantity according to the distance deviation square sum, wherein the target characteristic quantity is an alignment result.
Optionally, the determining unit includes:
the first calculating unit is used for calculating the sum of squares of the distance deviations of the distance array and the reference distance array respectively;
and the solving unit is used for carrying out reverse iterative calculation and solving through a gradient descent algorithm, and if the sum of squares of the distance deviations reaches a threshold value or reaches the maximum iteration number, the characteristic quantity result at the moment is the target characteristic quantity.
Optionally, the determining unit includes:
the second calculation unit is used for calculating the sum of squares of the distance deviations of the distance array and the reference distance array respectively;
and a first determination unit for determining the feature quantity which generates the least distance deviation squared sum as the target feature quantity.
According to the technical scheme, the alignment method and the alignment device based on the minimum distance deviation are shown in the embodiment of the application, firstly, the alignment method based on the minimum distance deviation directly starts with the specification of an alignment product as a target, and the calculated alignment result is more in line with the evaluation standard of production; secondly, the alignment method disclosed by the embodiment of the application can effectively solve the non-centered alignment problem and is suitable for alignment scenes of irregular-shaped products in actual production. The contraposition method mainly comprises the following steps: and product feature quantity positioning, feature quantity offset iteration and distance deviation square sum calculation. The image characteristic point line positioning is mainly to obtain the necessary characteristic points of the algorithm according to the image positioning algorithm and/or the coordinate representation of the straight line. The characteristic quantity migration iteration means that in the optimization distance alignment, the characteristic quantity can be continuously changed in an iteration mode, and after each iteration, the characteristic point needs to be migrated according to the latest characteristic quantity, so that the distance between the point and the corresponding straight line is continuously close to the reference distance. And (4) performing minimum distance difference calculation, namely performing iterative feature quantity calculation, and when the deviation between all the shifted points and the corresponding linear distance is small enough to the corresponding reference distance, determining the feature quantity at the moment as a final alignment result.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings needed in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings without creative efforts.
FIG. 1 is a schematic diagram of a conventional machine vision alignment system in accordance with a preferred embodiment;
FIG. 2 is a block diagram illustrating a method for alignment based on minimizing distance deviation according to a preferred embodiment;
FIG. 3 is a schematic illustration of a calibration reference position of the subject platform in accordance with a preferred embodiment;
FIG. 4 is a diagram illustrating a reference image, and coordinates of the reference image, in accordance with a preferred embodiment;
FIG. 5 is a diagram illustrating a θ coordinate in accordance with a preferred embodiment;
FIG. 6 is a diagram illustrating a product on an object platform in accordance with a preferred embodiment;
FIG. 7 is a schematic diagram illustrating a product after rotation in accordance with a preferred embodiment;
FIG. 8 is a schematic diagram illustrating product translation in accordance with a preferred embodiment
FIG. 9 is a schematic diagram illustrating a product after conversion in accordance with yet another preferred embodiment.
Detailed Description
In the following, the technical solutions shown in the embodiments of the present invention will be clearly and completely described with reference to the drawings in the technical solutions shown in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Machine vision is a technique for automatically acquiring analysis images to obtain data describing a scene or controlling some action. In the modern industrial automation production process, machine vision is becoming a key technology for improving production efficiency and ensuring product quality. The machine vision-based alignment system is a system for performing non-contact two-dimensional or three-dimensional coordinate alignment by using an image acquisition Device such as a Charge Coupled Device (CCD) camera and the like as an image sensor and comprehensively applying technologies such as image processing, motion algorithms and the like, is based on optics, integrates an electronic technology, a computer technology and an image processing technology, and has the advantages of high precision, high efficiency, high automation degree, low manufacturing cost and the like. The image processing technology is a key technology of a machine vision alignment system, comprises image smoothing processing, image sharpening, an image contrast enhancement algorithm and the like, and obtains data such as contour information, position information and the like of an acquired object by analyzing color features and the like of the acquired object image.
The machine vision alignment system is widely applied to the preparation and detection processes of LCDs and semiconductors, and generally comprises a CCD camera, a three-dimensional mobile platform, a stepping motor control unit, an image processing unit, a computer system control unit, a result output and feedback unit and the like. For example, in the assembling process of the liquid crystal display of the mobile phone, the machine vision alignment system acquires, filters, separates and identifies alignment marks of the liquid crystal display of the mobile phone and the backlight module to obtain the deviation amount between image mark positions, so that the object platform is controlled to move, and the accurate alignment of the liquid crystal display of the mobile phone and the backlight module is completed. In the actual production process, the machine vision alignment system generally designs the object platform into a platform capable of performing accurate movement with multiple degrees of freedom according to the integrated factors such as the overall mechanical design, the motion flow, the cost control and the like, and the target platform is generally a fixed stationary platform.
In order to ensure the accuracy of the machine vision alignment system, the machine vision alignment system must be calibrated before the product is produced. And in the calibration process, a coordinate mapping relation between an image plane shot by the CCD camera and the plane of the object platform and between the image plane shot by the CCD camera and the plane of the target platform is established. The accuracy degree of calibration directly determines the alignment accuracy, so that the quality of a product is influenced, and the calibration is only needed when the CCD camera is installed for the first time or the position of the CCD camera changes in the actual production process. According to the calibration method and device for the machine vision alignment system, provided by the embodiment of the invention, the object platform is controlled to move, the object image coordinate change of the reference point on the object platform and the object platform coordinate change are obtained, and the calibration of the object platform is accurately calculated; then acquiring a plurality of mark point images of a real-time object arranged on the object platform, and calculating to obtain the coordinate deviation of the object platform between adjacent mark points according to the calibration result of the object platform; and finally, attaching the real-time object to a target platform, acquiring the coordinate deviation of the target image between adjacent mark points, and calculating and determining the calibration result of the target platform by combining the coordinate deviation of the object platform of the mark points to finish the calibration of the target platform. In the whole calibration process, the calibration of the static target platform can be completed by calibrating the movable target platform, and the method has extremely high precision and applicability.
Referring to fig. 2, an embodiment of the present application illustrates an alignment method based on minimizing distance deviation, where the method includes:
s101, an object image acquisition device acquires a reference image on a target platform, a coordinate system is established on the reference image, and coordinates of a plurality of characteristic quantities of a product on the object platform are acquired, wherein the reference image comprises: a reference feature quantity, and a reference straight line.
In the technical solution shown in the embodiment of the present invention, the object image capturing device may be an industrial CCD camera or an industrial CMOS (Complementary Metal Oxide Semiconductor) camera, and parameters such as a focal length and an aperture of the camera are adjusted according to a size of the object platform and a distance from the camera to the object platform, so that a product that can clearly obtain a reference point of the object platform in a field of view of the camera. Fig. 3 is a schematic diagram of a calibration reference position of an object platform according to an embodiment of the present invention, where 11 is the object platform and 12 is a product; by adjusting the camera, the image of the product 12 on the object platform 11 can be clearly obtained in the field of view of the camera, and the pair of products 11 is ensured to be located at a position close to the center in the field of view. Of course, the position of the product 12 on the object platform 11 may be any point on the object platform, as long as the product 12 is fixed with respect to the object platform 11 and is easily acquired by the object image acquisition device, and the size and the position of the reference point of the object platform 11 are not limited in the embodiment of the present invention, and those skilled in the art may arbitrarily select the reference point according to actual needs.
In the technical solution shown in the embodiment of the present invention, the image of the product acquired by the reference image acquiring apparatus and the reference image are both planar images.
Then, a coordinate system of a reference image is established on the reference image, as shown in fig. 4, the coordinate system of the reference image may be established with one vertex angle of the field-of-view reference image as an origin, an X axis and a Y axis of the object image coordinate system are respectively parallel to outer edge lines of the reference image, the X axis is positive to the right, and the Y axis is positive to the up; the reference image coordinate system may also be established with any vertex of the reference image as an origin. Typically, the X-axis and the Y-axis of the reference image coordinate system are parallel to the edges of the object platform, respectively, the X-axis being positive to the right and the Y-axis being positive to the up.
It should be noted that, the present application shows a technical solution in real time, and in addition to the translation of the product, the present application also relates to the rotation of the product during the alignment process, and therefore, in the technical solution shown in the embodiment of the present application, the reference image coordinate system further includes the θ coordinate. Referring to fig. 5, the object platform is determined to rotate around the origin as the rotation center, and the clockwise rotation is positive and the counterclockwise rotation is negative. Of course, those skilled in the art may establish the object image coordinate system and the object platform coordinate system with any predetermined coordinate system origin and corresponding coordinate axis directions, and the coordinate systems are not limited to the rectangular coordinate system, and may also be other coordinate systems, such as a polar coordinate system.
The method for acquiring the coordinates of the characteristic quantities of the products on the object platform comprises the following steps:
obtaining an image of a product on the object platform, and obtaining a characteristic quantity of the image through an image positioning algorithm, wherein the characteristic quantity comprises: characteristic points, and/or, straight lines.
And obtaining the characteristic points and straight lines of the product image through an image positioning algorithm. For a triangular product, the vertex angle of the triangle is a characteristic point, and the side of the triangle is a straight line. In practical application, the characteristic quantity of the product can be positioned according to the shape of the product and an image positioning method.
S102, calculating a distance array of the characteristic quantity and a reference straight line according to the coordinates of the characteristic quantity and the reference straight line, and calculating a reference distance array according to the reference characteristic quantity and the reference straight line;
firstly, calculating a reference distance array according to the reference characteristic quantity and a reference straight line;
the quasi-distance array of the product can be obtained according to the production specification;
a reference feature of a reference image may be obtained by an image localization algorithm, where the reference feature includes: a reference feature point and/or a reference straight line.
Referring to fig. 4, the triangle in the drawing is the reference image, the rectangle on the right side, and the sides of the rectangle on the bottom side near the triangle are the reference lines, respectively, and the reference distance array of the reference image is { L1 ', L2 ', L3 ' }accordingto the production specification.
Then, a distance array of the feature amount and a reference straight line is calculated from the coordinates of the feature amount and the reference straight line.
Specifically, referring to fig. 6, fig. 6 illustrates an initial state of a triangle object on an object platform, where a, B, and C are three feature quantities of the object, respectively.
Constructing a coordinate system according to the reference image in advance, and taking an X axis and a Y axis of the coordinate system as a first reference line and a second reference line respectively;
in the coordinate system, the coordinates of the three points of the feature amounts a, B, and C in the initial state are (X10, Y10), (X20, Y20), (X30, Y30), respectively.
The correspondence between the feature quantity and the reference line may be that a shortest distance between the feature quantity and the reference line is determined as the reference line corresponding to the feature quantity.
In fig. 6, the reference line corresponding to B is the X axis, and the reference line corresponding to a is the Y axis.
The distance between the feature a and the reference straight line (Y axis) is: x10 is L10, and the distance between the feature value B and the reference straight line (X axis) is: y20 is L20, and the distance between the feature value C and the reference straight line (X axis) is: y30 ═ L30.
The distance array of the corresponding product is { L10, L20, L30 }.
The feature quantities are initialized. Taking the initial translation length and the rotation angle of the feature point, and generally taking the value of 0;
s103, controlling the object platform to move for a preset deviation amount, acquiring coordinates of a plurality of characteristic quantities of a product on the object platform again, and calculating a distance array of the characteristic quantities and a reference straight line;
and controlling the object platform to move for a fixed distance along the X, Y and theta directions of the object platform, and acquiring the coordinates of a plurality of characteristic quantities of the product on the object platform again.
As shown in fig. 7, the object is rotated by θ, and the coordinates of a plurality of feature quantities of the product on the object platform are obtained again, and the coordinates of the feature quantities a1, B1, and C1 are (X11, Y11), (X21, Y21), (X31, and Y31), respectively.
The distance between the feature a1 and the reference straight line (Y axis) is: x11 is L11, and the distance between the feature amount B1 and the reference straight line (X axis) is: y21 is L21, and the distance between the feature value C1 and the reference straight line (X axis) is: y31 ═ L31.
The second distance array is { L11, L21, L31}, where the second distance array corresponds to a feature quantity of θ.
Referring to fig. 8, which is a schematic diagram illustrating the positions of the reference points after the object platform moves a fixed distance along the X and Y directions, respectively, in the technical solution shown in the embodiment of the present invention, the object platform moves a fixed distance Δ Xd along the negative direction of the X axis of the object platform,
the object platform moves a fixed distance Δ Yd in the positive direction of the Y-axis of the object platform, and the coordinates of the corresponding feature quantities a3, B3, and C3 are (X12, Y12), (X22, Y22), (X32, Y32), respectively.
The distance between the feature a2 and the reference straight line (Y axis) is: x12 is L12, and the distance between the feature amount B1 and the reference straight line (X axis) is: y21 is L22, and the distance between the feature value C1 and the reference straight line (X axis) is: y31 ═ L32.
S104, calculating the distance deviation square sum of the distance array and the reference distance array, and determining a target characteristic quantity according to the distance deviation square sum, wherein the target characteristic quantity is an alignment result.
The reference distance arrays are { L1 ', L2 ', L3 ' };
the first set { L10, L20, L30 }.
The second set of distance arrays is { L11, L21, L31 };
the calculation formula for optimizing the distance deviation can be expressed as:
the distance deviation square sum of the distance array and the reference distance array can be respectively calculated;
and performing reverse iterative calculation and solving through a gradient descent algorithm, wherein if the sum of squares of the distance deviations reaches a threshold value or reaches the maximum iteration times, the characteristic quantity result at the moment is the target characteristic quantity.
Or respectively calculating the distance deviation square sum of the distance array and the reference distance array;
the feature quantity that yields the minimum distance deviation squared and the corresponding is determined as the target feature quantity.
The rotation angle θ is the target feature quantity.
Preferably, a point participating in centering alignment is added to the straight line to make compatibility with a positioning scene with one side being standard distance alignment and the other two sides being centering alignment.
Specifically, an image of a product on the object platform is acquired;
and positioning the characteristic quantity of the product according to the image of the product, wherein the characteristic quantity comprises a centered contraposition point or a centered straight line of the product.
Referring to fig. 9, L4 and L5 are actual distances involved in left-right centering alignment, and after centering alignment, the two distances become equal and equal to L'; and L6 participates in the standard distance alignment, which is close to the standard distance L6' after alignment.
A second aspect of the embodiments of the present application shows an alignment apparatus based on minimizing distance deviation, the apparatus including:
the acquisition unit is used for initializing the characteristic quantity, and the object image acquisition device acquires the coordinates of a reference image on the target platform and acquires the coordinates of a plurality of characteristic quantities of a product on the object platform;
a calculation unit, configured to acquire a reference image on a target platform by using an object image acquisition device, establish a coordinate system on the reference image, and acquire coordinates of a plurality of feature quantities of a product on the object platform, where the reference image includes: a reference characteristic amount, and a reference straight line;
an array calculation unit configured to calculate a distance array of the feature amount and a reference straight line based on the coordinates of the feature amount and the reference straight line, and calculate a reference distance array based on the reference feature amount and the reference straight line;
the control unit is used for controlling the object platform to move for presetting the deviation, acquiring the coordinates of a plurality of characteristic quantities of a product on the object platform again, and calculating the distance array of the characteristic quantities and the reference straight line;
and the determining unit is used for calculating the distance deviation square sum of the distance array and the reference distance array, and determining a target characteristic quantity according to the distance deviation square sum, wherein the target characteristic quantity is an alignment result.
Optionally, the determining unit includes:
the first calculating unit is used for calculating the sum of squares of the distance deviations of the distance array and the reference distance array respectively;
and the solving unit is used for carrying out reverse iterative calculation and solving through a gradient descent algorithm, and if the sum of squares of the distance deviations reaches a threshold value or reaches the maximum iteration number, the characteristic quantity result at the moment is the target characteristic quantity.
Optionally, the determining unit includes:
the second calculation unit is used for calculating the sum of squares of the distance deviations of the distance array and the reference distance array respectively;
a first determination unit for determining the feature quantity generating the least distance deviation squared sum as the target feature quantity
According to the technical scheme, the alignment method and the alignment device based on the minimum distance deviation are shown in the embodiment of the application, firstly, the alignment method based on the minimum distance deviation directly starts with the specification of an alignment product as a target, and the calculated alignment result is more in line with the evaluation standard of production; secondly, the alignment method disclosed by the embodiment of the application can effectively solve the non-centered alignment problem and is suitable for alignment scenes of irregular-shaped products in actual production. The contraposition method mainly comprises the following steps: and product feature quantity positioning, feature quantity offset iteration and distance deviation square sum calculation. The image characteristic point line positioning is mainly to obtain the necessary characteristic points of the algorithm according to the image positioning algorithm and/or the coordinate representation of the straight line. The characteristic quantity migration iteration means that in the optimization distance alignment, the characteristic quantity can be continuously changed in an iteration mode, and after each iteration, the characteristic point needs to be migrated according to the latest characteristic quantity, so that the distance between the point and the corresponding straight line is continuously close to the reference distance. And (4) performing minimum distance difference calculation, namely performing iterative feature quantity calculation, and when the deviation between all the shifted points and the corresponding linear distance is small enough to the corresponding reference distance, determining the feature quantity at the moment as a final alignment result.
Other embodiments of the invention will be apparent to those skilled in the art from consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the invention following, in general, the principles of the invention and including such departures from the present disclosure as come within known or customary practice within the art to which the invention pertains. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the invention being indicated by the following claims.
It will be understood that the invention is not limited to the precise arrangements described above and shown in the drawings and that various modifications and changes may be made without departing from the scope thereof. The scope of the invention is limited only by the appended claims.
Claims (6)
1. A contraposition method based on minimized distance deviation is characterized by comprising the following steps:
the method comprises the following steps that an object image acquisition device acquires a reference image on a target platform, a coordinate system is established on the reference image, and coordinates of a plurality of characteristic quantities of a product on the object platform are acquired, wherein the reference image comprises: a reference characteristic amount, and a reference straight line;
calculating a distance array { L10, L20 … … Ln0} of the feature quantity and a reference straight line according to the coordinates of the feature quantity and the reference straight line, and calculating a reference distance array { L1, L2 … … Ln } according to the reference feature quantity and the reference straight line;
controlling the object platform to move for a preset deviation amount, acquiring coordinates of a plurality of characteristic quantities of a product on the object platform again, and calculating a distance array { L11, L21 … … Ln1} of the characteristic quantities and a reference straight line;
respectively calculating the distance deviation square sum of { L10, L20 … … Ln0} and the { L1, L2 … … Ln } and the distance deviation square sum of { L11, L21 … … Ln1} and the { L1, L2 … … Ln };
the feature quantity that yields the minimum distance deviation squared and the corresponding is determined as the target feature quantity.
2. The method of claim 1, wherein the step of controlling the movement of the subject platform by a preset offset amount comprises:
the object platform is controlled to move a fixed distance in the X, Y, theta coordinate direction of the object platform and/or to rotate a fixed angle.
3. The method according to claim 1, characterized in that the characteristic quantities comprise: characteristic points, and/or, straight lines.
4. The method of claim 3, wherein the step of obtaining coordinates of a plurality of feature quantities of the product on the subject platform comprises:
acquiring an image of a product on the object platform;
and positioning the characteristic quantity of the product according to the image of the product, wherein the characteristic quantity comprises a centered contraposition point or a centered straight line of the product.
5. The method according to any one of claims 1 to 4, wherein the step of obtaining the coordinates of the plurality of feature quantities of the product on the object platform is specifically:
and obtaining the coordinates of the object characteristic quantity through an image positioning algorithm.
6. An alignment apparatus based on minimizing distance deviation, the apparatus comprising:
the acquisition unit is used for initializing the characteristic quantity, and the object image acquisition device acquires the coordinates of a reference image on the target platform and acquires the coordinates of a plurality of characteristic quantities of a product on the object platform;
the calculation unit is used for acquiring a reference image on a target platform by an object image acquisition device, establishing a coordinate system on the reference image, and acquiring coordinates of a plurality of characteristic quantities of a product on the object platform, wherein the reference image comprises: a reference characteristic amount, and a reference straight line;
an array calculating unit for calculating an array { L10, L20 … … Ln0} of distances between the feature quantity and a reference straight line based on the coordinates of the feature quantity and the reference straight line, and calculating an array { L1, L2 … … Ln } of reference distances based on the reference feature quantity and the reference straight line;
the control unit is used for controlling the object platform to move for a preset deviation amount, acquiring the coordinates of a plurality of characteristic quantities of the product on the object platform again, and calculating a distance array { L11, L21 … … Ln1} of the characteristic quantities and a reference straight line;
and a determination unit which respectively calculates the distance deviation square sum of { L10, L20 … … Ln0} and { L1, L2 … … Ln } and the distance deviation square sum of { L11, L21 … … Ln1} and { L1, L2 … … Ln }, and determines the feature quantity corresponding to the minimum distance deviation square sum as the target feature quantity.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811358794.3A CN109191527B (en) | 2018-11-15 | 2018-11-15 | Alignment method and device based on minimum distance deviation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811358794.3A CN109191527B (en) | 2018-11-15 | 2018-11-15 | Alignment method and device based on minimum distance deviation |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109191527A CN109191527A (en) | 2019-01-11 |
| CN109191527B true CN109191527B (en) | 2021-06-11 |
Family
ID=64939460
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811358794.3A Active CN109191527B (en) | 2018-11-15 | 2018-11-15 | Alignment method and device based on minimum distance deviation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109191527B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110928235B (en) * | 2019-11-18 | 2021-04-06 | 广东利元亨智能装备股份有限公司 | Workpiece bonding method and device, electronic equipment and workpiece bonding system |
| CN112845115A (en) * | 2020-12-31 | 2021-05-28 | 华芯智造微电子(重庆)有限公司 | Efficient automatic sorting method |
| CN112845187B (en) * | 2020-12-31 | 2023-01-10 | 华芯智造微电子(重庆)有限公司 | Automatic sorting machine |
| CN112976559B (en) * | 2021-02-05 | 2022-11-25 | 广东科升智能装备有限公司 | Alignment method and film pasting method based on linear array CCD |
| CN115355822B (en) * | 2022-10-19 | 2023-01-17 | 成都新西旺自动化科技有限公司 | Special-shaped alignment calculation method and system |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101680756A (en) * | 2008-02-12 | 2010-03-24 | 松下电器产业株式会社 | Compound eye imaging device, distance measurement device, parallax calculation method and distance measurement method |
| CN102037800A (en) * | 2008-03-25 | 2011-04-27 | 伊利诺斯工具制品有限公司 | Method and apparatus for dispensing material on a substrate |
| CN102183205A (en) * | 2011-01-19 | 2011-09-14 | 北京航空航天大学 | Method for matching optimal assembly poses of large-sized parts |
| CN105427288A (en) * | 2015-11-10 | 2016-03-23 | 凌云光技术集团有限责任公司 | Calibration method and device of machine vision alignment system |
| CN106127722A (en) * | 2016-05-03 | 2016-11-16 | 深圳视觉龙智能传感器有限公司 | The demarcation of polyphaser and para-position applying method |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| SE530854C2 (en) * | 2005-12-30 | 2008-09-30 | Ericsson Telefon Ab L M | Alignment of optical fibers in their jointing |
-
2018
- 2018-11-15 CN CN201811358794.3A patent/CN109191527B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101680756A (en) * | 2008-02-12 | 2010-03-24 | 松下电器产业株式会社 | Compound eye imaging device, distance measurement device, parallax calculation method and distance measurement method |
| CN102037800A (en) * | 2008-03-25 | 2011-04-27 | 伊利诺斯工具制品有限公司 | Method and apparatus for dispensing material on a substrate |
| CN102183205A (en) * | 2011-01-19 | 2011-09-14 | 北京航空航天大学 | Method for matching optimal assembly poses of large-sized parts |
| CN105427288A (en) * | 2015-11-10 | 2016-03-23 | 凌云光技术集团有限责任公司 | Calibration method and device of machine vision alignment system |
| CN106127722A (en) * | 2016-05-03 | 2016-11-16 | 深圳视觉龙智能传感器有限公司 | The demarcation of polyphaser and para-position applying method |
Non-Patent Citations (3)
| Title |
|---|
| A Line Feature Matching Technique Based on an Eigenvector Approach;Sang Ho Park 等;《Computer Vision and Image Understanding》;20000331;第77卷(第3期);263-283 * |
| 基于多传感器的大口径器件自动对准策略;卢金燕 等;《自动化学报》;20151031;第41卷(第10期);1711-1722 * |
| 视觉检测技术在矫直机在线检测系统中的应用;于海霞 等;《大连工业大学学报》;20150531;第34卷(第3期);221-224 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109191527A (en) | 2019-01-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109191527B (en) | Alignment method and device based on minimum distance deviation | |
| CN107014312B (en) | A kind of integral calibrating method of mirror-vibrating line laser structured light three-dimension measuring system | |
| CN111536902B (en) | Galvanometer scanning system calibration method based on double checkerboards | |
| US10690492B2 (en) | Structural light parameter calibration device and method based on front-coating plane mirror | |
| CN108571971B (en) | AGV visual positioning system and method | |
| CN109612390B (en) | Large-size workpiece automatic measuring system based on machine vision | |
| CN109767476B (en) | Automatic focusing binocular camera calibration and depth calculation method | |
| CN109859272B (en) | Automatic focusing binocular camera calibration method and device | |
| CN105716542B (en) | A kind of three-dimensional data joining method based on flexible characteristic point | |
| CN108717715A (en) | A kind of line-structured light vision system automatic calibration method for arc welding robot | |
| CN113674345B (en) | Two-dimensional pixel-level three-dimensional positioning system and positioning method | |
| CN105066884A (en) | Robot tail end positioning deviation correction method and system | |
| CN107339935B (en) | Target space intersection measuring method for full-view scanning measuring system | |
| CN105427288A (en) | Calibration method and device of machine vision alignment system | |
| CN113724337B (en) | Camera dynamic external parameter calibration method and device without depending on tripod head angle | |
| CN107534715B (en) | Camera production method and advanced driving assistance system | |
| CN112258583B (en) | Distortion calibration method for close-range image based on equal distortion partition | |
| CN112658643B (en) | Connector assembly method | |
| CN104167001B (en) | Large-visual-field camera calibration method based on orthogonal compensation | |
| CN108805940B (en) | Method for tracking and positioning zoom camera in zooming process | |
| CN114001651A (en) | Large-scale long and thin cylinder type component pose in-situ measurement method based on binocular vision measurement and prior detection data | |
| CN113155084A (en) | Binocular vision distance measuring device and method based on laser cross standard line assistance | |
| CN111028298B (en) | Convergent binocular system for rigid coordinate system space transformation calibration | |
| WO2020113978A1 (en) | Method for calculating center position of hole located on plane | |
| CN114485479B (en) | Structured light scanning and measuring method and system based on binocular camera and inertial navigation |
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 | ||
| CB02 | Change of applicant information | ||
| CB02 | Change of applicant information |
Address after: 100094 Beijing city Haidian District Cui Hunan loop 13 Hospital No. 7 Building 7 room 701 Applicant after: Lingyunguang Technology Co., Ltd Address before: 100094 Beijing city Haidian District Cui Hunan loop 13 Hospital No. 7 Building 7 room 701 Applicant before: LUSTER LIGHTTECH GROUP Co.,Ltd. |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant |