CN103810458A - Image recognition method - Google Patents
Image recognition method Download PDFInfo
- Publication number
- CN103810458A CN103810458A CN201210439246.XA CN201210439246A CN103810458A CN 103810458 A CN103810458 A CN 103810458A CN 201210439246 A CN201210439246 A CN 201210439246A CN 103810458 A CN103810458 A CN 103810458A
- Authority
- CN
- China
- Prior art keywords
- gray
- area
- ordinate
- pixel
- point
- 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.)
- Pending
Links
Images
Landscapes
- Image Analysis (AREA)
Abstract
The invention aims at providing an image recognition method. According to the image recognition method, welding spots for welding silver wire on the surface of a silver foil in an acquired image are recognized. The acquired image comprises at least one first region corresponding to the silver foil and at least one second region corresponding to the silver wire. The image recognition method comprises the following steps of acquiring the gray value of each pixel in the first region and the second region, acquiring characteristic points of the first region according to the gray value of each pixel in the first region, and seeking the welding points in the second region according to the gray value of each pixel in the second region and the characteristic points. In comparison with the prior art, recognition is carried out according to the gray difference between the silver foil and the silver wire, the position of the silver wire is sought around mass points of the silver foil, the welding spots are finally determined, and the positions of the welding spots are then conveniently, quickly and automatically recognized.
Description
?
Technical field
The present invention relates to field of image recognition, relate in particular to a kind of visible sensation image processing process, for realizing the automatic welding of solar array.
Background technology
Domestic space substantially all adopts artificial welding with the parts welding of solar array, adopt artificial locator meams to utilize resistance spot welding to weld silver-colored wire on the silver foil of the sun wing windsurfing back side, aim at after solder joint control operation box mobile welding structure again by human eye, each solder joint need repetition coarse adjustment, aiming, fine setting, examination to press down to finely tune again this process could accurately move to solder joint directly over.
This method efficiency consuming time is low, and the welding that completes solder joint on a silver foil approximately needs the time of 2 minutes, has affected greatly manufacturing schedule.While, artificial long-time welding was easy to be tired, so welding quality can not get fine guarantee because solar array surface wire, component number are many.
Therefore, industry needs a kind of picture recognition module for solar array robot welding system, to can be fast and accurate for positioning, can efficiently complete the reliable welding of solar array surface electronic components and parts.
Summary of the invention
In order to solve the artificial positioning welding inefficiency of solar array, problem that reliability is low, take automatic locator meams butt welding point to position.Due to the stochastic distribution of solder joint, can not provide in advance position (coordinate) information accurately, must adopt the coordinate position of machine vision butt welding point to identify.The invention provides image recognition technology, operating personnel do not need butt welding point repeatedly to debug, aim at, and have improved efficiency and the reliability of operation.
According to an aspect of the present invention, a kind of image-recognizing method is provided, it identifies silver-colored wire bonds at the lip-deep solder joint of silver foil in the collection image through gathering, and described collection image comprises at least one first area corresponding to described silver foil and at least one second area corresponding to described silver-colored wire.Described image-recognizing method comprises the steps: that (a) obtains the gray-scale value of each pixel in described first area and described second area; (b) gray-scale value based on each pixel in described first area, obtains the unique point of described first area; And (c) based on described unique point, in described second area, look for described solder joint according to the gray-scale value of each pixel in described second area.
Preferably, described step (a) also comprises: (A) gray-scale value of the each pixel to described first area and described second area carries out greyscale transformation, to increase the difference between the gray-scale value of each pixel in the gray-scale value of each pixel in described first area and described second area, and, step (A) comprises the gray-scale value that reduces each pixel in described first area, and increases the gray-scale value of each pixel in described second area.
Preferably, described step (b) comprising: (b1), according to the first predetermined gray-scale value, determine the elemental area of described first area, thereby determine the surface area of described first area; And (b2) obtain the barycenter of the surface area of described first area, as described unique point.
Preferably, described step (b1) comprising: (b11) described collection image is carried out to Region Segmentation; (b12) carry out region growing from the initial pixel of each cut zone, wherein said region growing criterion is, in the time that the gray-scale value of pixel exceedes predetermined growing threshold, to carry out region growing, when the gray-scale value of pixel is during lower than predetermined growing threshold, stop growing; And (b13) screen the region that described growth obtains, and the selection result is carried out to area statistics, to determine the surface area of described first area.
Preferably, in described step (c), describedly comprise take the horizontal ordinate point of described unique point or ordinate point as benchmark based on described unique point.
Preferably, in described step (c), take described horizontal ordinate point as benchmark in the situation that, within the scope of ordinate traversal, carry out ordinate traversal along axis of ordinates, the ordinate place pixel around more a certain ordinate in predetermined ordinate scope (M1) and the gray-scale value of described a certain ordinate place pixel; If the difference of gray-scale value exceedes setting threshold, represent that described a certain ordinate place pixel has gray scale sudden change, and this ordinate and the described horizontal ordinate point as benchmark definite place pixel be catastrophe point; Described catastrophe point is calculated, wherein, in the time that the counting of described catastrophe point arrives a certain predetermined value, corresponding this catastrophe point ordinate point and described horizontal ordinate point are defined as to the coordinate points of described solder joint.
Preferably, in described step (c), take described ordinate point as benchmark in the situation that, within the scope of ordinate traversal, carry out horizontal ordinate traversal along abscissa axis, the horizontal ordinate place pixel around more a certain horizontal ordinate in predetermined ordinate scope (M2) and the gray-scale value of described a certain horizontal ordinate place pixel; If the difference of gray-scale value exceedes setting threshold, represent that described a certain horizontal ordinate place pixel has gray scale sudden change, and this horizontal ordinate and the described ordinate point as benchmark definite place pixel be catastrophe point; Described catastrophe point is calculated, wherein, in the time that the counting of described catastrophe point arrives a certain predetermined value, corresponding this catastrophe point horizontal ordinate point and the described point of the ordinate as benchmark are defined as to the coordinate points of described solder joint.
Of the present invention is a kind of welding system of solar array more on the one hand, comprising: image capture module, and it needs the silver-colored wire of welding and the image of silver foil for gathering; Picture recognition module, it receives the collection image that described image capture module gathers, and identifies silver-colored wire bonds at the lip-deep solder joint of silver foil, and described picture recognition module is as aforementioned pattern recognition device; Welding mechanism, it welds described silver-colored wire and described silver foil at described solder joint place; And motion-control module, the solder joint that it identifies according to described picture recognition module, controls described welding mechanism and moves to described solder joint position, completes the welding of described solder joint.
Compared with prior art, gray difference according to the image-recognizing method of the embodiment of the present invention based between silver foil and silver-colored wire is identified, around the particle of silver foil, look for guide line position, and finally determine solder joint, can automatically identify easily and quickly thus bond pad locations.
By reference to the accompanying drawings, according to below illustrate that by example the description of purport of the present invention can know other aspects of the present invention and advantage.
Accompanying drawing explanation
By reading the detailed description that non-limiting example is done of doing with reference to the following drawings, it is more obvious that other features, objects and advantages of the present invention will become:
Fig. 1 is the process flow diagram according to image-recognizing method of the present invention;
The collection image that Fig. 2 shows to identify;
Fig. 3 is the histogram of the gray-scale value of described collection image;
Fig. 4 be after greyscale transform process through gather image;
Fig. 5 is that wherein bond pad locations is the position at the little black fork place on wire in figure through the image of identification;
Fig. 6 shows the identification of image-recognizing method for variety classes silver foil; And
Fig. 7 is the calcspar according to pattern recognition device of the present invention.
In accompanying drawing, same or analogous Reference numeral represents same or analogous parts.
Embodiment
Referring to the accompanying drawing that the embodiment of the present invention is shown, the present invention below will be described in more detail.But the present invention can be with many multi-form realizations, and should not be construed as the restriction of the embodiment being subject in this proposition.On the contrary, it is abundant and complete open in order to reach proposing these embodiment, and makes those skilled in the art understand scope of the present invention completely.In these accompanying drawings, for clarity sake, may amplify size and the relative size in layer and region.
Refer now to Fig. 1, describe in detail according to image-recognizing method of the present invention.
According to the kind image-recognizing method of the embodiment of the present invention, it identifies silver-colored wire bonds at the lip-deep solder joint of silver foil in the collection image through gathering, and described collection image comprises at least one first area corresponding to described silver foil and at least one second area corresponding to described silver-colored wire.
As shown in Figure 1, according in the image-pickup method of the embodiment of the present invention,
At step S101, obtain the gray-scale value of each pixel in described first area and described second area.
In the present embodiment, can, by described collection image being carried out to the embodiment statistics of grey level histogram, obtain the gray-scale value of each pixel in described first area and described second area.Fig. 2 shows described collection image, and it shows the overlapping silver foil and the silver-colored wire that need identification.As shown in Figure 2, silver foil and silver-colored wire gray scale are variant, and silver foil surface pixels grey value profile is more even, and silver foil surface pixels point gray-scale value is greater than silver-colored wire gray-scale value.
Fig. 3 is the histogram of the gray-scale value of described collection image.For adding up the intensity profile situation of whole picture, first described collection image is carried out to the histogram calculation of gray scale, thereby obtain Fig. 3.As can be seen from Figure 3, gray-scale value is concentrated and is distributed in 30-40,120-140, these three regions of 145-165.By the gray-scale value of background area, silver-colored wire, silver foil is extracted, known background area gray-scale value is mainly distributed as 30-40, and silver-colored wire gray-scale value is mainly distributed in 120-140, and the grey value profile of silver foil is at 145-165.As can be seen here, there is larger difference in silver foil, silver-colored wire gray scale and background gray levels, and the gray scale of silver-colored wire and silver foil exists less difference.
Owing to being that gray difference based between silver foil and silver-colored wire is identified according to the image-recognizing method of the embodiment of the present invention, therefore, in preferred embodiment, the replacement of obtaining as gray-scale value or in addition, also can carry out the linear transformation of gray scale to the described first area of described collection image and second area, finally obtain the image that contrast strengthens, strengthened the gray difference between silver foil and silver-colored wire, be convenient to the identification in later stage.
In the present embodiment, the gray-scale value of the each pixel to described first area and described second area carries out greyscale transformation, to increase the difference between the gray-scale value of each pixel in the gray-scale value of each pixel in described first area and described second area.
In the present embodiment, by reducing the gray-scale value of each pixel in described first area, and increase the gray-scale value of each pixel in described second area, to increase the difference between the gray-scale value of each pixel in the gray-scale value of each pixel in described first area and described second area.
Below in detail the processing of greyscale transformation will be described.
Fig. 4 be after greyscale transform process through gather image.
Setting linear greyscale transformation function is, wherein f
afor the slope of linear function, f
bfor linear function is at the intercept of y axle, D
afor the gray scale of input picture, D
brepresent the gray scale of output image, the pretreated Main Function of image is the gray difference strengthening between silver foil and silver-colored wire, takes to improve silver foil gray-scale value, and the method that reduces the gray-scale value of silver-colored wire strengthens both difference before.Set a f
bvalue, D
b1represent the gray scale of silver foil output image, D
b2the gray scale that represents silver-colored wire output image, Simultaneous Equations is as follows:
Solve,
At f
acodomain scope select a f
avalue, according to f
a, f
bvalue size, carries out the linear transformation of gray scale to former collection image, and once linear conversion is complete, then carries out iteration according to formula, carries out secondary linear transformation.
Individual in preferred embodiment, can improve contrast to the complete image of gray scale linear transformation, set a f
athe value of > 1, makes f
b=0, the gray scale of integral image is increased, image is brighter.As can be seen from Figure 4, through the pretreated picture of image, background picture gray scale reduces, and silver foil and silver-colored wire seem brighter, and the gray difference between silver foil and silver-colored wire is more obvious.
In step S103, based on the gray-scale value of each pixel in described first area, obtain the unique point of described first area.
In the present embodiment, according to the first predetermined gray-scale value, determine the elemental area of described first area, thereby determine the surface area of described first area.
Particularly, first, described collection image is carried out to Region Segmentation.After described collection image is divided into multiple regions, then determine the elemental area of described first area, thereby determine the surface area of described first area.
Then, carry out region growing from the initial pixel of each cut zone, wherein said region growing criterion is, in the time that the gray-scale value of pixel exceedes predetermined growing threshold, carries out region growing, when the gray-scale value of pixel is during lower than predetermined growing threshold, stops growing.In the present embodiment, selecting growing point grey scale pixel value is 50.In other words, the gray-scale value that selective area growth criterion is pixel exceedes 50, carries out region growing, stops growing lower than 50.
Finally, screen the region that described growth obtains, and the selection result is carried out to area statistics, to determine the surface area of described first area.In the present embodiment, because the area of silver foil region is larger, substantially, more than 10000, therefore determine that the condition threshold value of screening area is 10000, filter out the surface area at silver foil place.
Because silver-colored wire is generally in about silver foil central point, and barycenter overlaps substantially with silver foil central point, therefore take the barycenter of silver foil as benchmark.That is, obtain the barycenter of the surface area of described first area, as described unique point.
In step S105, based on described unique point, in described second area, look for described solder joint according to the gray-scale value of each pixel in described second area.As previously mentioned, because silver-colored wire is generally in about silver foil central point, and barycenter overlaps substantially with silver foil central point, therefore allows horizontal stroke, ordinate travel through in scope around barycenter take the barycenter of silver foil as benchmark, searching solder joint.
In the present embodiment, describedly comprise take the horizontal ordinate point of described unique point or ordinate point as benchmark based on described unique point.
Like this, take described horizontal ordinate point as benchmark in the situation that, within the scope of ordinate traversal, carry out ordinate traversal along axis of ordinates, the ordinate place pixel around more a certain ordinate in predetermined ordinate scope M1 and the gray-scale value of described a certain ordinate place pixel; If the difference of gray-scale value exceedes setting threshold, represent that described a certain ordinate place pixel has gray scale sudden change, and this ordinate and the described horizontal ordinate point as benchmark definite place pixel be catastrophe point; Described catastrophe point is calculated, wherein, in the time that the counting of described catastrophe point arrives a certain predetermined value, corresponding this catastrophe point ordinate point and described horizontal ordinate point are defined as to the coordinate points of described solder joint.
Or, take described ordinate point as benchmark in the situation that, within the scope of ordinate traversal, carry out horizontal ordinate traversal along abscissa axis, the horizontal ordinate place pixel around more a certain horizontal ordinate in predetermined ordinate scope N1 and the gray-scale value of described a certain horizontal ordinate place pixel; If the difference of gray-scale value exceedes setting threshold, represent that described a certain horizontal ordinate place pixel has gray scale sudden change, and this horizontal ordinate and the described ordinate point as benchmark definite place pixel be catastrophe point; Described catastrophe point is calculated, wherein, in the time that the counting of described catastrophe point arrives a certain predetermined value, corresponding this catastrophe point horizontal ordinate point and the described point of the ordinate as benchmark are defined as to the coordinate points of described solder joint.
Specifically, for horizontal silver foil, fixing barycenter horizontal ordinate, allows ordinate from traversing than the position of the little N1 of a silver foil ordinate maximal value pixel wide than the position of the many N1 of a silver foil ordinate minimum value pixel wide.In the present embodiment, N1 can set as required.Then, setting a suitable pixel wide is ruler, and determined pixel width is M1, in the time that ordinate travels through silver foil, compares with the gray-scale value of ordinate point and the gray-scale value of this ordinate within the scope of ordinate fore-and-aft distance M1.In the time that both gray scale differences exceed setting threshold, represent that this point has the sudden change of gray scale, the number of statistics catastrophe point, if catastrophe point exceedes a certain number of, think that this ordinate is solder joint position ordinate, is no more than a certain number of, allow ordinate then travel through silver foil, until obtain satisfactory point.The ordinate of trying to achieve adds that barycenter horizontal ordinate obtains solder joint coordinate on silver-colored wire.
Similarly, perpendicular recognition methods of putting silver-colored wire on silver foil is similar to horizontal silver-colored wire.Fixing barycenter ordinate, allows horizontal ordinate from traversing than the position of the little N2 of a silver foil horizontal ordinate maximal value pixel wide than the position of the many N2 of a silver foil horizontal ordinate minimum value pixel wide.In the present embodiment, N2 can set as required.Then setting a suitable pixel wide is ruler, determined pixel width is M2, in the time that horizontal ordinate travels through silver foil, compare with gray-scale value and this horizontal ordinate gray-scale value of the horizontal ordinate point within the scope of horizontal ordinate lateral separation M2, gray scale difference exceedes setting threshold between the two, represents that this point has the sudden change of gray scale.The number of statistics catastrophe point, a certain number of if catastrophe point exceedes, represent that this horizontal ordinate is solder joint position horizontal ordinate, be no more than a certain number ofly, allow horizontal ordinate then travel through silver foil, until obtain satisfactory point.The horizontal ordinate of trying to achieve adds that barycenter ordinate obtains silver-colored wire solder joint coordinate.
In preferred embodiment, for improving bonding speed, reduce operation complexity, take the single job identification all silver foil solder joint coordinates on image that gather., can be to comprising horizontal silver foil, perpendicular put silver foil, gather the incomplete silver foil of image and identify.
Fig. 5 is figure that Fig. 2 image acquisition is arrived through greyscale transformation, image recognition and the final solder joint identification figure obtaining, and bond pad locations is the position at the little black fork place on wire in figure, and Fig. 6 has shown the identification of picture recognition module for variety classes silver foil.
Referring now to Fig. 7 describes according to pattern recognition device of the present invention, it identifies silver-colored wire bonds at the lip-deep solder joint of silver foil in the collection image through gathering, and described collection figure comprises at least one first area corresponding to described silver foil and at least one second area corresponding to described silver-colored wire.
As shown in Figure 7, described pattern recognition device comprises gray-scale value acquiring unit, and it obtains the gray-scale value of each pixel in described first area and described second area; Feature extraction unit, its gray-scale value based on each pixel in described first area, obtains the unique point of described first area; And solder joint determining unit, it looks for described solder joint according to the gray-scale value of each pixel in described second area in described second area.
In the present embodiment, the gray-scale value of the each pixel of described gray-scale value acquiring unit to described first area and described second area carries out greyscale transformation, to strengthen the difference between the gray-scale value of each pixel in the gray-scale value of each pixel in described first area and described second area.In addition, described gray-scale value acquiring unit reduces the gray-scale value of each pixel in described first area, and increases the gray-scale value of each pixel in described second area.
In the present embodiment, described feature extraction unit, according to the first predetermined gray-scale value, is determined the elemental area of described first area, thereby determines the surface area of described first area.And described feature extraction unit obtains the barycenter of the surface area of described first area, as described unique point.
In the present embodiment, described feature extraction unit is carried out Region Segmentation to described collection image, and carry out region growing from the initial pixel of each cut zone, wherein said region growing criterion is, in the time that the gray-scale value of pixel exceedes predetermined growing threshold, carry out region growing, when the gray-scale value of pixel is during lower than predetermined growing threshold, stop growing.In addition, described feature extraction unit is screened the region that described growth obtains, and the selection result is carried out to area statistics, to determine the surface area of described first area.
In the present embodiment, the described of described solder joint determining unit comprises take the horizontal ordinate point of described unique point or ordinate point as benchmark based on described unique point.
In the present embodiment, in described solder joint determining unit, take described horizontal ordinate point as benchmark in the situation that, within the scope of ordinate traversal, carry out ordinate traversal along axis of ordinates, the ordinate place pixel around more a certain ordinate in predetermined ordinate scope (M1) and the gray-scale value of described a certain ordinate place pixel; If the difference of gray-scale value exceedes setting threshold, represent that described a certain ordinate place pixel has gray scale sudden change, and this ordinate and the described horizontal ordinate point as benchmark definite place pixel be catastrophe point; Described catastrophe point is calculated, wherein, in the time that the counting of described catastrophe point arrives a certain predetermined value, corresponding this catastrophe point ordinate point and described horizontal ordinate point are defined as to the coordinate points of described solder joint.
In the present embodiment, in described solder joint determining unit, take described ordinate point as benchmark in the situation that, within the scope of ordinate traversal, carry out horizontal ordinate traversal along abscissa axis, the horizontal ordinate place pixel around more a certain horizontal ordinate in predetermined ordinate scope (N1) and the gray-scale value of described a certain horizontal ordinate place pixel; If the difference of gray-scale value exceedes setting threshold, represent that described a certain horizontal ordinate place pixel has gray scale sudden change, and this horizontal ordinate and the described ordinate point as benchmark definite place pixel be catastrophe point; Described catastrophe point is calculated, wherein, in the time that the counting of described catastrophe point arrives a certain predetermined value, corresponding this catastrophe point horizontal ordinate point and the described point of the ordinate as benchmark are defined as to the coordinate points of described solder joint.
According to the attainable function of the operation of the pattern recognition device of the embodiment of the present invention and its and corresponding according to the image-recognizing method of the embodiment of the present invention, then this does not carry out too much description.
Now describe according to the robot welding system of the solar array of the embodiment of the present invention.Described robot welding system comprises image capture module, and it needs the silver-colored wire of welding and the image of silver foil for gathering; Picture recognition module, it receives the collection image that described image capture module gathers, and identifies silver-colored wire bonds at the lip-deep solder joint of silver foil; Welding mechanism, it welds described silver-colored wire and described silver foil at described solder joint place; And motion-control module, the solder joint that it identifies according to described picture recognition module, controls described welding mechanism and moves to described solder joint position, completes the welding of described solder joint.
In the present embodiment, described image capture module comprises low angle light source, to utilize described low angle light source to throw light on to described silver-colored wire and silver foil in the time gathering.Because silver-colored wire and silver foil surface color are all silvery white, difference is very little, but silver-colored wire is on silver foil surface, and both are highly inconsistent, are the silver-colored wire on identification silver foil surface, adopt low angle light source.For making silver foil surface uniform gray level, reduce light and shade difference, be convenient to later image identification, adopt annular light source, the present embodiment is selected annular low angle light source.
The pattern recognition device of described picture recognition module as described in any one in aforementioned claim 1～7, and described motion-control module and welding mechanism be the common result of industry, do not repeat them here.
Tool of the present invention has the following advantages:
(1) according to the image-recognizing method of the embodiment of the present invention, the gray difference based between silver foil and silver-colored wire is identified, around the particle of silver foil, look for guide line position, and finally determine solder joint, can automatically identify easily and quickly thus bond pad locations;
(2) according to the image-recognizing method of the embodiment of the present invention, by image being carried out to the statistics of grey level histogram, then image is carried out to the linear transformation of a series of gray scales, finally obtain the image that contrast strengthens, strengthen the gray difference between silver foil and silver-colored wire, be convenient to the identification in later stage;
(3), according to the solar array robot welding system of the embodiment of the present invention, can realize the accurate location to required solder joint, thereby realize automatic welding.Adopt picture recognition module, operating personnel do not need butt welding point repeatedly to debug, aim at, and have improved efficiency and the reliability of operation;
(4), according to the solar array robot welding system piece of the embodiment of the present invention, because picture recognition module is very high to the requirement of light source, the configuration of light source has directly affected image pre-service and the graphical analysis in later stage.Picture recognition module need to be identified the silver-colored wire on silver foil surface, and silver foil is identical with silver-colored colour of insulated conductor, and ordinary light source is difficult to identification, and the height of silver foil and silver-colored wire is inconsistent, adopts low angle annular light source can strengthen the contrast of wire and silver foil gray scale.
To those skilled in the art, obviously the invention is not restricted to the details of above-mentioned one exemplary embodiment, and in the situation that not deviating from spirit of the present invention or essential characteristic, can realize the present invention with other concrete form.Therefore, no matter from which point, all should regard embodiment as exemplary, and be nonrestrictive, scope of the present invention is limited by claims rather than above-mentioned explanation, is therefore intended to all changes that drop in the implication and the scope that are equal to important document of claim to include in the present invention.Any Reference numeral in claim should be considered as limiting related claim.In addition, obviously other unit or step do not got rid of in " comprising " word, and odd number is not got rid of plural number.Multiple unit of stating in system claim or device also can be realized by software or hardware by a unit or device.The first, the second word such as grade is used for representing title, and does not represent any specific order.
Claims (9)
1. an image-recognizing method, it identifies silver-colored wire bonds at the lip-deep solder joint of silver foil in the collection image through gathering, described collection image comprises at least one first area corresponding to described silver foil and at least one second area corresponding to described silver-colored wire, it is characterized in that, comprise the steps:
(a) obtain the gray-scale value of each pixel in described first area and described second area;
(b) gray-scale value based on each pixel in described first area, obtains the unique point of described first area; And
(c), based on described unique point, in described second area, look for described solder joint according to the gray-scale value of each pixel in described second area.
2. recognition methods according to claim 1, is characterized in that, step (a) also comprises:
(A) gray-scale value of the each pixel to described first area and described second area carries out greyscale transformation, to increase the difference between the gray-scale value of each pixel in the gray-scale value of each pixel in described first area and described second area,
And step (A) comprises the gray-scale value that reduces each pixel in described first area, and increase the gray-scale value of each pixel in described second area.
3. according to the recognition methods described in right 1, it is characterized in that, described step (b) comprising:
(b1) according to the first predetermined gray-scale value, determine the elemental area of described first area, thereby determine the surface area of described first area; And
(b2) obtain the barycenter of the surface area of described first area, as described unique point.
4. according to the recognition methods described in right 3, it is characterized in that, described step (b1) comprising:
(b11) described collection image is carried out to Region Segmentation;
(b12) carry out region growing from the initial pixel of each cut zone, wherein said region growing criterion is, in the time that the gray-scale value of pixel exceedes predetermined growing threshold, to carry out region growing, when the gray-scale value of pixel is during lower than predetermined growing threshold, stop growing; And
(b13) screen the region that described growth obtains, and the selection result is carried out to area statistics, to determine the surface area of described first area.
5. according to the recognition methods described in right 1, it is characterized in that, in described step (c), describedly comprise take the horizontal ordinate point of described unique point or ordinate point as benchmark based on described unique point.
6. according to the recognition methods described in right 5, it is characterized in that, in described step (c), take described horizontal ordinate point as benchmark in the situation that, within the scope of ordinate traversal, carry out ordinate traversal along axis of ordinates, the ordinate place pixel around more a certain ordinate in predetermined ordinate scope (M1) and the gray-scale value of described a certain ordinate place pixel; If the difference of gray-scale value exceedes setting threshold, represent that described a certain ordinate place pixel has gray scale sudden change, and this ordinate and the described horizontal ordinate point as benchmark definite place pixel be catastrophe point; Described catastrophe point is calculated, wherein, in the time that the counting of described catastrophe point arrives a certain predetermined value, corresponding this catastrophe point ordinate point and described horizontal ordinate point are defined as to the coordinate points of described solder joint.
7. according to the recognition methods described in right 5, it is characterized in that, in described step (c), take described ordinate point as benchmark in the situation that, within the scope of ordinate traversal, carry out horizontal ordinate traversal along abscissa axis, the horizontal ordinate place pixel around more a certain horizontal ordinate in predetermined ordinate scope (M2) and the gray-scale value of described a certain horizontal ordinate place pixel; If the difference of gray-scale value exceedes setting threshold, represent that described a certain horizontal ordinate place pixel has gray scale sudden change, and this horizontal ordinate and the described ordinate point as benchmark definite place pixel be catastrophe point; Described catastrophe point is calculated, wherein, in the time that the counting of described catastrophe point arrives a certain predetermined value, corresponding this catastrophe point horizontal ordinate point and the described point of the ordinate as benchmark are defined as to the coordinate points of described solder joint.
8. a welding system for solar array, is characterized in that, comprising:
Image capture module, it needs the silver-colored wire of welding and the image of silver foil for gathering;
Picture recognition module, it receives the collection image that described image capture module gathers, and identifies silver-colored wire bonds at the lip-deep solder joint of silver foil, and the pattern recognition device of described picture recognition module as described in any one in aforementioned claim 1～7;
Welding mechanism, it welds described silver-colored wire and described silver foil at described solder joint place; And
Motion-control module, the solder joint that it identifies according to described picture recognition module, controls described welding mechanism and moves to described solder joint position, completes the welding of described solder joint.
9. according to the welding system described in right 8, it is characterized in that, described image capture module comprises low angle light source, to utilize described low angle light source to throw light on to described silver-colored wire and silver foil in the time gathering.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210439246.XA CN103810458A (en) | 2012-11-07 | 2012-11-07 | Image recognition method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210439246.XA CN103810458A (en) | 2012-11-07 | 2012-11-07 | Image recognition method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN103810458A true CN103810458A (en) | 2014-05-21 |
Family
ID=50707206
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201210439246.XA Pending CN103810458A (en) | 2012-11-07 | 2012-11-07 | Image recognition method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103810458A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103846538A (en) * | 2012-11-29 | 2014-06-11 | 上海航天设备制造总厂 | Image recognition device and battery array welding system utilizing same |
| CN103846539A (en) * | 2012-11-29 | 2014-06-11 | 上海航天设备制造总厂 | Image recognition method |
| CN108335299A (en) * | 2018-03-26 | 2018-07-27 | 戴惠英 | Model selection platform based on image procossing |
| CN109623151A (en) * | 2019-02-20 | 2019-04-16 | 中铝瑞闽股份有限公司 | A kind of aluminium volume end face welding system based on line laser |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101159017A (en) * | 2007-11-27 | 2008-04-09 | 清华大学 | Welding line automatic recognition visible sensation method based on partial image texture characteristic matched |
| CN101358836A (en) * | 2008-09-28 | 2009-02-04 | 西安理工大学 | Method for recognising weld spot central position based on computer vision |
| US20090097162A1 (en) * | 2007-10-11 | 2009-04-16 | Nhk Spring Co., Ltd. | Head suspension, load beam, and method of manufacturing load beam |
| CN101666621A (en) * | 2009-09-24 | 2010-03-10 | 浙江大学 | Gauge point extracting method without reference to illumination |
| CN101961819A (en) * | 2009-07-22 | 2011-02-02 | 中国科学院沈阳自动化研究所 | Device for realizing laser welding and seam tracking and control method thereof |
| CN102179598A (en) * | 2011-03-31 | 2011-09-14 | 中国科学院自动化研究所 | Method and device for centering initial welding positions of flat narrow welding lines and tracking welding lines |
| CN102663781A (en) * | 2012-03-23 | 2012-09-12 | 南昌航空大学 | Sub-pixel level welding center extraction method based on visual sense |
-
2012
- 2012-11-07 CN CN201210439246.XA patent/CN103810458A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090097162A1 (en) * | 2007-10-11 | 2009-04-16 | Nhk Spring Co., Ltd. | Head suspension, load beam, and method of manufacturing load beam |
| CN101159017A (en) * | 2007-11-27 | 2008-04-09 | 清华大学 | Welding line automatic recognition visible sensation method based on partial image texture characteristic matched |
| CN101358836A (en) * | 2008-09-28 | 2009-02-04 | 西安理工大学 | Method for recognising weld spot central position based on computer vision |
| CN101961819A (en) * | 2009-07-22 | 2011-02-02 | 中国科学院沈阳自动化研究所 | Device for realizing laser welding and seam tracking and control method thereof |
| CN101666621A (en) * | 2009-09-24 | 2010-03-10 | 浙江大学 | Gauge point extracting method without reference to illumination |
| CN102179598A (en) * | 2011-03-31 | 2011-09-14 | 中国科学院自动化研究所 | Method and device for centering initial welding positions of flat narrow welding lines and tracking welding lines |
| CN102663781A (en) * | 2012-03-23 | 2012-09-12 | 南昌航空大学 | Sub-pixel level welding center extraction method based on visual sense |
Non-Patent Citations (3)
| Title |
|---|
| 孙太生 等: "基于改进区域生长算法的焊缝图像分割", 《现代焊接》 * |
| 王婳懿: "基于eVision的全自动定位焊点的机器视觉系统", 《焊接设备与材料》 * |
| 陈方昕: "基于区域生长法的图像分割技术", 《科技信息（科学教研）》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103846538A (en) * | 2012-11-29 | 2014-06-11 | 上海航天设备制造总厂 | Image recognition device and battery array welding system utilizing same |
| CN103846539A (en) * | 2012-11-29 | 2014-06-11 | 上海航天设备制造总厂 | Image recognition method |
| CN108335299A (en) * | 2018-03-26 | 2018-07-27 | 戴惠英 | Model selection platform based on image procossing |
| CN109623151A (en) * | 2019-02-20 | 2019-04-16 | 中铝瑞闽股份有限公司 | A kind of aluminium volume end face welding system based on line laser |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103810459A (en) | Image recognition device and solar array welding system by using same | |
| CN101443791B (en) | Improved foreground/background separation in digitl images | |
| CN109871763B (en) | Specific target tracking method based on YOLO | |
| CN105758397B (en) | A kind of aircraft camera positioning method | |
| DE102009005861B4 (en) | Vehicle environment detection system | |
| CN102646193B (en) | Segmentation method of character images distributed in ring shape | |
| CN103354599B (en) | A kind of auto focusing method being applied to dynamic light source scene and device | |
| CN103810458A (en) | Image recognition method | |
| CN104463117A (en) | Sample collection method and system used for face recognition and based on video | |
| WO2016175234A1 (en) | Color image processing method, color image processing program, and object recognition method and device | |
| CN103929588A (en) | Rapid and automatic focusing method and system for camera zoom | |
| DE102019114533A1 (en) | DETECTION OF VISIBLE LIGHT COMMUNICATION SOURCES OVER A HIGH DYNAMIC RANGE | |
| CN102646343A (en) | Vehicle detection apparatus | |
| CN109447066B (en) | Rapid and accurate single-stage target detection method and device | |
| CN104636760A (en) | Positioning method for welding seam | |
| DE102012108764A1 (en) | Image processing device | |
| CN104897671A (en) | Identification system for fruit stem and calyx of fruit | |
| CN102608130A (en) | Smart card stain detecting system based on image feature matching technology and detection and detecting method | |
| CN102622578A (en) | Passenger counting system and passenger counting method | |
| CN103364398A (en) | Joint inspection apparatus | |
| CN106506946B (en) | A kind of camera automatic focusing method and video camera | |
| CN105825219A (en) | Machine vision detection system | |
| Lee et al. | Pedestrian detection based on deep fusion network using feature correlation | |
| EP3812953A1 (en) | Code reader and method for reading optical codes | |
| CN103176668B (en) | A kind of shooting method for correcting image for camera location touch system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20140521 |
|
| WD01 | Invention patent application deemed withdrawn after publication |