CN105405117B - Angular Point Extracting Method and device based on image outline - Google Patents
Angular Point Extracting Method and device based on image outline Download PDFInfo
- Publication number
- CN105405117B CN105405117B CN201510673750.XA CN201510673750A CN105405117B CN 105405117 B CN105405117 B CN 105405117B CN 201510673750 A CN201510673750 A CN 201510673750A CN 105405117 B CN105405117 B CN 105405117B
- Authority
- CN
- China
- Prior art keywords
- curvature
- point
- value
- profile
- angle 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.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
-
- 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
- G06T2207/20—Special algorithmic details
- G06T2207/20112—Image segmentation details
- G06T2207/20164—Salient point detection; Corner detection
-
- 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
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
- G06T2207/30141—Printed circuit board [PCB]
Abstract
Angular Point Extracting Method disclosed by the invention based on image outline, including：After image outline is extracted, " the k cosine curvature " method of use extracts notable angle point from the profile point of image outline, wherein, the notable angle point is more than the point of preset curvature for curvature；Data are described using the corresponding contour curve of profile point between notable angle point described in fitting algorithm generation each two；When the contour curve describes the difference of data and normal data more than predetermined threshold value, weak angle point is extracted from the profile point according to preset rules, wherein, the weak angle point is less than the point of preset curvature for curvature.The invention also discloses a kind of angle point grid devices based on image outline.Compared with prior art, this programme more can accurately extract the angle point in image outline comprehensively, and the contour segment ensured is more accurate, and then can improve the accuracy of detection of PCB substrate.
Description
Technical field
The present invention relates to technical field of image detection, and in particular to a kind of Angular Point Extracting Method and dress based on image outline
It puts.
Background technology
Substrates of the PCB (Printed Circuit Board, printed wiring board) as assembling electronic component, was making
Cheng Zhong by process such as hole machined, electroless copper, electro-coppering, etchings, obtains required circuitous pattern.It is arranged on PCB bases
The circuitous pattern of plate forms electric coupling area, and the predeterminated position in electric coupling area is provided with the welding hole of circuit components, is inciting somebody to action
After related circuit component is welded in PCB substrate, available circuit board is obtained.
Since the shape of electric coupling area, size and position directly affect the performance of PCB substrate and reliable in PCB substrate
Property, therefore, after PCB substrate completes, usually to members such as the straight wire of electric coupling area, circle pad, square pad, electric areas
Plain type is detected.Specifically, can the profile of PCB substrate be mapped to image detection software obtains image outline, wherein,
Image outline is made of profile point, differently contoured section of intersecting obtained angle point is then extracted from profile point, with what is extracted
Image outline is divided into contour segment by angle point as endpoint, each element type is detected by the size and shape for detecting contour segment
Whether corresponding conditions are met.
In the relevant technologies, the cosine value of angle is formed by calculating each profile point and its front and rear k profile point and is carried
Angle point is taken, is known as " k- cosine curvature " method in the related art.However, there are the angle point of many angle very littles in image outline,
For example, the angle point that straight line is formed with the tangent point of contact of circular arc and straight line with straight line intersection but the intersection point of turning very little, uses " k-
Cosine curvature " method can not be accurately identified, and cause obtained contour segment inaccurate, and the examination criteria of different contour segments is not
Together, the corresponding condition met also differs, therefore, once the detection that obtained contour segment inaccuracy will influence contour segment, and then
Cause the testing result low precision of PCB substrate.
Invention content
In view of technical problem present in the relevant technologies, an embodiment of the present invention provides a kind of angle points based on image outline
Extracting method and device, PCB substrate testing result low precision asks caused by solve due to can not accurately extract segment angle point
Topic.
In a first aspect, an embodiment of the present invention provides a kind of Angular Point Extracting Method based on image outline, including：Work as extraction
After image outline, " the k- cosine curvature " method of use extracts notable angle point from the profile point of image outline, wherein, it is described notable
Angle point is more than the point of preset curvature for curvature；It is corresponding using profile point between notable angle point described in fitting algorithm generation each two
Contour curve describes data；When the contour curve describes the difference of data and normal data more than predetermined threshold value, according to pre-
If rule extracts weak angle point from the profile point, wherein, the weak angle point is less than the point of preset curvature for curvature.
Preferably, it is described that weak angle point is extracted from the profile point according to preset rules, including：Call the default rule of the first son
Then；The weak angle point is extracted from the profile point according to the described first sub- preset rules；When according to the default rule of the described first son
When not extracting the weak angle point from the profile point then, the second sub- preset rules are called；According to the default rule of the described second son
The weak angle point is then extracted from the profile point.
Preferably, it is described that the weak angle point is extracted from the profile point according to the described first sub- preset rules, including：From
Curvature peak value is obtained in the corresponding curvature curve of the profile point；Calculate the continuum that the curvature peak value both sides curvature value is zero
The length value of section；It is preset when the length value of continuous section that the curvature peak value both sides curvature value is zero is more than or equal to first
During value, the corresponding profile point of the curvature peak value is determined as weak angle point.
Preferably, the weak angle point is extracted from the profile point according to the described second sub- preset rules, including：From described
Trapezoidal curvature curve is obtained in the corresponding curvature curve of profile point；Calculate the length of the trapezoidal curvature curve mean curvature peak value
Value；When the length value of the trapezoidal curvature curve mean curvature peak value is more than or equal to the second preset value, determine described trapezoidal
Curvature curve mean curvature value rises section and curvature value declines the corresponding profile point range of section；Respectively to the profile point range
Interior profile point is fitted operation and obtains description data；By the description data and the profile point of the difference minimum of normal data
It is determined as weak angle point.
Preferably, after the length value for calculating the trapezoidal curvature curve mean curvature peak value, the trapezoidal song is being determined
Before rate curve mean curvature value rises section and the corresponding profile point range of curvature value decline section, further include：It is calculated using fitting
Method calculates the radius of the corresponding contour curve of the profile point；Judge whether the radius is less than default support width, wherein, institute
When stating default support width to calculate profile point curvature, the span value of profile point to be calculated and selected profile point；It is if described
Radius is less than the default support width, performs the determining trapezoidal curvature curve mean curvature value and rises section and curvature value
The step of declining section corresponding profile point range；If the radius is more than or equal to default support width, by default step
The long N number of detection support width value of setting, and according to the curvature value of the first detection support width value calculating profile point, wherein, N
For natural number, detection support width value gradually increases from 1 to N；When the quantity of the profile point of curvature value continuous phase etc. is more than third
During preset value, the profile point of the curvature value continuous phase etc. is determined as profile point to be detected；Sequentially supported according to the second detection
Width value computes repeatedly the curvature value of the profile point to be detected to N detections support width value；It is wide when detecting support according to I
Whens the obtained curvature value of angle value and discontinuous phase etc., I -1 is detected into support width and is worth the rising of corresponding curvature curve, only takes turns
Exterior feature point is determined as weak angle point, wherein, N >=I>1.
Preferably, it further includes：Profile segment type between the weak angle point of each two is determined by fitting algorithm；When adjacent two-wheeled
When the type of wide section is identical, the weak angle point between two contour segments is deleted.
Second aspect, an embodiment of the present invention provides a kind of angle point grid device based on image outline, including：Extract mould
Block, for after image outline is extracted, " the k- cosine curvature " method of use to extract notable angle point from the profile point of image outline,
Wherein, the notable angle point is more than the point of preset curvature for curvature；Generation module, for generating the extraction using fitting algorithm
The corresponding contour curve of profile point describes data between the notable angle point of each two that module is extracted；The extraction module, is also used
When the contour curve generated when the generation module describes the difference of data and normal data more than predetermined threshold value, according to pre-
If rule extracts weak angle point from the profile point, wherein, the weak angle point is less than the point of preset curvature for curvature.
Preferably, the extraction module includes call unit and extraction unit, wherein, the call unit, for calling
First sub- preset rules；The extraction unit, described in being extracted from the profile point according to the described first sub- preset rules
Weak angle point；The call unit is additionally operable to described when not extracted from the profile point according to the described first sub- preset rules
During weak angle point, the second sub- preset rules are called；The extraction unit is additionally operable to according to the described second sub- preset rules from the wheel
The weak angle point is extracted in exterior feature point.
Preferably, the extraction unit includes obtaining subelement, computation subunit and determination subelement, wherein, it is described to obtain
Subelement is taken, for obtaining curvature peak value from the corresponding curvature curve of the profile point；The computation subunit, for calculating
The length value of continuous section that the curvature peak value both sides curvature value is zero；The determination subelement, for working as the curvature peak
When the length value for the continuous section that value both sides curvature value is zero is more than or equal to the first preset value, the curvature peak value is corresponded to
Profile point be determined as weak angle point.
Preferably, the extraction unit further includes：Fitting operation subelement, wherein, the acquisition subelement, be additionally operable to from
Trapezoidal curvature curve is obtained in the corresponding curvature curve of the profile point；The computation subunit is additionally operable to calculate described trapezoidal
The length value of curvature curve mean curvature peak value；The determination subelement is additionally operable to when the trapezoidal curvature curve mean curvature peak value
Length value when being more than or equal to the second preset value, determine that the trapezoidal curvature curve mean curvature value rises section and curvature value
Decline the corresponding profile point range of section；The fitting operation subelement, for respectively to the profile in the range of the profile point
Point carries out fitting operation and obtains description data；The determination subelement is additionally operable to the difference by the description data and normal data
The profile point of value minimum is determined as weak angle point.
Summary, to solve the problems, such as that the relevant technologies can not accurately extract angle point, the technical solution of the embodiment of the present invention
After image outline is extracted, " k- cosine curvature " method is used to extract notable angle point from the profile point of image outline first,
In, the notable angle point is more than the point of preset curvature for curvature；Then, using notable angle point described in fitting algorithm generation each two
Between profile point corresponding contour curve data are described, when the contour curve describe data and normal data difference be more than it is pre-
If during threshold value, illustrate in the profile point there may be it is unidentified go out angle point, so as to according to preset rules from the profile point
The middle weak angle point of extraction, wherein, the weak angle point is less than the point of preset curvature for curvature.It is obvious that the technology of the embodiment of the present invention
Scheme extracts the angle point in image outline according to amount of curvature in two stages, moreover, the extraction conditions in two extraction stages
Correspond to the angle point setting of different condition respectively so that formed between two extraction stages it is complementary, so as to will be in image outline
Angle point more accurately extract comprehensively, the contour segment ensured is more accurate, and then can improve the detection of PCB substrate
Precision.
Description of the drawings
It in order to illustrate more clearly about the embodiment of the present invention or technical scheme of the prior art, below will be to institute in embodiment
Attached drawing to be used is needed to be briefly described, it should be apparent that, the accompanying drawings in the following description is only some implementations of the present invention
Example, for those of ordinary skill in the art, without creative efforts, can also obtain according to these attached drawings
Obtain other attached drawings.By the way that shown in attached drawing, above and other purpose of the invention, feature and advantage will be more clear.In whole
Identical reference numeral indicates identical part in attached drawing.Deliberately attached drawing, emphasis are not drawn by actual size equal proportion scaling
It is purport of the invention is shown.
Fig. 1 is the method flow diagram of the Angular Point Extracting Method provided in an embodiment of the present invention based on image outline；
Fig. 2 is the side of second of embodiment of the Angular Point Extracting Method provided in an embodiment of the present invention based on image outline
Method flow chart；
Fig. 3 is the exemplary plot of partial trace in image outline provided by the invention；
Fig. 4 is the curvature curve schematic diagram of all profile points in image outline provided by the invention；
Fig. 5 is image outline exemplary plot provided in an embodiment of the present invention；
Fig. 6 is the curvature curve exemplary plot of the weak angle point of the first kind provided in an embodiment of the present invention；
Fig. 7 is the curvature curve exemplary plot of the weak angle point of the second class provided in an embodiment of the present invention；
Fig. 8 is the structure diagram of the angle point grid device provided in an embodiment of the present invention based on image outline.
Specific embodiment
By the relevant technologies it is found that the image outline that PCB substrate maps is made of contour segment, contour segment mainly includes straight
Two class of line and circular arc, therefore, angle point can include straight line and the intersection point of straight line, the intersection point of straight line and circular arc, circular arc and circular arc
The point of contact of intersection point, straight line and circular arc.Due to can not accurately extract all angle points using the relevant technologies, the present invention is real
Type of the example according to angle point is applied, the angle point by two stages comprehensively, in accurate extraction image outline.
Below in conjunction with the attached drawing in the embodiment of the present invention, the technical solution in the embodiment of the present invention is carried out clear, complete
Whole description, it is clear that described embodiment is only part of the embodiment of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, those of ordinary skill in the art are obtained every other without making creative work
Embodiment shall fall within the protection scope of the present invention.
Fig. 1 is referred to, Fig. 1 is the method flow of the Angular Point Extracting Method provided in an embodiment of the present invention based on image outline
Figure, the method includes following step：
Step S101, after image outline is extracted, " the k- cosine curvature " method of use is carried from the profile point of image outline
Take notable angle point.
Wherein, for PCB substrate after completing, image detection software extracts the image outline of PCB substrate.It needs to illustrate
, for image detection software after PCB substrate image is obtained, the marginal point of tracking PCB substrate image obtains one continuously
Edge point set, which forms image outline, and marginal point is the profile point described in the embodiment of the present invention.Due to
The technology that the technology of extraction image outline is well known to those skilled in the art, therefore, details are not described herein again for the embodiment of the present invention.
After image outline is obtained, " k- cosine curvature " method is used to calculate the song of each profile point in image outline first
Rate, specifically, the process for calculating profile point curvature is introduced with reference to Fig. 3.
Fig. 3 shows the partial trace in image outline, wherein, profile point PiWheel for curvature to be calculated in the present embodiment
It is wide, calculating profile point PiCurvature before, be equipped beforehand with width K, when the default support width is calculates curvature,
The profile point to be calculated of setting and the span value of selected profile point, accordingly, it is determined that profile point PiLater, profile point P is determinediBefore,
Each K profile point P afterwardsi-KWith profile point Pi+K, profile point Pi-KWith profile point PiLine formed vectorProfile point Pi+KWith
Profile point PiLine formed vectorVectorWith vectorForm angle theta i, profile point PiCurvature be the angle theta
The cosine value of i.
It should be noted that the preferable example of the present invention is above are only, the calculating for other profile point curvature, with
Foregoing description is similar, and details are not described herein again for the embodiment of the present invention.In addition, in embodiments of the present invention, presetting support width K can be with
It is 4-6 pixels, specifically, carrying out different settings according to different image outlines, the embodiment of the present invention is without limitation.
In this step, after the curvature of all profile points is calculated, curvature curve schematic diagram as shown in Figure 4 is obtained,
When being more than certain value due to curvature peak value, the angle angle for forming the contour segment generation of the corresponding profile point of curvature peak value is opposite
Smaller, the angle point in image outline as shown in Figure 5 shown in angle point 5, angle point 6, angle point 7 and angle point 8 can explicitly should very much
Curvature peak value correspondence profile point is determined as angle point；However, when curvature peak value is less than the value, the corresponding wheel of curvature peak value is formed
The angle angle that the contour segment of exterior feature point generates is relatively large, angle point 1, angle point 2, angle point 3 and angle in image outline as shown in Figure 5
Angle point shown in point 4, can not explicitly judge whether the corresponding profile point of curvature peak value is angle point very much.Therefore, when obtaining
After curvature curve, can curvature threshold be set according to curvature curve, will be greater than the corresponding angle point of curvature peak value of curvature threshold
It is determined as notable angle point, and continues curvature curve of the detection less than curvature threshold, it is weak less than curvature threshold to extract curvature
Angle point.
Technical solution described in this step, can be by the angle point preliminary classification in image outline, and by notable angle point grid
Go out, so that it is determined that range existing for weak angle point, in order to preferably extract weak angle point.
Step S102 is retouched using the corresponding contour curve of profile point between notable angle point described in fitting algorithm generation each two
State data.
Wherein, after notable angle point is extracted, profile between the notable angle point of each two is primarily determined by fitting algorithm
The line style of curve.Fitting algorithm is the mathematical model established according to specific line style, the location information for obtaining notable angle point it
Afterwards, the profile dot position information between two adjacent notable angle points is substituted into fitting formula, obtains the description data of contour curve, and
The difference of description data standard value corresponding with mathematical model that digital simulation obtains, judges whether the difference is more than default threshold
Contour curve if difference is not more than predetermined threshold value, is determined as the corresponding line style of respective standard value, if difference is more than by value
Predetermined threshold value, then the line style of contour curve line style corresponding with respective standard value do not meet.
It should be noted that the line style of contour curve includes straight line and curve, and the fitting algorithm model of straight line and curve
Difference, corresponding standard value is also different, accordingly, it is determined that being needed in order during the line style of contour curve between two notable angle points
Description data are calculated respectively using two kinds of fitting algorithms.When the line style that the first fitting algorithm is used to can determine contour curve
When, stop calculating；When using the first fitting algorithm that can not determine the line style of contour curve, second of fitting algorithm pair is used
The description data of contour curve are calculated, if the difference of description data and the standard value of second of fitting algorithm is not more than default threshold
Value, the line style of contour curve is the corresponding line style of second of fitting algorithm, if description data and the mark of second of fitting algorithm
The difference of quasi- value is more than predetermined threshold value, then it is assumed that there are weak angle point in profile point between the two notable angle point, and therefrom extracts weak
Angle point.
Wherein, the first fitting algorithm can be Algorithm of fitting a straight line in the above process, and the second fitting algorithm is curve matching
Algorithm, alternatively, the first fitting algorithm be curve fitting algorithm, the second fitting algorithm be Algorithm of fitting a straight line, the embodiment of the present invention
It is without limitation.
The description data of contour curve are calculated by fitting algorithm for this step, so as to according to describing data
The profile segment limit there are weak angle point is filtered out, and then sufficient preparation is made that extract weak angle point.
Step S103, when the contour curve describes the difference of data and normal data more than predetermined threshold value, according to pre-
If rule extracts weak angle point from the profile point.
Wherein, according to foregoing description it is found that weak angle point is the larger point of angle, such angle point is divided into two classes, including：Straight line
The angle point that the larger intersection point of angle formed with straight line is formed, such as the angle point 3 in Fig. 5 and angle point 4 and the point of contact of straight line and circular arc
The angle point of formation, such as the angle point 1 and angle point 2 in 5.
Due to two kinds of weak angle point have the characteristics that it is different, the technical program be directed to the weak angle point of two types
Two sets of extracting rules are provided with, after profile point range existing for weak angle point is determined, sequentially using two kinds of extracting rules to weak
Angle point extracts.Specifically, calling the first sub- preset rules first, and extracted from profile point according to the first sub- preset rules
Weak angle point；If weak angle point can not be extracted using the first sub- preset rules, the second sub- preset rules are called, and according to first
Sub- preset rules extract weak angle point from profile point.
Wherein, in embodiments of the present invention, the first sub- preset rules can be the weak angle point that extraction straight line is formed with straight line
Rule, the second sub- preset rules are to extract the rule for the weak angle point that straight line and curve are formed, alternatively, the first sub- preset rules can
Think the rule for the weak angle point that extraction straight line is formed with curve, the second sub- preset rules are to extract the weak angle that straight line is formed with straight line
The rule of point, the embodiment of the present invention are without limitation.
Technical solution described in the present embodiment extracts the angle point in image outline according to amount of curvature in two stages,
Moreover, the extraction conditions in two extraction stages correspond to the angle point setting of different condition respectively so that shape between two extraction stages
Into complementation, so as to which the angle point in image outline is more accurately extracted comprehensively, the contour segment ensured is more accurate
Really, and then the accuracy of detection of PCB substrate can be improved.
Above-described embodiment describes technical scheme of the present invention on the whole, in order to make those skilled in the art more clear
Chu, the technical solution of the detailed understanding embodiment of the present invention, with reference to weak angle point type to the extraction process of weak angle point into
Row detailed description.
It should be noted that description of the present embodiment based on above-described embodiment, is the supplement to above-described embodiment, therefore,
Part same as the previously described embodiments in the present embodiment, refers to the description of above-described embodiment, and the present embodiment repeats no more.
Referring to Fig. 2, Fig. 2 is second of implementation of the Angular Point Extracting Method provided in an embodiment of the present invention based on image outline
The method flow diagram of mode, the described method comprises the following steps：
Step S201, after image outline is extracted, " the k- cosine curvature " method of use is carried from the profile point of image outline
Take notable angle point.
Wherein, referring to Fig. 5, in this step, " the k- cosine curvature " method of use extracts notable angle point angle point 5, angle point 6, angle
Point 7 and angle point 8.Specific extraction process refers to the description of above-described embodiment, and details are not described herein again for the present embodiment.
Step S202 is retouched using the corresponding contour curve of profile point between notable angle point described in fitting algorithm generation each two
State data.
Wherein, after notable angle point angle point 5, angle point 6, angle point 7 and angle point 8 are extracted, respectively to each two angle point it
Between contour line be fitted operation in the way of first fitting a straight line again curve matching, obtain the description data of contour curve,
By describing data and standard value difference determines contour curve between angle point 5 and angle point 6 and between angle point 7 and angle point 8
For straight line, the contour curve between angle point 5 and angle point 8 is curve.Since the contour curve between angle point 6 and angle point 7 is not one
Bar line, therefore, the description data obtained using Algorithm of fitting a straight line and curve fitting algorithm, with the difference of respective standard value compared with
Greatly, so, the profile point between angle point 6 and angle point 7 is determined as to the profile point of weak angle point to be extracted.
It should be noted that in this step, can description data first be obtained using Algorithm of fitting a straight line, reuse curve
Fitting algorithm obtains description data, first can also obtain description data using curve fitting algorithm, reuse Algorithm of fitting a straight line
Description data are obtained, specifically, being performed according to preset program, the embodiment of the present invention is without limitation.
Step S203 when the difference that the contour curve describes data and normal data is more than predetermined threshold value, calls the
One sub- preset rules.
Wherein, according to the description of above-described embodiment it is found that weak angle point includes two types, therefore, the embodiment of the present invention carries
It is provided with the rule of two sets of weak angle points of extraction.
Specifically, according to the type of weak angle point, the first sub-rule can be that extraction straight line is formed with straight line in the present embodiment
Weak angle point rule, the second sub-rule can be the rule of weak angle point that extraction straight line and curve are formed.
Step S204 extracts the weak angle point according to the described first sub- preset rules from the profile point.
Wherein, in the weak angle point formed in straight line and straight line, such as the angle point 3 and angle point 4 in Fig. 5, due to straight line generally all
It is long, it therefore, at least can there are a longer zero curvature section, such as Fig. 6 at the curvature peak value both ends less than curvature threshold
The weak angle point curvature curve exemplary plot of the shown first kind, the profile point corresponding to curvature peak value 001 are that straight line is formed with straight line
Weak angle point.It therefore, can be using curvature peak value and zero curvature section length as extraction foundation.
Specifically, curvature peak value is obtained in the corresponding curvature curve of profile point between angle point 6 and angle point 7, due to angle
Point is 6 more complicated with the contour curve of angle point 7, and image detection software when calculating profile point curvature there are disturbing factor, therefore,
At least two curvature peak values can be got, after curvature peak value is obtained, calculate each curvature peak value both sides curvature value respectively
The length value for the continuous section for being zero, when the length value of continuous section that curvature peak value both sides curvature value is zero is more than or equal to
During the first preset value, which is determined as weak angle point, so as to extract angle point 6 and angle point 7.
It should be pointed out that in a preferred embodiment of the invention, the first preset value can be the value more than 30 pixels, can
Specifically to be set according to concrete application scene, the embodiment of the present invention is without limitation
Step S205, when not extracting the weak angle point from the profile point according to the described first sub- preset rules,
Call the second sub- preset rules.
However, if the weak angle point that straight line is formed with straight line is not present in profile point, the type of weak angle point is straight line and song
The point of contact of line, such as the angle point 1 and angle point 2 in Fig. 5, then, it will be unable to find by the process of step S204 qualified
Curvature peak value can call the second sub- preset rules at extraction point of contact to carry out the extraction of weak angle point at this time.
Step S206 extracts the weak angle point according to the described second sub- preset rules from the profile point.
Referring to Fig. 7, Fig. 7 is the curvature curve exemplary plot of the weak angle point of the second class provided in an embodiment of the present invention, in general, using
The curvature curve that " k- cosine curvature " method obtains, when there are during the profile point of straight line and contact of a curve, the profile point with before and after it
The curvature curve of several profile points shows as trapezoidal rising edge or failing edge, as shown in Fig. 7 002 and 003.Therefore, exist
When extracting the weak angle point that point of contact is formed, the characteristic can be based on and carried out.
Specifically, it is bent to obtain trapezoidal curvature in the corresponding curvature curve of profile point from Fig. 5 between angle point 6 and angle point 7
Line calculates the length value of the trapezoidal curvature curve mean curvature peak value, when the length value of the trapezoidal curvature curve mean curvature peak value is big
When the second preset value, it may be determined that there are weak angle points in the corresponding profile point of the trapezoidal curvature curve.
Since straight line profile point corresponding with the point of contact of circular arc is difficult to be determined by trapezoidal curvature curve, but point of contact is located at
Determined by the rising edge and failing edge of trapezoidal curvature curve in range, therefore, in the length of trapezoidal curvature curve mean curvature peak value
When angle value is more than or equal to the second preset value, determine that the trapezoidal curvature curve mean curvature value rises section and curvature value declines
The corresponding profile point range of section, it is then possible to search description data and standard value from profile point range by fitting operation
The profile point of difference minimum, and the profile point is determined as weak angle point.
It should be pointed out that in a preferred embodiment of the invention, the second preset value can be the value more than 30 pixels, can
Specifically to be set according to concrete application scene, the embodiment of the present invention is without limitation.
Certainly, the above-mentioned weak Angular Point Extracting Method for selected default support width in the case of comparatively ideal, due to pre-
If support sector width is fixed value, when default support sector width opposing arcs radius is smaller, the curvature value being calculated is by dry
Being affected for information is disturbed, several fractures will occur in obtained trapezoidal curvature curve, influence the extraction of weak angle point, therefore, upper
On the basis of stating description, after the length value for calculating the trapezoidal curvature curve mean curvature peak value, trapezoidal curvature curve is being determined
Before mean curvature value rises section and the corresponding profile point range of curvature value decline section, the present invention also provides another to carry
Take mode.
In order to avoid due to presetting the influence that brings of support width fixation, in the present embodiment, first with fitting algorithm meter
The radius of the corresponding contour curve of profile point to be extracted is calculated, and judges whether the radius is less than default support width, if radius
Less than default support width, then, obtained trapezoidal curvature curve is comparatively preferable, it may be determined that in trapezoidal curvature curve
Curvature value rises section and curvature value declines the corresponding profile point range of section, and weak angle point is extracted from profile point range.
If radius is more than or equal to default support width, then, obtained trapezoidal curvature curve is disturbed information
Be affected, influence profile point range determine.At this point, by the N number of detection support width value of default step-length setting, wherein, N is
Natural number, detection support width value gradually increase from 1 to N, which is more than radius, is partly less than
Radius.And third preset value is set, third preset value can be 2.5 times of the first detection support width value.According to the first detection
Support width value calculates the curvature value of profile point, since the first detection support width value is more than contour curve radius, is taken turns calculating
During the curvature of each profile point on wide curve, according to the profile point selected by the first detection support width value across weak
Angle point, therefore, the obtained curvature value of each profile point are equal, when the quantity of the profile point of curvature value continuous phase etc. is more than the
During three preset values, weak angle point is included in, and therefore, the profile point of curvature value continuous phase etc. is determined as profile point to be detected.
Then, sequentially according to the second detection support width value the profile point to be detected is computed repeatedly to N detections support width value
Curvature value, due to from second detection support width value to N detection support width value quantity be gradually reduced, and support width value by
More than radius to radius is less than, therefore, when being worth obtained curvature value and discontinuous phase etc. according to I detection support width,
I -1 is detected into the rising of the corresponding curvature curve of support width value, stops profile point and is determined as weak angle point, wherein, N >=I>1.
Weak Angular Point Extracting Method described in the present embodiment, corresponding different types of weak angle point set different extracting rules,
So as to which the weak angle point of image outline is comprehensively and accurately extracted, so that identified contour segment is more accurate, from
And the precision of PCB substrate testing result can be greatly improved.
Furthermore, it is necessary to explanation, supports sector width to be more than contour curve radius, exists on extraction curve when preset
Weak angle point when, it is easy to occur be not weak angle point profile point accidentally extracted, therefore, corresponding line style be curve profile
Duan Shi is extracted after weak angle point completes, and profile segment type between the weak angle point of each two is determined by fitting algorithm, specifically, can
To calculate the radius length of contour segment and center location, if the type of two contour segments is identical, then it is assumed that between two contour segments
Weak angle point is the point of accidentally extraction, deletes the weak angle point, merges two contour segments, so as to do further inspection to the weak angle point extracted
It tests.
Summary, to solve the problems, such as that the relevant technologies can not accurately extract angle point, the technical solution of the embodiment of the present invention
After image outline is extracted, " k- cosine curvature " method is used to extract notable angle point from the profile point of image outline first,
In, the notable angle point is more than the point of preset curvature for curvature；Then, using notable angle point described in fitting algorithm generation each two
Between profile point corresponding contour curve data are described, when the contour curve describe data and normal data difference be more than it is pre-
If during threshold value, illustrate in the profile point there may be it is unidentified go out angle point, so as to according to preset rules from the profile point
The middle weak angle point of extraction, wherein, the weak angle point is less than the point of preset curvature for curvature.It is obvious that the technology of the embodiment of the present invention
Scheme extracts the angle point in image outline according to amount of curvature in two stages, moreover, the extraction conditions in two extraction stages
Correspond to the angle point setting of different condition respectively so that formed between two extraction stages it is complementary, so as to will be in image outline
Angle point more accurately extract comprehensively, the contour segment ensured is more accurate, and then can improve the detection of PCB substrate
Precision.
Corresponding with above-mentioned implementation method, the embodiment of the present invention additionally provides a kind of angle point grid based on image outline
Device, referring to Fig. 8, structure diagrams of the Fig. 8 for the angle point grid device provided in an embodiment of the present invention based on image outline, institute
Device is stated to include：Extraction module 11 and generation module 12, wherein, extraction module 11 is used for after image outline is extracted, and is used
" k- cosine curvature " method extracts notable angle point from the profile point of image outline, wherein, the notable angle point is more than pre- for curvature
If the point of curvature；Generation module 12, for generating the notable angle of each two that the extraction module 11 extracted using fitting algorithm
The corresponding contour curve of profile point describes data between point；In the present embodiment, extraction module 11 are additionally operable to when the generation mould
When the difference that the contour curve that block 12 is generated describes data and normal data is more than predetermined threshold value, according to preset rules from described
Weak angle point is extracted in profile point, wherein, the weak angle point is less than the point of preset curvature for curvature.
In the present embodiment, extraction module 11 includes call unit and extraction unit, wherein, call unit, for calling
First sub- preset rules；Extraction unit, for extracting the weak angle from the profile point according to the described first sub- preset rules
Point；In the present embodiment, it when not extracting the weak angle point from the profile point according to the described first sub- preset rules, adjusts
With unit, it is additionally operable to call the second sub- preset rules, correspondingly, extraction unit, is additionally operable to according to the described second sub- preset rules
The weak angle point is extracted from the profile point.
Specifically, extraction unit includes obtaining subelement, computation subunit and determination subelement, wherein, subelement is obtained,
For obtaining curvature peak value from the corresponding curvature curve of the profile point；Computation subunit, for calculating the curvature peak value
The length value for the continuous section that both sides curvature value is zero；Determination subelement is zero for working as the curvature peak value both sides curvature value
Continuous section length value be more than or equal to the first preset value when, the corresponding profile point of the curvature peak value is determined as weak
Angle point.
On the basis of above-described embodiment, extraction unit further includes：Fitting operation subelement, wherein, in the present embodiment,
The acquisition subelement is additionally operable to obtain trapezoidal curvature curve from the corresponding curvature curve of the profile point；It is described to calculate son
Unit is additionally operable to calculate the length value of the trapezoidal curvature curve mean curvature peak value；The determination subelement is additionally operable to when described
When the length value of trapezoidal curvature curve mean curvature peak value is more than or equal to the second preset value, determine in the trapezoidal curvature curve
Curvature value rises section and curvature value declines the corresponding profile point range of section；The fitting operation subelement, for right respectively
Profile point in the range of the profile point is fitted operation and obtains description data；The determination subelement, being additionally operable to will be described
The profile point of description data and the difference minimum of normal data is determined as weak angle point.
With reference to above-described embodiment, in another embodiment, the extraction unit is specifically additionally operable to utilize fitting algorithm meter
Calculate the radius of the corresponding contour curve of the profile point；Judge whether the radius is less than default support width, wherein, it is described pre-
If support width is calculates profile point curvature, the span value of profile point to be calculated and selected profile point；If the radius
Less than the default support width, perform the determining trapezoidal curvature curve mean curvature value and rise section and curvature value decline
The step of section corresponding profile point range；If the radius is more than or equal to default support width, set by default step-length
N number of detection support width value is put, and the curvature value of the profile point is calculated according to the first detection support width value, wherein, N is certainly
So number, detection support width value gradually increase from 1 to N；When the quantity of the profile point of curvature value continuous phase etc. is preset more than third
During value, the profile point of the curvature value continuous phase etc. is determined as profile point to be detected；Sequentially according to the second detection support width
It is worth the curvature value that N detections support width value computes repeatedly the profile point to be detected；When according to I detection support width value
Whens obtained curvature value and discontinuous phase etc., I -1 is detected into support width and is worth the rising of corresponding curvature curve, only profile point
It is determined as weak angle point, wherein, N >=I>1.
In another embodiment, the extraction unit further includes angle point and deletes subelement, in the present embodiment, the plan
Operation subelement is closed, is additionally operable to determine profile segment type between the weak angle point of each two by fitting algorithm；The angle point deletes son
Unit is additionally operable to when the type of two adjacent contour segments is identical, deletes the weak angle point between two contour segments.
The function of modules and unit and the realization process of effect refer to corresponding reality in the above method in described device
Existing process, details are not described herein.
Summary, to solve the problems, such as that the relevant technologies can not accurately extract angle point, the technical solution of the embodiment of the present invention
After image outline is extracted, " k- cosine curvature " method is used to extract notable angle point from the profile point of image outline first,
In, the notable angle point is more than the point of preset curvature for curvature；Then, using notable angle point described in fitting algorithm generation each two
Between profile point corresponding contour curve data are described, when the contour curve describe data and normal data difference be more than it is pre-
If during threshold value, illustrate in the profile point there may be it is unidentified go out angle point, so as to according to preset rules from the profile point
The middle weak angle point of extraction, wherein, the weak angle point is less than the point of preset curvature for curvature.It is obvious that the technology of the embodiment of the present invention
Scheme extracts the angle point in image outline according to amount of curvature in two stages, moreover, the extraction conditions in two extraction stages
Correspond to the angle point setting of different condition respectively so that formed between two extraction stages it is complementary, so as to will be in image outline
Angle point more accurately extract comprehensively, the contour segment ensured is more accurate, and then can improve the detection of PCB substrate
Precision.
The above is only the specific embodiment of the present invention, it is noted that for the ordinary skill people of the art
For member, various improvements and modifications may be made without departing from the principle of the present invention, these improvements and modifications also should
It is considered as protection scope of the present invention.
Claims (8)
1. a kind of Angular Point Extracting Method based on image outline, which is characterized in that including：
After image outline is extracted, " the k- cosine curvature " method of use extracts notable angle point from the profile point of image outline,
In, the notable angle point is more than the point of preset curvature for curvature；
Data are described using the corresponding contour curve of profile point between notable angle point described in fitting algorithm generation each two；
When the difference that the contour curve describes data and normal data is more than predetermined threshold value, according to preset rules from the wheel
Weak angle point is extracted in exterior feature point, wherein, the weak angle point is less than the point of preset curvature for curvature；
It is described that weak angle point is extracted from the profile point according to preset rules, including：
Call the first sub- preset rules；
The weak angle point is extracted from the profile point according to the described first sub- preset rules；
It is described that the weak angle point is extracted from the profile point according to the described first sub- preset rules, including：
Curvature peak value is obtained from the corresponding curvature curve of the profile point；
Calculate the length value of continuous section that the curvature peak value both sides curvature value is zero；
It, will when the length value of continuous section that the curvature peak value both sides curvature value is zero is more than or equal to the first preset value
The corresponding profile point of the curvature peak value is determined as weak angle point.
2. according to the method described in claim 1, it is characterized in that, it is described extracted from the profile point according to preset rules it is weak
Angle point further includes：
When not extracting the weak angle point from the profile point according to the described first sub- preset rules, call the second son default
Rule；
The weak angle point is extracted from the profile point according to the described second sub- preset rules.
3. according to the method described in claim 2, it is characterized in that, according to the described second sub- preset rules from the profile point
The weak angle point is extracted, including：
Trapezoidal curvature curve is obtained from the corresponding curvature curve of the profile point；
Calculate the length value of the trapezoidal curvature curve mean curvature peak value；
When the length value of the trapezoidal curvature curve mean curvature peak value is more than or equal to the second preset value, determine described trapezoidal
Curvature curve mean curvature value rises section and curvature value declines the corresponding profile point range of section；
Operation is fitted to the profile point in the range of the profile point respectively and obtains description data；
The profile point of the description data and the difference minimum of normal data is determined as weak angle point.
4. according to the method described in claim 3, it is characterized in that, in the length for calculating the trapezoidal curvature curve mean curvature peak value
After angle value, determining that the trapezoidal curvature curve mean curvature value rises section and curvature value declines the corresponding profile point model of section
Before enclosing, further include：
The radius of the corresponding contour curve of the profile point is calculated using fitting algorithm；
Judge whether the radius is less than default support width, wherein, when the default support width is calculates profile point curvature,
The span value of profile point to be calculated and selected profile point；
If the radius is less than the default support width, performs the determining trapezoidal curvature curve mean curvature value and rise
Section and curvature value decline the step of section corresponding profile point range；
If the radius is more than or equal to default support width, by the N number of detection support width value of default step-length setting, and press
The curvature value of the profile point is calculated according to the first detection support width value, wherein, N is natural number, detection support width value from 1 to
N gradually increases；
When curvature value continuous phase when profile point quantity be more than third preset value when, by the profile of the curvature value continuous phase etc.
Point is determined as profile point to be detected；
The song of the profile point to be detected is sequentially computed repeatedly to N detections support width value according to the second detection support width value
Rate value；
When being worth obtained curvature value and discontinuous phase etc. according to I detection support width, I -1 is detected into support width value
Corresponding curvature curve rise, only profile point be determined as weak angle point, wherein, N >=I>1.
5. method according to claim 3 or 4, which is characterized in that further include：
Profile segment type between the weak angle point of each two is determined by fitting algorithm；
When the type of two adjacent contour segments is identical, the weak angle point between two contour segments is deleted.
6. a kind of angle point grid device based on image outline, which is characterized in that including：
Extraction module, for after image outline is extracted, " the k- cosine curvature " method of use to be carried from the profile point of image outline
Notable angle point is taken, wherein, the notable angle point is more than the point of preset curvature for curvature；
Generation module, for generating profile point between the notable angle point of each two that the extraction module extracted using fitting algorithm
Corresponding contour curve describes data；
The extraction module, the contour curve for being additionally operable to be generated when the generation module describe data and the difference of normal data
During more than predetermined threshold value, weak angle point is extracted from the profile point according to preset rules, wherein, the weak angle point is less than for curvature
The point of preset curvature；
The extraction module includes call unit and extraction unit, wherein,
The call unit, for calling the first sub- preset rules；
The extraction unit, for extracting the weak angle point from the profile point according to the described first sub- preset rules；
The extraction unit includes obtaining subelement, computation subunit and determination subelement, wherein,
The acquisition subelement, for obtaining curvature peak value from the corresponding curvature curve of the profile point；
The computation subunit, for calculating the length value of continuous section that the curvature peak value both sides curvature value is zero；
The determination subelement, for work as the length value of continuous section that the curvature peak value both sides curvature value is zero be more than or
During equal to the first preset value, the corresponding profile point of the curvature peak value is determined as weak angle point.
7. device according to claim 6, which is characterized in that the extraction module includes call unit and extraction unit,
Wherein,
The call unit is additionally operable to work as and does not extract the weak angle from the profile point according to the described first sub- preset rules
During point, the second sub- preset rules are called；
The extraction unit is additionally operable to extract the weak angle point from the profile point according to the described second sub- preset rules.
8. device according to claim 6, which is characterized in that the extraction unit further includes：Fitting operation subelement,
In,
The acquisition subelement is additionally operable to obtain trapezoidal curvature curve from the corresponding curvature curve of the profile point；
The computation subunit is additionally operable to calculate the length value of the trapezoidal curvature curve mean curvature peak value；
The determination subelement is additionally operable to be more than or equal to second when the length value of the trapezoidal curvature curve mean curvature peak value
During preset value, determine that the trapezoidal curvature curve mean curvature value rises section and curvature value declines the corresponding profile point model of section
It encloses；
The fitting operation subelement, is described for being fitted operation to the profile point in the range of the profile point respectively
Data；
The determination subelement, the profile point for being additionally operable to the difference minimum by the description data and normal data are determined as weak angle
Point.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510673750.XA CN105405117B (en) | 2015-10-16 | 2015-10-16 | Angular Point Extracting Method and device based on image outline |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510673750.XA CN105405117B (en) | 2015-10-16 | 2015-10-16 | Angular Point Extracting Method and device based on image outline |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105405117A CN105405117A (en) | 2016-03-16 |
| CN105405117B true CN105405117B (en) | 2018-07-03 |
Family
ID=55470584
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510673750.XA Active CN105405117B (en) | 2015-10-16 | 2015-10-16 | Angular Point Extracting Method and device based on image outline |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105405117B (en) |
Families Citing this family (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106845494B (en) * | 2016-12-22 | 2019-12-13 | 歌尔科技有限公司 | Method and device for detecting contour corner points in image |
| CN109509165B (en) * | 2017-09-11 | 2021-01-29 | 凌云光技术股份有限公司 | Image positioning area selection method and device |
| CN109509166B (en) * | 2017-09-15 | 2021-05-04 | 凌云光技术股份有限公司 | Printed circuit board image detection method and device |
| CN109543665B (en) * | 2017-09-22 | 2020-10-16 | 凌云光技术集团有限责任公司 | Image positioning method and device |
| CN107967679B (en) * | 2017-11-21 | 2020-07-03 | 凌云光技术集团有限责任公司 | Method for automatically selecting positioning core based on vector graph of PCB product |
| CN108010018A (en) * | 2017-11-27 | 2018-05-08 | 浙江华睿科技有限公司 | The notch detection method and device of a kind of silicon chip |
| CN108010043A (en) * | 2017-12-06 | 2018-05-08 | 凌云光技术集团有限责任公司 | A kind of image outline dividing method and device |
| JP2019199663A (en) * | 2018-05-16 | 2019-11-21 | Juki株式会社 | Inspection device |
| CN109886182A (en) * | 2019-02-19 | 2019-06-14 | 北方民族大学 | Brake disc identification and localization method under a kind of complex environment |
| CN111390248B (en) * | 2020-03-18 | 2021-12-10 | 广州兴森快捷电路科技有限公司 | Processing method and device for circuit board convex angle, computer equipment and storage medium |
| CN112308797B (en) * | 2020-10-30 | 2024-02-02 | 维沃移动通信有限公司 | Corner detection method and device, electronic equipment and readable storage medium |
| CN112700443A (en) * | 2021-02-10 | 2021-04-23 | 北京平恒智能科技有限公司 | Arc corner detection method based on two-dimensional image |
| CN113449644A (en) * | 2021-06-29 | 2021-09-28 | 中铁隧道集团三处有限公司 | Method for extracting contour of common building in city by airborne laser point cloud |
| CN114666498A (en) * | 2022-02-28 | 2022-06-24 | 超级视线科技有限公司 | Video image acquisition jitter processing method and system |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101339606A (en) * | 2008-08-14 | 2009-01-07 | 北京中星微电子有限公司 | Human face critical organ contour characteristic points positioning and tracking method and device |
| CN101887586A (en) * | 2010-07-30 | 2010-11-17 | 上海交通大学 | Self-adaptive angular-point detection method based on image contour sharpness |
| CN102542558A (en) * | 2010-12-30 | 2012-07-04 | 北京大学 | Method and system for detecting angular point of figure |
| CN103345628A (en) * | 2013-07-29 | 2013-10-09 | 苏州大学 | Target recognition and shape retrieval method based on hierarchical description |
| CN103345755A (en) * | 2013-07-11 | 2013-10-09 | 北京理工大学 | Chessboard angular point sub-pixel extraction method based on Harris operator |
| CN103487034A (en) * | 2013-09-26 | 2014-01-01 | 北京航空航天大学 | Method for measuring distance and height by vehicle-mounted monocular camera based on vertical type target |
| CN103499337A (en) * | 2013-09-26 | 2014-01-08 | 北京航空航天大学 | Vehicle-mounted monocular camera distance and height measuring device based on vertical target |
| CN104424473A (en) * | 2013-09-06 | 2015-03-18 | 北京三星通信技术研究有限公司 | Method and device for identifying and editing freehand sketch |
-
2015
- 2015-10-16 CN CN201510673750.XA patent/CN105405117B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101339606A (en) * | 2008-08-14 | 2009-01-07 | 北京中星微电子有限公司 | Human face critical organ contour characteristic points positioning and tracking method and device |
| CN101887586A (en) * | 2010-07-30 | 2010-11-17 | 上海交通大学 | Self-adaptive angular-point detection method based on image contour sharpness |
| CN102542558A (en) * | 2010-12-30 | 2012-07-04 | 北京大学 | Method and system for detecting angular point of figure |
| CN103345755A (en) * | 2013-07-11 | 2013-10-09 | 北京理工大学 | Chessboard angular point sub-pixel extraction method based on Harris operator |
| CN103345628A (en) * | 2013-07-29 | 2013-10-09 | 苏州大学 | Target recognition and shape retrieval method based on hierarchical description |
| CN104424473A (en) * | 2013-09-06 | 2015-03-18 | 北京三星通信技术研究有限公司 | Method and device for identifying and editing freehand sketch |
| CN103487034A (en) * | 2013-09-26 | 2014-01-01 | 北京航空航天大学 | Method for measuring distance and height by vehicle-mounted monocular camera based on vertical type target |
| CN103499337A (en) * | 2013-09-26 | 2014-01-08 | 北京航空航天大学 | Vehicle-mounted monocular camera distance and height measuring device based on vertical target |
Non-Patent Citations (2)
| Title |
|---|
| 一种改进的角点提取方法;王建琦 等;《上海交通大学学报》;20000731;第34卷(第7期);第913-916页 * |
| 一种新的Harris多尺度角点检测;张小洪 等;《电子与信息学报》;20070731;第29卷(第7期);第1735-1738页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105405117A (en) | 2016-03-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105405117B (en) | Angular Point Extracting Method and device based on image outline | |
| CN107702663B (en) | Point cloud registration method based on rotating platform with mark points | |
| CN101692257A (en) | Method for registering complex curved surface | |
| CN111292379A (en) | Depth camera correction device and method thereof | |
| CN114220757B (en) | Wafer detection alignment method, device and system and computer medium | |
| CN109447061A (en) | Reactor oil level indicator recognition methods based on crusing robot | |
| CN104517092A (en) | Method and system for detecting QR (quick response) code correction graphics | |
| CN105787939A (en) | Rapid positioning detection method for PCB circular hole | |
| CN108871185A (en) | Method, apparatus, equipment and the computer readable storage medium of piece test | |
| CN105160107B (en) | It is a kind of to wait resistance wiring method and device for flat-panel monitor | |
| TW201500978A (en) | Method for determining incline angle of electromagnetic pointer | |
| CN102663756A (en) | Registration method of special shaped elements and high-density packing components in printed circuit board | |
| CN104550046B (en) | Point stack device of circuit board and point heaping method of circuit board | |
| CN102510676B (en) | Method for adding teardrop in printed circuit board (PCB) during computer aided manufacturing (CAM) | |
| CN116206993A (en) | Wafer ink dot marking method and device | |
| CN105912783A (en) | Method and device for detection of PCB design scheme | |
| CN108960226A (en) | A kind of pointer instrument class expression value number reading method and device | |
| CN107527843B (en) | Method and system for realizing wafer laser identification process based on computer aided design | |
| CN114266162A (en) | Method for solving reflection flux in etching simulation model | |
| CN112581478A (en) | Centroid-based road center line extraction method | |
| CN106507588A (en) | A kind of method of automatic detection unnecessary component in PCB | |
| CN110377978A (en) | The method and system of 0402 resistor-capacitor unit is placed in a kind of PCB design under BGA package | |
| CN110264531A (en) | A kind of catching for X-comers takes method, apparatus, system and readable storage medium storing program for executing | |
| CN116563419B (en) | Correction method and device for wafer map configuration data, electronic equipment and storage medium | |
| CN116070071B (en) | Wall function selection method, device, terminal equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CP03 | Change of name, title or address | ||
| CP03 | Change of name, title or address |
Address after: 100094 Beijing city Haidian District Cui Hunan loop 13 Hospital No. 7 Building 7 room 701 Patentee after: Lingyunguang Technology Co., Ltd Address before: 100094 Beijing city Haidian District road Tamatsu wisdom Valley Center Building 2 Patentee before: LUSTER LIGHTTECH GROUP Co.,Ltd. |