CN109325930A - Detection method, device and the detection device of boundary defect - Google Patents
Detection method, device and the detection device of boundary defect Download PDFInfo
- Publication number
- CN109325930A CN109325930A CN201811062811.9A CN201811062811A CN109325930A CN 109325930 A CN109325930 A CN 109325930A CN 201811062811 A CN201811062811 A CN 201811062811A CN 109325930 A CN109325930 A CN 109325930A
- Authority
- CN
- China
- Prior art keywords
- image
- boundary
- defect
- straight line
- region
- 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.)
- Granted
Links
- 230000007547 defect Effects 0.000 title claims abstract description 177
- 238000001514 detection method Methods 0.000 title claims abstract description 134
- 238000000034 method Methods 0.000 claims abstract description 29
- 239000000284 extract Substances 0.000 claims description 22
- 238000013507 mapping Methods 0.000 claims description 12
- 238000001914 filtration Methods 0.000 claims description 10
- 238000000605 extraction Methods 0.000 claims description 9
- 238000004891 communication Methods 0.000 description 9
- 238000010586 diagram Methods 0.000 description 7
- 230000001788 irregular Effects 0.000 description 7
- 238000007689 inspection Methods 0.000 description 5
- 238000012545 processing Methods 0.000 description 5
- 238000003708 edge detection Methods 0.000 description 4
- 230000015572 biosynthetic process Effects 0.000 description 3
- 230000002950 deficient Effects 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000002093 peripheral effect Effects 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 239000007787 solid Substances 0.000 description 2
- 208000003464 asthenopia Diseases 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000000802 evaporation-induced self-assembly Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 238000001028 reflection method Methods 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 230000001052 transient effect Effects 0.000 description 1
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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/10—Image enhancement or restoration using non-spatial domain filtering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/13—Edge detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
- G06T7/62—Analysis of geometric attributes of area, perimeter, diameter or volume
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/70—Determining position or orientation of objects or cameras
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Geometry (AREA)
- Quality & Reliability (AREA)
- Image Processing (AREA)
- Image Analysis (AREA)
Abstract
The present invention provides a kind of detection method of boundary defect, device and detection devices, wherein method includes obtaining image to be detected；The borderline region in two adjacent images region is extracted, to obtain borderline region image；Borderline region graph line is unfolded, to form straight line unfolded image；Based on straight line unfolded image, the detection of boundary defect is carried out；When detecting that straight line unfolded image there are when boundary defect, using borderline region image and straight line unfolded image, determines position of the boundary defect in image to be detected.Extend the first pre-determined distance and the second pre-determined distance to two adjacent images region respectively by the normal direction of pixel each along boundary line, obtain the first boundary line and the second boundary line, i.e. by the way that suitable border detection range is arranged, all boundary defects of image to be detected are divided in borderline region, it is subsequent by borderline region straight line expansion after, it can completely detect the defect across zone boundary, improve the accuracy of detection.
Description
Technical field
The present invention relates to defects detection fields, and in particular to a kind of detection method of boundary defect, device and detection device.
Background technique
Due to industrial products surface defect can aesthetics, comfort level and service performance etc. to product bring bad shadow
It rings, therefore manufacturing enterprise can detect the surface defect of product before product export, to find and to be controlled in time
System.Wherein, most common in surface defect is exactly boundary defect, and the type that boundary defect is generated according to defect is divided into again draws
Trace, chipping, foreign matter, dirty etc..
Traditional defects detection relies on artificial detection more, not only low efficiency, but also can be because of human eye fatigue or master
Reason in sight causes erroneous detection or missing inspection.Therefore, with the rapid development of Digital image technology, the defect inspection based on machine vision
Survey method obtains more and more extensive research and application due to its high accuracy and high efficiency in modern industry.
In the prior art frequently with the detection method of boundary defect be the method compared based on template image, specifically:
Obtaining a width first does not have defective image as template image, then passes through image alignment algorithm calculation template image and detection
Relative positional relationship between image calculates the difference of detection image and template image, differential image is obtained, in differential image
Carry out Defect Search.
However, inventor has found in the research process of the detection method to above-mentioned boundary defect, due to industrial environment
Uncertainty, the diversity of product, the complexity of technique make the robustness of existing defect inspection method poor, so as to cause existing
The accuracy of defective detection method is lower.
Summary of the invention
In view of this, the embodiment of the invention provides a kind of detection method of boundary defect, device and detection device, with solution
The lower problem of the accuracy of certainly existing defect inspection method.
For this purpose, the embodiment of the invention provides following technical solutions:
First aspect present invention provides a kind of detection method of boundary defect, comprising:
Obtain image to be detected；Wherein, described image to be detected includes at least two image-regions；
The borderline region in two neighboring described image region is extracted, to obtain borderline region image；
The borderline region graph line is unfolded, to form straight line unfolded image；
Based on the straight line unfolded image, the detection of the boundary defect is carried out；
When detecting that the straight line unfolded image there are when the boundary defect, utilizes the borderline region image and institute
Straight line unfolded image is stated, determines position of the boundary defect in described image to be detected.
The detection method of boundary defect provided in an embodiment of the present invention passes through the borderline region to two adjacent images region
Straight line expansion is carried out, i.e., irregular borderline region image is converted into the borderline region image of rule, can be improved detection side
The robustness of method, and then improve the accuracy of detection method；In addition, can not only be realized by the defects detection to borderline region
Immediate area boundary or on zone boundary defect detection, additionally it is possible to realize the complete detection of the defect on trans-regional boundary,
Detection accuracy with higher.
With reference to first aspect, in first aspect first embodiment, the two neighboring described image region of extraction
Borderline region, to obtain borderline region image, comprising:
Obtain the normal and seat of each pixel on the boundary line and the boundary line in two neighboring described image region
Mark；Wherein, the coordinate is uniquely determined by the first reference frame；
The normal direction of each pixel along the boundary line, by the boundary line respectively to two neighboring described image
Region deviates the first pre-determined distance and the second pre-determined distance, to obtain the first boundary line and the second boundary line；
The region between first boundary line and the second boundary line is extracted, to obtain the borderline region figure
Picture.
The detection method of boundary defect provided in an embodiment of the present invention passes through the normal side of pixel each along boundary line
To the first pre-determined distance and the second pre-determined distance is extended to two adjacent images region respectively, obtain the first boundary line and
All boundary defects of image to be detected are divided in side that is, by the way that suitable border detection range is arranged by the second boundary line
It is subsequent after by the expansion of borderline region straight line in battery limit (BL) domain, it can completely detect the defect across zone boundary, improve
The accuracy of detection.
First embodiment with reference to first aspect, it is described by the borderline region in first aspect second embodiment
Graph line expansion, to form straight line unfolded image, comprising:
Using presetted pixel point on the boundary line as starting point, boundary point is sequentially mapped to second with reference to seat along first straight line
Mark is fastened；Wherein, the boundary point is the pixel of the normal and the boundary line point of intersection, the normal and the boundary
Point corresponds；
Successively calculate the distance between each pixel on the normal and the corresponding boundary point；
Using the distance, all pixels point on the borderline region image is successively mapped to described second with reference to seat
Mark is fastened, to form the straight line unfolded image；Wherein, the pixel in the borderline region image and the straight line expanded view
Pixel as in corresponds.
The detection method of boundary defect provided in an embodiment of the present invention, using normal by all pictures on borderline region image
Vegetarian refreshments maps to the second reference coordinate and fastens, since normal is vertical with the tangent line of corresponding pixel points, using normal as straight line exhibition
Open according to accuracy with higher.
Second embodiment with reference to first aspect, it is described to utilize the distance in first aspect third embodiment, according to
The secondary all pixels point by the borderline region image is mapped to second reference coordinate and fastens, comprising:
Successively extract the pixel on the normal；
Inquire mapped boundaries point corresponding with the normal；Wherein, the mapped boundaries point is corresponding with the normal
The mapping point that the boundary point is fastened in second reference coordinate；
The pixel extracted second reference coordinate is mapped to along second straight line to fasten；Wherein, it extracts
The pixel mapping point and the mapped boundaries point be separated by the distance, the second straight line is vertical with first straight line,
The second straight line and the mapped boundaries point correspond.
First embodiment with reference to first aspect, it is described to be based on the straight line exhibition in the 4th embodiment of first aspect
Image is opened, the detection of the boundary defect is carried out, comprising:
The straight line unfolded image is filtered, filtering image is formed；
Compare the straight line unfolded image and the filtering image, obtains differential image；
Filter out the region for meeting preset condition in the differential image；Wherein, the region filtered out is to detect
The boundary defect region.
The detection method of boundary defect provided in an embodiment of the present invention, in differential image, since boundary defect is corresponding
Pixel has a greater change relative to background, and corresponding difference value is larger, by preset condition (for example, preset defect
Area, length threshold etc.) region for meeting condition that filters out is the defect detected；It is determined using apparent difference value
The defect for meeting condition out can be improved the efficiency of detection method.
First embodiment with reference to first aspect, it is described to utilize the frontier district in the 5th embodiment of first aspect
Area image and the straight line unfolded image, determine position of the boundary defect in described image to be detected, comprising:
Extract the coordinate that all pixels on the region of the boundary defect o'clock are fastened in the second reference coordinate；
The coordinate that the coordinate extracted and the first reference coordinate are fastened is matched, with the determination boundary defect described
Position in image to be detected.
The detection method of boundary defect provided in an embodiment of the present invention can be determined quickly using the method for coordinate matching
Position of the boundary defect in image to be detected out, detection efficiency and Detection accuracy with higher.
First embodiment with reference to first aspect, it is described to obtain two neighboring institute in first aspect sixth embodiment
State the boundary line of image-region, comprising:
Obtain zero defect template image；Wherein, the zero defect template image and described image to be detected correspond；
Extract the boundary line in the two adjacent images region of the zero defect template image；
It is aligned the zero defect image and described image to be detected；
The boundary line extracted is mapped in described image to be detected；Wherein, the boundary line extracted is
The boundary line in the two neighboring described image region of described image to be detected.
The detection method of boundary defect provided in an embodiment of the present invention is mapped as using the boundary line of zero defect template image
The boundary line of image to be detected, the division that can be avoided the boundary line caused by the boundary defect of image to be detected are inaccurate
Really, and then subsequent detection accuracy is influenced.
According to second aspect, the embodiment of the invention also provides a kind of detection devices of boundary defect, comprising:
Module is obtained, for obtaining image to be detected；Wherein, described image to be detected includes at least two image-regions；
Extraction module, for extracting the borderline region in two neighboring described image region, to obtain borderline region image；
Module is unfolded in straight line, for the borderline region graph line to be unfolded, to form straight line unfolded image；
Detection module carries out the detection of the boundary defect for being based on the straight line unfolded image；
Boundary defect determining module, for when detecting that the straight line unfolded image there are when the boundary defect, utilizes
The borderline region image and the straight line unfolded image, determine position of the boundary defect in described image to be detected
It sets.
The detection device of boundary defect provided in an embodiment of the present invention passes through the borderline region to two adjacent images region
Straight line expansion is carried out, i.e., irregular borderline region image is converted into the borderline region image of rule, can be improved detection side
The robustness of method, and then improve the accuracy of detection method；In addition, can not only be realized by the defects detection to borderline region
Immediate area boundary or on zone boundary defect detection, additionally it is possible to realize the complete detection of the defect on trans-regional boundary,
Detection accuracy with higher.
According to the third aspect, the embodiment of the invention also provides a kind of detection devices, comprising: at least one processor；With
And the memory being connect at least one described processor communication；Wherein, be stored with can be by one processing for the memory
The instruction that device executes, described instruction are executed by least one described processor, so that the execution of at least one described processor is above-mentioned
The detection method of boundary defect described in first aspect or first aspect any embodiment.
According to fourth aspect, the embodiment of the invention also provides a kind of computer readable storage mediums, are stored thereon with meter
The instruction of calculation machine, the instruction are executed by processor boundary defect described in above-mentioned first aspect or first aspect any embodiment
Detection method the step of.
Detailed description of the invention
It, below will be to specific in order to illustrate more clearly of the specific embodiment of the invention or technical solution in the prior art
Embodiment or attached drawing needed to be used in the description of the prior art be briefly described, it should be apparent that, it is described below
Attached drawing is some embodiments of the present invention, for those of ordinary skill in the art, before not making the creative labor
It puts, is also possible to obtain other drawings based on these drawings.
Fig. 1 is the flow chart of the detection method of boundary defect according to an embodiment of the present invention；
Fig. 2 is the schematic diagram of image to be detected according to an embodiment of the present invention；
Fig. 3 is the schematic diagram of borderline region image according to an embodiment of the present invention；
Fig. 4 is the flow chart of the detection method of boundary defect according to an embodiment of the present invention；
Fig. 5 is the schematic diagram of borderline region image according to an embodiment of the present invention；
Fig. 6 is the schematic diagram of straight line unfolded image according to an embodiment of the present invention；
Fig. 7 a- Fig. 7 c is the formation schematic diagram of differential image according to an embodiment of the present invention；
Fig. 8 is the flow chart of the BL method according to an embodiment of the present invention for obtaining two adjacent images region；
Fig. 9 is the flow chart of the detection method of boundary defect according to an embodiment of the present invention；
Figure 10 is the structure chart of the detection device of boundary defect according to an embodiment of the present invention；
Figure 11 is the structure chart of the detection device of boundary defect according to an embodiment of the present invention；
Figure 12 is the hardware structural diagram of detection device provided in an embodiment of the present invention；
Appended drawing reference:
The boundary line 10-；The first boundary 11- line；12- the second boundary line；13- normal；
10 '-mapped boundaries lines；11 '-the first boundary lines of mapping；12 '-mapping the second boundary lines；13 '-reflection methods
Line；
A1, A2, A3, A4, A5- boundary point；A6- pixel；
A1 ', A2 ', A3 ', A4 ', A5 '-mapped boundaries point；A6 '-maps pixel；
B1, B2- defect；
B1 ', B2 '-maps defects.
Specific embodiment
In order to make the object, technical scheme and advantages of the embodiment of the invention clearer, below in conjunction with the embodiment of the present invention
In attached drawing, technical scheme in the embodiment of the invention is clearly and completely described, it is clear that described embodiment is
A part of the embodiment of the present invention, instead of all the embodiments.Based on the embodiments of the present invention, those skilled in the art are not having
Every other embodiment obtained under the premise of creative work is made, shall fall within the protection scope of the present invention.
According to embodiments of the present invention, a kind of detection method embodiment of boundary defect is provided, it should be noted that attached
The step of process of figure illustrates can execute in a computer system such as a set of computer executable instructions, though also,
So logical order is shown in flow charts, but in some cases, it can be to be different from shown by sequence execution herein
Or the step of description.
A kind of detection method of boundary defect is provided in the present embodiment, can be used in detection device, and Fig. 1 is according to this
The flow chart of the boundary defect detection of inventive embodiments, as shown in Figure 1, the process includes the following steps:
S11 obtains image to be detected.
Wherein, image to be detected includes at least two image-regions.
Detection device can get image to be detected through a variety of ways, for example, can be clapped by CCD or CMOS camera
It takes the photograph, be also possible to other image preprocessings as a result, need to only guarantee that image to be detected accessed by the detection device includes extremely
Few two image-regions.
So-called image-region, i.e., in region the value of pixel clearly distinguish with region outside pixel, for example, acquired
To image to be detected in image-region be divided into: background image locating for the image and product to be detected of product to be detected.
For example, as shown in Fig. 2, image to be detected includes two image-regions, i.e. region 1 and region 2, wherein region 1 is to be detected
The image of product, region 2 are that background image, region 1 and region 2 are divided by boundary line.
S12 extracts the borderline region in two adjacent images region, to obtain borderline region image.
Detection device successively extracts the borderline region in two adjacent images region after getting image to be detected, with
Obtain borderline region image.Wherein, borderline region is the image that two adjacent regions form.
For example, being deviated into region 2 as shown in figure 3, a distance can be deviated into region 1 on the basis of boundary line
A distance forms the region of fill part in Fig. 3, as borderline region image.In addition, borderline region image, is also possible to
Part is extracted from region 1, while extracting part from region 2, need to only guarantee the part phase extracted from two regions respectively
It is adjacent.
Alternatively, part borderline region can also be extracted only, such as has learned that exist in the part range on boundary in advance
Defect, then when carrying out borderline region extraction, it is only necessary to the borderline region for extracting existing defects, it is whole without extracting
A borderline region.
Borderline region graph line is unfolded S13, to form straight line unfolded image.
Detection device is unfolded after extracting borderline region image, by borderline region image along straight line, will be irregular
Borderline region image is converted into the straight line unfolded image of rule.It specifically, can be with a certain pixel in borderline region image
As the starting point of expansion, the neighbor pixel of a certain pixel is then successively extracted, is closed according to the position between two pixels
System places along straight line；It is also possible to the starting point using all pixels point on boundary line as expansion, then using on boundary line
All pixels point on the corresponding normal of pixel and normal successively realizes that all pixels point is again in borderline region image
Arrangement, to form straight line unfolded image；Alternatively, can also realize using other modes, only it need to guarantee that detection device will be irregular
Borderline region image be converted into rule straight line unfolded image.
S14 is based on straight line unfolded image, carries out the detection of boundary defect.
Detection device can carry out the inspection of boundary defect after forming straight line unfolded image to the straight line unfolded image
It surveys；For example, zero defect image corresponding with image to be detected can be used, the matching of zero defect image is extracted according to above-mentioned steps
Image compares straight line unfolded image and matching image, can be detected out the boundary defect on straight line unfolded image；It can also be to straight
Line unfolded image is filtered, and is then utilized the differential image of straight line unfolded image and filtered image, is carried out boundary defect
Detection；Or the detection etc. of boundary defect is carried out using other modes.
S15, when detecting that straight line unfolded image there are when boundary defect, utilizes borderline region image and straight line expanded view
Picture determines position of the boundary defect in image to be detected.
When detection device detects in straight line unfolded image there are when boundary defect, according to the operation opposite with S13, by side
Boundary defect in battery limit (BL) area image navigates in straight line unfolded image, that is, can determine that boundary defect in image to be detected
Position, to realize the detection to the boundary defect of image to be detected.
The detection method of boundary defect provided in an embodiment of the present invention passes through the borderline region to two adjacent images region
Straight line expansion is carried out, i.e., irregular borderline region image is converted into the borderline region image of rule, can be improved detection side
The robustness of method, and then improve the accuracy of detection method；In addition, can not only be realized by the defects detection to borderline region
Immediate area boundary or on zone boundary defect detection, additionally it is possible to realize the complete detection of the defect on trans-regional boundary,
Detection accuracy with higher.
It, can also be with it should be noted that region included by image to be detected in the embodiment of the present invention is not limited to two
It is 3 or multiple, the detection method of above-mentioned boundary defect is used for wherein every two adjacent region, carries out boundary defect
Detection, can be detected out all boundary defects in image to be detected.
The embodiment of the invention also provides a kind of detection methods of boundary defect, as shown in figure 4, this method comprises:
S21 obtains image to be detected.
Wherein, image to be detected includes at least two image-regions.The S11 of embodiment illustrated in fig. 1 is referred to, herein
It repeats no more.
S22 extracts the borderline region in two adjacent images region, to obtain borderline region image.
Wherein, for two adjacent images region borderline region extraction, specifically includes the following steps:
S221 obtains the normal and coordinate of each pixel on the boundary line and boundary line in two adjacent images region.
Wherein, image to be detected is made of several pixels.Detection device is using edge detection operator (for example, can adopt
With Canny edge detection operator or other edge detection operators), extract the boundary line and side in two adjacent images region
The normal of each pixel in boundary line.
In addition, the coordinate of each pixel is uniquely determined by the first reference frame, coordinate is for indicating mapping to be checked
Relative positional relationship as between pixel.Specifically, the origin of the first reference frame can be in image to be detected
Heart point；Or, the origin of the first reference frame can be the upper left corner of image to be detected when image to be detected is rectangle, or
The pixel in the lower right corner, or any pixel point etc. in image to be detected；Only need the first reference frame of utility
The coordinate of each pixel on boundary line can be uniquely determined out.Side can not only be determined by the first reference frame
The coordinate of each pixel in boundary line, is also capable of determining that the coordinate of each pixel in image to be detected.
S222, the normal direction of each pixel along boundary line are inclined to two adjacent images region respectively by boundary line
The first pre-determined distance and the second pre-determined distance are moved, to obtain the first boundary line and the second boundary line.
Specifically as shown in figure 5, extracting adjacent two using edge detection operator when knowing the Position Approximate of defect in advance
Segment boundary line 10 is extended the first pre-determined distance L1 to two adjacent images region respectively by the boundary line 10 of a image-region,
And the second pre-determined distance L2.Wherein, the specific distance of L1 and L2 can be specifically arranged according to the actual situation, only need to protect
It demonstrate,proves L1 and is greater than the maximum distance that defect B1 and defect B2 arrives boundary line 10 in region 1, L2 is greater than defect B2 and arrives in region 2
The maximum distance of boundary line 10.Wherein, when L1, which is greater than defect B1, arrives the maximum distance of boundary line 10 in region 1, after
Continue after straight line expansion, can effectively detect the defect at borderline region；Boundary is arrived in region 2 when L2 is greater than defect B2
It is subsequent after straight line expansion when the maximum distance of line 10, it can completely detect the defect in crossing the boundary region.
It, can be whole pre- to the offset of region 1 first by boundary line 10 when not knowing the defects of image to be detected in advance
If distance, the second pre-determined distance is deviated to region 2, specifically may refer to shown in Fig. 3.
S223 extracts the region between the first boundary line and the second boundary line, to obtain borderline region image.
As shown in figure 5, between the first boundary line, the second boundary line and boundary point A1 and boundary point A5 is corresponding
The enclosed region of normal composition, as borderline region image, detection device extract the borderline region image, so that later use should
The detection of borderline region image progress boundary defect.
Borderline region graph line is unfolded S23, to form straight line unfolded image.
Specifically, the straight line unfolded image formed after the expansion of borderline region graph line is as shown in Figure 6.Wherein, correspond to
Fig. 5, mapped boundaries line 10 ' are corresponding with boundary line 10；It is corresponding with the first boundary line 11 to map the first boundary line 11 '；Mapping
The second boundary line 12 ' is corresponding with the second boundary line 12；It is corresponding with normal 13 to map normal 13 '；Maps defects B1 ' and
B2 ' is corresponding with defect B1 and B2；Mapped boundaries point A1 '-A5 ' is corresponding with boundary point A1-A5；Map pixel A6 ' and pixel
Point A6 is corresponding.
Remaining refers to the S13 of embodiment illustrated in fig. 1, and details are not described herein.
S24 is based on straight line unfolded image, carries out the detection of boundary defect.
Detection device is filtered straight line unfolded image and binary conversion treatment after forming straight line unfolded image
Afterwards, the detection of boundary defect can be carried out.It specifically includes:
S241 is filtered straight line unfolded image, forms filtering image.
Fig. 7 a shows straight line unfolded image, straight line unfolded image is filtered (for example, can using box filtering,
Mean filter or gaussian filtering etc.), it is as shown in Figure 7b to be formed by filtering image.
S242 compares straight line unfolded image and filtering image, obtains differential image.
After detection device compares straight line unfolded image and filtering image, differential image as shown in Figure 7 c is obtained.
S243 filters out the region for meeting preset condition in differential image.
As shown in Figure 7 c, in differential image, since the corresponding pixel of defect has a greater change relative to background, difference
It is worth larger, after doing binarization segmentation to differential image, the region of the condition of satisfaction is filtered out using preset condition, wherein filter out
Region be the boundary defect detected region.The defect for meeting condition is determined using apparent difference value, can be improved
The efficiency of detection method.
Specifically, the area threshold and/or length threshold that setting defect may be implemented meet when existing in differential image
When the region of any of the above-described preset condition, which is the region of the boundary defect detected.
S25, when detecting that straight line unfolded image there are when boundary defect, utilizes borderline region image and straight line expanded view
Picture determines position of the boundary defect in image to be detected.The S15 of embodiment illustrated in fig. 1 is referred to, it is no longer superfluous herein
It states.
Compared with embodiment illustrated in fig. 1, the detection method of boundary defect provided in this embodiment passes through every along boundary line
The normal direction of a pixel extends the first pre-determined distance and the second pre-determined distance to two adjacent images region respectively, obtains
First boundary line and the second boundary line, i.e., by the way that suitable border detection range is arranged, by all of image to be detected
Boundary defect is divided in borderline region, subsequent after by the expansion of borderline region straight line, can completely be detected across region
The defect on boundary improves the accuracy of detection.
As a kind of optional embodiment of the present embodiment, wherein can for the first boundary line and the second boundary line
To be arbitrary straight line, it is not limited to the offset of boundary line, need to only guarantee the area between the first boundary line and the second boundary line
Domain can cover all defect in image to be detected.
A kind of optional embodiment as the present embodiment, wherein as shown in figure 8, obtaining two adjacent images in S221
In the step of boundary line in region, for image to be detected of the product formation of irregular shape, comprising the following steps:
S41 obtains zero defect template image.
Wherein, zero defect template image and image to be detected correspond.
S42 extracts the boundary line in the two adjacent images region of zero defect template image.
S43, snap to none defect image and image to be detected.
Generally, due to it is mechanical or itself the reason of, product to be detected is formed by image to be detected, relative to zero defect
Template image may exist certain rotation, scaling, offset etc. distortion, can using the template matching method based on profile into
Row contraposition, to reduce the influence of distortion.
S44 maps to the boundary line extracted in image to be detected.
Wherein, the boundary line extracted is the boundary line in the two adjacent images region of image to be detected.
This method is mapped as the boundary line of image to be detected using the boundary line of zero defect template image, can be avoided due to
The division inaccuracy of boundary line caused by the boundary defect of image to be detected, and then influence subsequent detection accuracy.
As another optional embodiment of the present embodiment, for the product formation of regular shape image to be detected and
Speech, for example, in image to be detected two adjacent images region boundary be straight border, circular boundary, elliptical boundary etc.,
The extraction that boundary line can be directly carried out to image to be detected, then obtains boundary point by way of being fitted to borderline region
Coordinate and normal.
The embodiment of the invention also provides a kind of detection methods of boundary defect, as shown in figure 9, this method comprises:
S51 obtains image to be detected.
Wherein, image to be detected includes at least two image-regions.The S21 of embodiment illustrated in fig. 4 is referred to, herein
It repeats no more.
S52 extracts the borderline region in two adjacent images region, to obtain borderline region image.Refer to Fig. 4 institute
Show the S22 of embodiment, details are not described herein.
Borderline region graph line is unfolded S53, to form straight line unfolded image.
Wherein, borderline region image is as shown in figure 5, the straight line unfolded image formed after straight line expansion is as shown in Figure 6.Detection
Each pixel and the corresponding normal of pixel on equipment utilization boundary line carry out straight line expansion to frontier district area image.
Specifically, comprising:
Boundary point is sequentially mapped to the second reference along first straight line using presetted pixel point on boundary line as starting point by S531
On coordinate system.
Wherein, boundary point is the pixel of normal and boundary line point of intersection, and normal and boundary point correspond.
Specifically, as shown in figure 5, pixel A1-A5 on boundary line, the corresponding normal of each pixel, normal with
The intersection point of boundary line, that is, boundary point (or pixel A1-A5).For example, using boundary point A1 as starting point (i.e. presetted pixel point), successively
All boundary points on boundary line are mapped to the second reference frame along first straight line, form mapped boundaries point A1 '-A5 ', i.e.,
As shown in Figure 6.
S532 successively calculates the distance between each pixel and corresponding boundary point on normal.
As shown in figure 5, the corresponding normal 13 of boundary point A1, extracts normal 13 in the first boundary line 11 and the second boundary
The coordinate of all pixels point on line segment between line 12 is calculated using the coordinate of the coordinate and boundary point A1 that extract
The distance between each pixel and boundary point A1 on line segment.Successively calculate the pixel on the corresponding normal of remaining boundary point
Point, the distance between corresponding boundary point can be obtained between the pixel and corresponding boundary point on borderline region image
Relationship.
All pixels point on borderline region image is successively mapped to the second reference coordinate and fastened by S533 using distance,
To form straight line unfolded image.
Wherein, the pixel in borderline region image and the pixel in straight line unfolded image correspond.It specifically includes:
(1) pixel on normal is successively extracted.
For example, extracting the pixel A6 on the normal of boundary point A4, and calculated between A4 and A6 in S333
Distance is L.
(2) mapped boundaries point corresponding with normal is inquired；Wherein, mapped boundaries point is the boundary point corresponding with normal
In the mapping point that the second reference coordinate is fastened；
Using the pixel A6 extracted, inquire its corresponding boundary point in borderline region image is detection device
The mapped boundaries point of A4, A4 in the second reference frame is A4 '.Wherein, the origin of the second reference frame can be default
Pixel, or other points；Only need the second reference frame of utility that can uniquely determine out every on mapped boundaries line
The coordinate of a pixel.Each pixel on mapped boundaries line can not only be determined subsequently through the second reference frame
Coordinate, be also capable of determining that the coordinate of each pixel on straight line unfolded image.
(3) pixel extracted the second reference coordinate is mapped to along second straight line to fasten；Wherein, the pixel extracted
The mapping point and mapped boundaries point standoff distance of point, second straight line is vertical with first straight line, second straight line and mapped boundaries point one
One is corresponding.
For example, as shown in fig. 6, mapped boundaries point A4 ' is upwardly formed second straight line in the side vertically with first straight line, by picture
Vegetarian refreshments A6 is mapped in second straight line, wherein mapping the distance between pixel A6 ' and mapped boundaries point A4 ' are L.
It constantly repeats the above steps, until all pixels point in borderline region image all maps to the second reference coordinate
Until in system.
Detection device is recorded after the mapping for completing all pixels point and is formed by all pictures in straight line unfolded image
Coordinate of the vegetarian refreshments in the second reference frame, for position of the subsequent determining boundary defect in image to be detected.Wherein, it sits
Mark can use the number of the pixel between two pixels to determine.For example, the coordinate of A1 ' is (0,0), A1 ' and A4 ' it
Between have 100 pixels, then the coordinate of A4 ' be (101,0).
S54 is based on straight line unfolded image, carries out the detection of boundary defect.The S24 of embodiment illustrated in fig. 4 is referred to,
Details are not described herein.
S55, when detecting that straight line unfolded image there are when boundary defect, utilizes borderline region image and straight line expanded view
Picture determines position of the boundary defect in image to be detected.
As shown in Figure 7 c, when detection device detects to mark defective bit there are when boundary defect in straight line unfolded image
The all pixels point set extracts coordinate of the marked pixel in the second reference frame, utilizes the coordinate extracted
It is matched with the coordinate in the first reference frame, that is, can determine that position of the boundary defect in image to be detected.Specifically
Include:
S551 extracts the coordinate that all pixels on the region of boundary defect o'clock are fastened in the second reference coordinate.
S552 matches the coordinate that the coordinate extracted and the first reference coordinate are fastened, to determine boundary defect to be detected
Position in image.
Due to getting coordinate of each pixel in the first reference frame in borderline region image in S331, together
When, the pixel in pixel and straight line unfolded image in borderline region image corresponds, therefore, can be using coordinate
The position of boundary defect is determined in image to be detected.
Compared with embodiment illustrated in fig. 4, the detection method of boundary defect provided in this embodiment utilizes the side of coordinate matching
Method can determine position of the boundary defect in image to be detected quickly, and detection efficiency with higher and detection are accurate
Rate.
As a kind of optional embodiment of the present embodiment, above-mentioned S52 and S53 can also be replaced using following steps:
(1) boundary line in two adjacent images region is extracted, on boundary line on the normal of each pixel and boundary line
The coordinate that each pixel is fastened in the first reference coordinate.
(2) it is based on boundary line, the borderline region straight line of image to be detected is unfolded, to form straight line unfolded image.
Wherein, step (2) specifically includes:
Step (2.1): using presetted pixel point on boundary line as starting point, boundary point is sequentially mapped to second along first straight line
Reference coordinate is fastened.
Step (2.2): along normal direction, in two adjacent images region, respectively extract with corresponding boundary point apart
The all pixels of first pre-determined distance and the second pre-determined distance point.
For example, normal extends L1 to region 2 as shown in figure 5, boundary point A1 corresponding normal 13, extend L2 to region 1,
Extract normal 13 line segment length be L1+L2 on all pixels and all pixels point and boundary point A1 between away from
From.
Wherein, borderline region is that each normal extends L1 to region 2 respectively, extends L2 to region 1, is formed by closed area
Domain.
The S532 of embodiment illustrated in fig. 9 is referred to, details are not described herein.
Step (2.3): using the distance between pixel and the corresponding boundary point on normal, the institute that will successively extract
There is pixel to be mapped to the second reference coordinate to fasten, to form straight line unfolded image.Refer to embodiment illustrated in fig. 9
S533, details are not described herein.
Additionally provide a kind of boundary defect detection device in the present embodiment, the device is for realizing above-described embodiment and excellent
Embodiment is selected, the descriptions that have already been made will not be repeated.As used below, predetermined function may be implemented in term " module "
Software and/or hardware combination.Although device described in following embodiment is preferably realized with software, hardware,
Or the realization of the combination of software and hardware is also that may and be contemplated.
The present embodiment provides a kind of boundary defect detection devices, as shown in Figure 10, comprising:
Module 61 is obtained, for obtaining image to be detected；Wherein, described image to be detected includes at least two image districts
Domain.
Extraction module 62, for extracting the borderline region in two adjacent images region, to obtain borderline region image.
Module 63 is unfolded in straight line, for borderline region graph line to be unfolded, to form straight line unfolded image.
Detection module 64 carries out the detection of boundary defect for being based on straight line unfolded image.
Boundary defect determining module 65, for when detecting that straight line unfolded image there are when boundary defect, utilizes frontier district
Area image and straight line unfolded image determine position of the boundary defect in image to be detected.
The detection device of boundary defect provided in an embodiment of the present invention passes through the borderline region to two adjacent images region
Straight line expansion is carried out, i.e., irregular borderline region image is converted into the borderline region image of rule, can be improved detection side
The robustness of method, and then improve the accuracy of detection method；In addition, can not only be realized by the defects detection to borderline region
Immediate area boundary or on zone boundary defect detection, additionally it is possible to realize the complete detection of the defect on trans-regional boundary,
Detection accuracy with higher.
In some optional embodiments of the present embodiment, as shown in figure 11, wherein extraction module 62 includes:
Acquiring unit 621, it is each on the boundary line and the boundary line for obtaining two neighboring described image region
The normal of pixel.
Offset units 622, for the normal direction of pixel each along the boundary line, by the boundary line respectively to
Two neighboring described image region deviates the first pre-determined distance and the second pre-determined distance, to obtain the first boundary line and second
Boundary line.
Extraction unit 623, for extracting the region between first boundary line and the second boundary line, with
To the borderline region image.
Still optionally further, as shown in figure 11, wherein straight line is unfolded module 63 and includes:
First map unit 631 is used for using presetted pixel point is starting point on the boundary line, by boundary point along first straight line
The second reference coordinate is sequentially mapped to fasten；Wherein, the boundary point is the pixel of the normal and the boundary line point of intersection
Point, the normal and the boundary point correspond.
Computing unit 632, for successively calculating between each pixel on the normal and the corresponding boundary point
Distance.
Second map unit 633, for utilizing the distance, successively by all pixels point on the borderline region image
It is mapped to second reference coordinate to fasten, to form the straight line unfolded image；Wherein, the picture in the borderline region image
Pixel in vegetarian refreshments and the straight line unfolded image corresponds.
Boundary defect detection device in the present embodiment is presented in the form of functional unit, and unit here refers to
ASIC circuit, execute one or more softwares or fixed routine processor and memory and/or other above-mentioned function can be provided
The device of energy.
The further function description of above-mentioned modules is identical as above-mentioned corresponding embodiment, and details are not described herein.
The embodiment of the present invention also provides a kind of detection device, has the detection of boundary defect shown in above-mentioned Figure 10 and Figure 11
Device.
Figure 12 is please referred to, Figure 12 is a kind of structural schematic diagram for detection device that alternative embodiment of the present invention provides, and is such as schemed
Shown in 12, which may include: at least one processor 71, such as CPU (Central Processing Unit, in
Central processor), at least one communication interface 73, memory 74, at least one communication bus 72.Wherein, communication bus 72 is used for
Realize the connection communication between these components.Wherein, communication interface 73 may include display screen (Display), keyboard
(Keyboard), optional communication interface 73 can also include standard wireline interface and wireless interface.Memory 74 can be high speed
RAM memory (Random Access Memory, effumability random access memory), is also possible to non-labile storage
Device (non-volatile memory), for example, at least a magnetic disk storage.Memory 74 optionally can also be at least one
It is located remotely from the storage device of aforementioned processor 71.Wherein processor 71 can combine device described in Figure 10 and Figure 11,
Application program is stored in memory 74, and processor 71 calls the program code stored in memory 74, with above-mentioned for executing
Either method step.
Wherein, communication bus 72 can be Peripheral Component Interconnect standard (peripheral component
Interconnect, abbreviation PCI) bus or expanding the industrial standard structure (extended industry standard
Architecture, abbreviation EISA) bus etc..Communication bus 72 can be divided into address bus, data/address bus, control bus etc..
Only to be indicated with a thick line in Figure 12, it is not intended that an only bus or a type of bus convenient for indicating.
Wherein, memory 74 may include volatile memory (English: volatile memory), such as arbitrary access
Memory (English: random-access memory, abbreviation: RAM)；Memory also may include nonvolatile memory (English
Text: non-volatile memory), for example, flash memory (English: flash memory), hard disk (English: hard disk
Drive, abbreviation: HDD) or solid state hard disk (English: solid-state drive, abbreviation: SSD)；Memory 74 can also include
The combination of the memory of mentioned kind.
Wherein, processor 71 can be central processing unit (English: central processing unit, abbreviation: CPU),
The combination of network processing unit (English: network processor, abbreviation: NP) or CPU and NP.
Wherein, processor 71 can further include hardware chip.Above-mentioned hardware chip can be specific integrated circuit
(English: application-specific integrated circuit, abbreviation: ASIC), programmable logic device (English:
Programmable logic device, abbreviation: PLD) or combinations thereof.Above-mentioned PLD can be Complex Programmable Logic Devices
(English: complex programmable logic device, abbreviation: CPLD), field programmable gate array (English:
Field-programmable gate array, abbreviation: FPGA), Universal Array Logic (English: generic array
Logic, abbreviation: GAL) or any combination thereof.
Optionally, memory 74 is also used to store program instruction.Processor 71 can be instructed with caller, realize such as this Shen
It please boundary defect detection method shown in Fig. 1, Fig. 4, Fig. 8 and Fig. 9 embodiment.
The embodiment of the invention also provides a kind of non-transient computer storage medium, the computer storage medium is stored with
The boundary defect detection side in above-mentioned any means embodiment can be performed in computer executable instructions, the computer executable instructions
Method.Wherein, the storage medium can be magnetic disk, CD, read-only memory (Read-Only Memory, ROM), deposit at random
Store up memory body (Random Access Memory, RAM), flash memory (Flash Memory), hard disk (Hard Disk
Drive, abbreviation: HDD) or solid state hard disk (Solid-State Drive, SSD) etc.；The storage medium can also include above-mentioned
The combination of the memory of type.
Although being described in conjunction with the accompanying the embodiment of the present invention, those skilled in the art can not depart from the present invention
Spirit and scope in the case where various modifications and variations can be made, such modifications and variations are each fallen within by appended claims institute
Within the scope of restriction.
Claims (10)
1. a kind of detection method of boundary defect characterized by comprising
Obtain image to be detected；Wherein, described image to be detected includes at least two image-regions；
The borderline region in two neighboring described image region is extracted, to obtain borderline region image；
The borderline region graph line is unfolded, to form straight line unfolded image；
Based on the straight line unfolded image, the detection of the boundary defect is carried out；
When detecting that the straight line unfolded image there are when the boundary defect, utilizes the borderline region image and described straight
Line unfolded image determines position of the boundary defect in described image to be detected.
2. the method according to claim 1, wherein the frontier district for extracting two neighboring described image region
Domain, to obtain borderline region image, comprising:
Obtain the normal and coordinate of each pixel on the boundary line and the boundary line in two neighboring described image region；Its
In, the coordinate is uniquely determined by the first reference frame；
The normal direction of each pixel along the boundary line, by the boundary line respectively to two neighboring described image region
The first pre-determined distance and the second pre-determined distance are deviated, to obtain the first boundary line and the second boundary line；
The region between first boundary line and the second boundary line is extracted, to obtain the borderline region image.
3. according to the method described in claim 2, it is characterized in that, described the borderline region graph line is unfolded, with shape
Be in line unfolded image, comprising:
Using presetted pixel point on the boundary line as starting point, boundary point is sequentially mapped to the second reference frame along first straight line
On；Wherein, the boundary point is the pixel of the normal and the boundary line point of intersection, the normal and the boundary point one
One is corresponding；
Successively calculate the distance between each pixel on the normal and the corresponding boundary point；
Using the distance, all pixels point on the borderline region image is successively mapped to second reference frame
On, to form the straight line unfolded image；Wherein, in the pixel in the borderline region image and the straight line unfolded image
Pixel correspond.
4. according to the method described in claim 3, it is characterized in that, described utilize the distance, successively by the borderline region
All pixels point on image is mapped to second reference coordinate and fastens, comprising:
Successively extract the pixel on the normal；
Inquire mapped boundaries point corresponding with the normal；Wherein, the mapped boundaries point is corresponding with the normal described
The mapping point that boundary point is fastened in second reference coordinate；
The pixel extracted second reference coordinate is mapped to along second straight line to fasten；Wherein, the institute extracted
The mapping point and the mapped boundaries point for stating pixel are separated by the distance, and the second straight line is vertical with first straight line, described
Second straight line and the mapped boundaries point correspond.
5. according to the method described in claim 2, it is characterized in that, it is described be based on the straight line unfolded image, carry out the side
The detection of boundary's defect, comprising:
The straight line unfolded image is filtered, filtering image is formed；
Compare the straight line unfolded image and the filtering image, obtains differential image；
Filter out the region for meeting preset condition in the differential image；Wherein, the region filtered out is the institute detected
State the region of boundary defect.
6. according to the method described in claim 2, it is characterized in that, described utilize the borderline region image and the straight line
Unfolded image determines position of the boundary defect in described image to be detected, comprising:
Extract the coordinate that all pixels on the region of the boundary defect o'clock are fastened in the second reference coordinate；
The coordinate that the coordinate extracted and the first reference coordinate are fastened is matched, with the determination boundary defect described to be checked
Position in altimetric image.
7. according to the method described in claim 2, it is characterized in that, the boundary for obtaining two neighboring described image region
Line, comprising:
Obtain zero defect template image；Wherein, the zero defect template image and described image to be detected correspond；
Extract the boundary line in the two adjacent images region of the zero defect template image；
It is aligned the zero defect image and described image to be detected；
The boundary line extracted is mapped in described image to be detected；Wherein, the boundary line extracted is described
The boundary line in the two neighboring described image region of image to be detected.
8. a kind of detection device of boundary defect characterized by comprising
Module is obtained, for obtaining image to be detected；Wherein, described image to be detected includes at least two image-regions；
Extraction module, for extracting the borderline region in two neighboring described image region, to obtain borderline region image；
Module is unfolded in straight line, for the borderline region graph line to be unfolded, to form straight line unfolded image；
Detection module carries out the detection of the boundary defect for being based on the straight line unfolded image；
Boundary defect determining module, for when detecting the straight line unfolded image there are when the boundary defect, using described
Borderline region image and the straight line unfolded image, determine position of the boundary defect in described image to be detected.
9. a kind of detection device characterized by comprising at least one processor；And it is logical at least one described processor
Believe the memory of connection；Wherein, the memory is stored with the instruction that can be executed by one processor, and described instruction is by institute
The execution of at least one processor is stated, so that at least one described processor executes any side in the claims 1-7
The detection method of boundary's defect.
10. a kind of computer readable storage medium, is stored thereon with computer instruction, which is characterized in that the instruction is by processor
The step of detection method of any boundary defect in the claims 1-7 is realized when execution.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811062811.9A CN109325930B (en) | 2018-09-12 | 2018-09-12 | Boundary defect detection method, device and detection equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811062811.9A CN109325930B (en) | 2018-09-12 | 2018-09-12 | Boundary defect detection method, device and detection equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109325930A true CN109325930A (en) | 2019-02-12 |
| CN109325930B CN109325930B (en) | 2021-09-28 |
Family
ID=65265869
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811062811.9A Active CN109325930B (en) | 2018-09-12 | 2018-09-12 | Boundary defect detection method, device and detection equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109325930B (en) |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110717902A (en) * | 2019-09-29 | 2020-01-21 | 中山市瑞福达触控显示技术有限公司 | Processing method for displaying image edge |
| CN111353974A (en) * | 2020-02-20 | 2020-06-30 | 凌云光技术集团有限责任公司 | Method and device for detecting image boundary defects |
| CN111598074A (en) * | 2020-05-21 | 2020-08-28 | 杭州睿琪软件有限公司 | Edge detection method and apparatus, electronic device, and storage medium |
| CN111798449A (en) * | 2020-09-09 | 2020-10-20 | 江苏恒力化纤股份有限公司 | Spinneret plate residual impurity detection method based on image technology |
| CN111951290A (en) * | 2019-05-16 | 2020-11-17 | 杭州睿琪软件有限公司 | Edge detection method and device for object in image |
| CN112686919A (en) * | 2020-12-29 | 2021-04-20 | 上海闻泰信息技术有限公司 | Object boundary line determining method and device, electronic equipment and storage medium |
| CN113034527A (en) * | 2021-03-30 | 2021-06-25 | 长江存储科技有限责任公司 | Boundary detection method and related product |
| CN113538478A (en) * | 2020-04-15 | 2021-10-22 | 深圳市光鉴科技有限公司 | Image-based box boundary extraction method, system, equipment and storage medium |
| CN113658133A (en) * | 2021-08-16 | 2021-11-16 | 江苏鑫丰源机电有限公司 | Gear surface defect detection method and system based on image processing |
| CN114820594A (en) * | 2022-06-21 | 2022-07-29 | 中科慧远视觉技术(北京)有限公司 | Method for detecting edge sealing defects of plate based on image, related equipment and storage medium |
| CN115830043A (en) * | 2023-01-09 | 2023-03-21 | 北京阿丘机器人科技有限公司 | Boundary detection method, device and equipment of wireless magnet charging and storage medium |
| CN115880248A (en) * | 2022-12-13 | 2023-03-31 | 哈尔滨耐是智能科技有限公司 | Surface scratch defect identification method and visual detection equipment |
| CN117437233A (en) * | 2023-12-21 | 2024-01-23 | 山东润通齿轮集团有限公司 | Gear defect detection method and system based on image processing |
Citations (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1985147A (en) * | 2004-03-27 | 2007-06-20 | 特克斯玛格销售有限公司 | Apparatus for detecting joints in rubber sheets |
| CN102062739A (en) * | 2006-05-23 | 2011-05-18 | 麒麟工程技术***公司 | Surface examining device |
| CN102288620A (en) * | 2011-08-10 | 2011-12-21 | 天津大学 | Method and device for unfolding surface of steel ball on basis of multiple image sensors |
| US20130221556A1 (en) * | 2012-02-27 | 2013-08-29 | Canon Kabushiki Kaisha | Detector, imprint apparatus and method of manufacturing article |
| CN103499590A (en) * | 2013-10-17 | 2014-01-08 | 福州大学 | Method and system for detecting and screening end defects in circular parts |
| CN103745475A (en) * | 2014-01-22 | 2014-04-23 | 哈尔滨工业大学 | Detection and positioning method used for spherical pin element |
| CN104024792A (en) * | 2011-11-07 | 2014-09-03 | 株式会社神户制钢所 | Tire shape inspection method and tire shape inspection device |
| US8864283B1 (en) * | 2013-05-09 | 2014-10-21 | Xerox Corporation | System and method for visually detecting defective inkjets in an inkjet imaging apparatus |
| CN104568983A (en) * | 2015-01-06 | 2015-04-29 | 浙江工业大学 | Active-omni-directional-vision-based pipeline inside functional defect detection device and detection method |
| CN105066892A (en) * | 2015-08-05 | 2015-11-18 | 哈尔滨工业大学 | BGA element detecting and positioning method based on linear clustering analysis |
| CN105913415A (en) * | 2016-04-06 | 2016-08-31 | 吴晓军 | Image sub-pixel edge extraction method having extensive adaptability |
| CN106056546A (en) * | 2016-05-25 | 2016-10-26 | 广东工业大学 | Image restoration method and device based on sample block matching |
| CN106600600A (en) * | 2016-12-26 | 2017-04-26 | 华南理工大学 | Wafer defect detection method based on characteristic matching |
| CN107328781A (en) * | 2017-05-23 | 2017-11-07 | 镇江苏仪德科技有限公司 | A kind of columnar product detection method of surface flaw and device based on machine vision |
| CN108459027A (en) * | 2018-03-21 | 2018-08-28 | 华北电力大学 | A kind of blade of wind-driven generator detection method of surface flaw based on image procossing |
-
2018
- 2018-09-12 CN CN201811062811.9A patent/CN109325930B/en active Active
Patent Citations (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1985147A (en) * | 2004-03-27 | 2007-06-20 | 特克斯玛格销售有限公司 | Apparatus for detecting joints in rubber sheets |
| CN102062739A (en) * | 2006-05-23 | 2011-05-18 | 麒麟工程技术***公司 | Surface examining device |
| CN102288620A (en) * | 2011-08-10 | 2011-12-21 | 天津大学 | Method and device for unfolding surface of steel ball on basis of multiple image sensors |
| CN104024792A (en) * | 2011-11-07 | 2014-09-03 | 株式会社神户制钢所 | Tire shape inspection method and tire shape inspection device |
| US20130221556A1 (en) * | 2012-02-27 | 2013-08-29 | Canon Kabushiki Kaisha | Detector, imprint apparatus and method of manufacturing article |
| US8864283B1 (en) * | 2013-05-09 | 2014-10-21 | Xerox Corporation | System and method for visually detecting defective inkjets in an inkjet imaging apparatus |
| CN103499590A (en) * | 2013-10-17 | 2014-01-08 | 福州大学 | Method and system for detecting and screening end defects in circular parts |
| CN103745475A (en) * | 2014-01-22 | 2014-04-23 | 哈尔滨工业大学 | Detection and positioning method used for spherical pin element |
| CN104568983A (en) * | 2015-01-06 | 2015-04-29 | 浙江工业大学 | Active-omni-directional-vision-based pipeline inside functional defect detection device and detection method |
| CN105066892A (en) * | 2015-08-05 | 2015-11-18 | 哈尔滨工业大学 | BGA element detecting and positioning method based on linear clustering analysis |
| CN105913415A (en) * | 2016-04-06 | 2016-08-31 | 吴晓军 | Image sub-pixel edge extraction method having extensive adaptability |
| CN106056546A (en) * | 2016-05-25 | 2016-10-26 | 广东工业大学 | Image restoration method and device based on sample block matching |
| CN106600600A (en) * | 2016-12-26 | 2017-04-26 | 华南理工大学 | Wafer defect detection method based on characteristic matching |
| CN107328781A (en) * | 2017-05-23 | 2017-11-07 | 镇江苏仪德科技有限公司 | A kind of columnar product detection method of surface flaw and device based on machine vision |
| CN108459027A (en) * | 2018-03-21 | 2018-08-28 | 华北电力大学 | A kind of blade of wind-driven generator detection method of surface flaw based on image procossing |
Non-Patent Citations (6)
| Title |
|---|
| RAJITHA B 等: "A new local homogeneity analysis method based on pixel intensities for image defect detection", 《2015 IEEE 2ND INTERNATIONAL CONFERENCE ON RECENT TRENDS IN INFORMATION SYSTEMS》 * |
| 于军辉 等: "Zr-4合金涡流缺陷边界定位的新方法", 《科技视界》 * |
| 于海燕 等: "钢板表面缺陷图像检测与分类技术研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
| 刘文涛: "基于机器视觉的IC缺陷检测的研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
| 姜美华: "基于反射图案的钢球表面缺陷检测新方法的算法研", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
| 熊伟俊 等: "基于快速l1算法和LBP算法的木材缺陷识别", 《数据采集与处理》 * |
Cited By (23)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111951290A (en) * | 2019-05-16 | 2020-11-17 | 杭州睿琪软件有限公司 | Edge detection method and device for object in image |
| CN111951290B (en) * | 2019-05-16 | 2023-11-03 | 杭州睿琪软件有限公司 | Edge detection method and device for object in image |
| CN110717902B (en) * | 2019-09-29 | 2023-06-09 | 中山市瑞福达触控显示技术有限公司 | Processing method for display image edge |
| CN110717902A (en) * | 2019-09-29 | 2020-01-21 | 中山市瑞福达触控显示技术有限公司 | Processing method for displaying image edge |
| CN111353974A (en) * | 2020-02-20 | 2020-06-30 | 凌云光技术集团有限责任公司 | Method and device for detecting image boundary defects |
| CN111353974B (en) * | 2020-02-20 | 2023-08-18 | 苏州凌云光工业智能技术有限公司 | Method and device for detecting image boundary defects |
| CN113538478B (en) * | 2020-04-15 | 2024-06-04 | 深圳市光鉴科技有限公司 | Box boundary extraction method, system, equipment and storage medium based on image |
| CN113538478A (en) * | 2020-04-15 | 2021-10-22 | 深圳市光鉴科技有限公司 | Image-based box boundary extraction method, system, equipment and storage medium |
| CN111598074A (en) * | 2020-05-21 | 2020-08-28 | 杭州睿琪软件有限公司 | Edge detection method and apparatus, electronic device, and storage medium |
| CN111598074B (en) * | 2020-05-21 | 2023-07-07 | 杭州睿琪软件有限公司 | Edge detection method and device, electronic equipment and storage medium |
| CN111798449B (en) * | 2020-09-09 | 2021-02-05 | 江苏恒力化纤股份有限公司 | Spinneret plate residual impurity detection method based on image technology |
| CN111798449A (en) * | 2020-09-09 | 2020-10-20 | 江苏恒力化纤股份有限公司 | Spinneret plate residual impurity detection method based on image technology |
| CN112686919B (en) * | 2020-12-29 | 2022-10-21 | 上海闻泰信息技术有限公司 | Object boundary line determining method and device, electronic equipment and storage medium |
| CN112686919A (en) * | 2020-12-29 | 2021-04-20 | 上海闻泰信息技术有限公司 | Object boundary line determining method and device, electronic equipment and storage medium |
| CN113034527B (en) * | 2021-03-30 | 2022-05-03 | 长江存储科技有限责任公司 | Boundary detection method and related product |
| CN113034527A (en) * | 2021-03-30 | 2021-06-25 | 长江存储科技有限责任公司 | Boundary detection method and related product |
| CN113658133A (en) * | 2021-08-16 | 2021-11-16 | 江苏鑫丰源机电有限公司 | Gear surface defect detection method and system based on image processing |
| CN114820594A (en) * | 2022-06-21 | 2022-07-29 | 中科慧远视觉技术(北京)有限公司 | Method for detecting edge sealing defects of plate based on image, related equipment and storage medium |
| CN115880248A (en) * | 2022-12-13 | 2023-03-31 | 哈尔滨耐是智能科技有限公司 | Surface scratch defect identification method and visual detection equipment |
| CN115880248B (en) * | 2022-12-13 | 2024-02-09 | 哈尔滨耐是智能科技有限公司 | Surface scratch defect identification method and visual detection equipment |
| CN115830043A (en) * | 2023-01-09 | 2023-03-21 | 北京阿丘机器人科技有限公司 | Boundary detection method, device and equipment of wireless magnet charging and storage medium |
| CN117437233A (en) * | 2023-12-21 | 2024-01-23 | 山东润通齿轮集团有限公司 | Gear defect detection method and system based on image processing |
| CN117437233B (en) * | 2023-12-21 | 2024-03-26 | 山东润通齿轮集团有限公司 | Gear defect detection method and system based on image processing |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109325930B (en) | 2021-09-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109325930A (en) | Detection method, device and the detection device of boundary defect | |
| CN110210409B (en) | Method and system for detecting form frame lines in form document | |
| CN108921865B (en) | Anti-interference sub-pixel straight line fitting method | |
| CN106934803B (en) | method and device for detecting surface defects of electronic device | |
| CN103729632B (en) | A kind of localization method of Circular Mark point based on connected region filtering | |
| CN107203990A (en) | A kind of labeling damage testing method based on template matches and image quality measure | |
| CN112232391B (en) | Dam crack detection method based on U-net network and SC-SAM attention mechanism | |
| WO2022007431A1 (en) | Positioning method for micro qr two-dimensional code | |
| CN107092871B (en) | Remote sensing image building detection method based on multiple dimensioned multiple features fusion | |
| CN105913415A (en) | Image sub-pixel edge extraction method having extensive adaptability | |
| CN111210477A (en) | Method and system for positioning moving target | |
| CN101751572A (en) | Pattern detection method, device, equipment and system | |
| CN104240204A (en) | Solar silicon wafer and battery piece counting method based on image processing | |
| Zhou et al. | Exploring faster RCNN for fabric defect detection | |
| CN111582093A (en) | Automatic small target detection method in high-resolution image based on computer vision and deep learning | |
| JP2011043969A (en) | Method for extracting image feature point | |
| EP3163604B1 (en) | Position detection apparatus, position detection method, information processing program, and storage medium | |
| CN108960236B (en) | Pointer instrument identification method based on two-dimensional code matching and image identification | |
| CN108445009A (en) | A kind of solar panel crack detecting method | |
| CN103914829B (en) | Method for detecting edge of noisy image | |
| CN105354816B (en) | A kind of electronic units fix method and device | |
| CN113705564B (en) | Pointer type instrument identification reading method | |
| JP2017500662A (en) | Method and system for correcting projection distortion | |
| CN107808165B (en) | Infrared image matching method based on SUSAN corner detection | |
| CN114119437A (en) | GMS-based image stitching method for improving moving object distortion |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |