CN1746666A - Object based color image defect detection - Google Patents
Object based color image defect detection Download PDFInfo
- Publication number
- CN1746666A CN1746666A CN 200510092185 CN200510092185A CN1746666A CN 1746666 A CN1746666 A CN 1746666A CN 200510092185 CN200510092185 CN 200510092185 CN 200510092185 A CN200510092185 A CN 200510092185A CN 1746666 A CN1746666 A CN 1746666A
- Authority
- CN
- China
- Prior art keywords
- mentioned
- color
- region
- special processing
- physical presence
- 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
Images
Landscapes
- Image Processing (AREA)
- Image Analysis (AREA)
- Length Measuring Devices By Optical Means (AREA)
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
Abstract
The invention provides a defect detection technology based on the color image of an object, and improves the reliability of defect detection processing of a checking object. The invention comprises the following steps: a color image is obtained through shooting a checking object comprising a through hole; a specific area and an adjacent area around the specific area of the color image equivalent to the through hole are obtained; according to the color of the obtained adjacent area, the size of a mask area comprising the specific area is changed; moreover, specified defect detection processing of the area outside the mask area is carried out.
Description
Technical field
The present invention relates to use coloured image to detect the technology of the defective of checking object.
Background technology
At the printed base plate that is used to constitute electronic circuit, be provided with the conducting that is used for interlayer and parts insertion through hole and be used for the cut-out of printed base plate and the substrate aperture of location.Such printed base plate, the thickness variation of the resist of bore periphery is big sometimes, and printed base plate is made a video recording and color on the image that obtains changes.Therefore, when checking printed base plate by image testing device, printed base plate is made a video recording and in the image that obtains, the suitable zone of certain width part with the periphery in hole and hole, generally from the object of checking, remove (zone of removing is commonly referred to as " masks area ") from check object.
Patent documentation 4 is a TOHKEMY 2002-259667 communique.
But, without exception as masks area, can not detect the zone suitable sometimes and not be through hole but be present in defective around the substrate aperture with the peripheral certain width part in hole and hole.On the other hand, if only making the zone suitable with the hole is masks area, then because the change color that change allowed of the resist thickness around the through hole is defective printed base plate even might there not be the printed base plate of defective also to be mistaken as.That is to say that the coloured image that obtains taking printed base plate is analyzed and carried out defective when detecting, and has the such problem of erroneous judgement that defective is had or not and takes place.
Above-mentioned problem is not limited only to the inspection of the image testing device of printed base plate, in general, is common problem in the inspection of having used the coloured image of checking object.
Summary of the invention
The present invention produces in order to solve above-mentioned prior art problems, and its purpose is to improve the reliability that the defective of checking object detects processing.
At least a portion in order to achieve the above object, method of the present invention, be to use the inspection object that comprises through hole is made a video recording and the coloured image that obtains detects the method for the defective of above-mentioned inspection object, it is characterized in that, comprising: the operation of (a) obtaining the specific region suitable in the above-mentioned coloured image with above-mentioned through hole; (b) obtain above-mentioned specific region around the operation of adjacent area; (c) setting comprises the operation of the masks area of above-mentioned specific region; (d) at the zone beyond the aforementioned mask zone in the above-mentioned coloured image, the defective of stipulating detects the operation of processing, and the size in aforementioned mask zone changes corresponding to the color of above-mentioned adjacent area.
Constitute according to this, because the size of masks area changes corresponding to the color of adjacent area, so can improve the reliability that the defective of checking object detects processing.
The aforementioned mask zone has above-mentioned specific region and comprises the special processing region on every side of above-mentioned specific region, said method also comprises: (e) at above-mentioned special processing region, carry out detecting the operation of handling different special defect processing with the defective of afore mentioned rules and also be fine.
Constitute according to this, can detect the defective that occurs in the masks area.
Said method, it is characterized in that, also comprise: (f) in the color space of regulation, the physical presence color gamut of the scope of the color that the above-mentioned special processing region of setting expression obtains and the operation of the non-physical presence color gamut beyond the above-mentioned physical presence color gamut, above-mentioned operation (e) comprising: (e1) by judging which of above-mentioned physical presence color gamut and above-mentioned non-physical presence color gamut be each color of pixel that constitutes above-mentioned special processing region belong to, and detects the operation of the defective in the above-mentioned special processing region.
Constitute according to this, can detect the defective of the special processing region that satisfies defined terms.
Above-mentioned operation (f) also can comprise: preparation is as the operation of the master image of the coloured image of the standard of above-mentioned inspection object; The scope of the part of the color space of the afore mentioned rules that comprises the color of pixel that constitutes above-mentioned special processing region in the above-mentioned master image is set at the operation of above-mentioned physical presence color gamut.
Constitute according to this, because can set the physical presence color gamut, so the setting of physical presence color has just become easily by the color of pixel in the special processing region of master image.
Also can be that above-mentioned operation (f) comprises the operation of making question blank, this question blank with the color arbitrarily in the color space of afore mentioned rules as input, the above-mentioned color arbitrarily of output expression belongs to which the value in above-mentioned physical presence color gamut and the above-mentioned non-physical presence color gamut, above-mentioned operation (e1) comprising: with reference to above-mentioned question blank, judge that above-mentioned each color of pixel belongs to which the operation in above-mentioned physical presence color gamut and the above-mentioned non-physical presence color gamut.
Constitute according to this, because whether the color of pixel in the special processing region belongs to the judgement of physical presence color gamut can judge, so the inspection of special processing region can be carried out more at high speed by reference look-up tables.
It also can be that above-mentioned operation (b) comprising: the operation of borderline region of (b1) obtaining the periphery of above-mentioned specific region; When (b2) color of above-mentioned borderline region satisfies defined terms, comprise above-mentioned borderline region and obtain the operation of the above-mentioned adjacent area bigger than above-mentioned borderline region; When (b3) color of above-mentioned borderline region does not satisfy the condition of afore mentioned rules, the size in aforementioned mask zone is set at the operation of prescribed level.
Constitute according to this, because the color of borderline region is not when satisfying defined terms, corresponding to the color of adjacent area, the size of masks area changes, so can reduce the Flame Image Process amount of adjacent area.
The condition that also can be afore mentioned rules is the such condition of pixel that has first color gamut of regulation in above-mentioned borderline region, above-mentioned operation (c) comprising: the color of pixel in the above-mentioned adjacent area the regulation second color gamut outside the time, compare the operation that the aforementioned mask zone is increased with the situation in above-mentioned second color gamut.
Constitute according to this, can more suitably set the size of masks area.
Above-mentioned inspection object is that printed base plate is also passable.
Constitute according to this, the defective that can improve printed base plate detects the reliability of processing.
In addition, the present invention can realize with various forms, for example can be with defective method for detecting and the device of checking object, use this image checking method that detects the result and device, be used to realize the function of these the whole bag of tricks or device computer program, write down this computer program recording medium, comprise this computer program and in carrier wave, realized by the forms such as data-signal of specific implementation.
Description of drawings
Fig. 1 is the key diagram of expression as the structure of the printed base plate testing fixture 100 of one embodiment of the invention.
Fig. 2 is the key diagram of the surface state of expression printed base plate PCB.
Fig. 3 is the process flow diagram that the defective of printed base plate PCB among expression first embodiment detects the order of processing.
Fig. 4 A～Fig. 4 C is the key diagram of the situation about obtaining of expression adjacent area.
Fig. 5 A～Fig. 5 C is the key diagram of situation of the setting of the special processing region of expression.
Fig. 6 A, Fig. 6 B are illustrated in the key diagram that defective on the printed base plate that does not have defective in the special processing region detects the situation of processing.
Fig. 7 A, Fig. 7 B are illustrated in the key diagram that the defective on the defective printed base plate in the special processing region detects the situation of processing.
Fig. 8 A, Fig. 8 B are that expression is carried out the key diagram that defective detects the result of processing for the printed base plate PCB that does not have defective.
Fig. 9 is the process flow diagram that the defective of printed base plate PCB among expression second embodiment detects the order of processing.
Figure 10 A, Figure 10 B are the key diagrams of setting situation that the adjacent area of borderline region has been used in expression.
Embodiment
Then, be used to implement optimised form of the present invention based on embodiment according to following order explanation.
A. first embodiment
B. second embodiment
C. variation
A. first embodiment:
Fig. 1 is the key diagram of expression as the structure of the printed base plate testing fixture 100 of one embodiment of the present of invention.This printed base plate testing fixture 100 has: printed base plate PCB is carried out illuminating light source 20; The image pickup part 30 that the image of printed base plate PCB is made a video recording; Carry out the computing machine 40 of the control of device integral body.In computing machine 40, be connected with the external memory 50 of store various kinds of data and computer program.
Fig. 2 is the key diagram of expression as the situation of the printed base plate PCB that checks object.This printed base plate PCB is provided with the substrate aperture HL of the cut-out usefulness of 5 through hole TH1～TH5, printed base plate PCB.The surface of printed base plate PCB comprises: the serigraphy region R SG of white text in serigraphy on substrate; Be implemented gold-plated gold-plated region R GP; The substrate region R SB that exposes substrate; The substrate resist region R BR of coating resist on substrate; The figure resist region R PR of coating resist on the figure of copper wiring.In addition, on the figure of copper wiring, in the zone of coating resist, become the thin thin resist region R TR of resist around the through hole TH5.
In general, printed base plate PCB, because the working procedure of coating of resist, and the resist thickness around the through hole becomes unstable, sometimes at each through hole, the thickness of the resist of periphery is all inequality.Therefore, even there is not the printed base plate PCB of defective, as shown in Figure 2, produce the thin zone of resist in bore periphery.And the allowed band of the thickness of the resist of bore periphery is set widelyer than other zones of coating resist, allows the thin zone of the such resist of thin resist region R TR to exist.In addition, even there is not the printed base plate of defective, thin resist region R TR also can be owing to the change of resist thickness color difference as described later.That is, the thin resist region R TR adjacent with the through hole allowed band that is color is than other regional wide zones.
Fig. 3 is the process flow diagram of the order of the expression defective that detects printed base plate PCB among first embodiment.In step S100, image acquiring unit 210 (Fig. 1) is obtained the coloured image of printed base plate PCB from image pickup part 30.
In addition, in step S100, for the coloured image that is obtained, image acquiring unit 210 is carried out smoothing as required and is handled (setting off (Pot か) by contrast handles).In smoothing is handled, can use various smoothing light filters such as medium light filter, Gauss's light filter, moving average.Because by carry out smoothing handle can remove image data memory special pixel, so can obtain the few view data of impurity (noise composition).In addition, for the image that obtains in advance, when the later processing of execution in step S200, read view data from external memory 50 (Fig. 1) at step S100.
Fig. 4 A is that expression is made a video recording to printed base plate PCB (Fig. 2) and the key diagram of the situation of the coloured image IM that obtains.This coloured image IM comprises black region BK, white portion WH, golden regional GL, dark brown area B R, dirty-green zone GD, bright green area GB.In addition, in this manual, these image-regions BK, WH, GL, BR, GD, GB are generically and collectively referred to as " color region ".
On coloured image IM, because through hole TH1～TH5, substrate aperture HL have opened the hole on substrate, so represent with black region BK.Serigraphy region R SG, gold-plated region R GP and substrate base region R SB correspond respectively to the color of Facing material, represent with white portion WH, golden regional GL, dark brown area B R.Substrate resist region R BR, because on dark brown substrate, applied green resist, so GD represents with the dirty-green zone, figure resist region R PR, because below the resist is the copper wiring figure of coppery, represents so use than the high bright green area GB of substrate resist region R BR brightness.And the thin thin resist region R TR of the resist on the figure, because show the color of copper wiring, so represent with golden regional GL.
In addition, in the coloured image that obtains reality printed base plate PCB is made a video recording, corresponding to the golden regional GL of thin resist region R TR and indeterminate corresponding to the border of the bright green area GB of the figure resist region R PR around it, in Fig. 4 A, in order to illustrate conveniently, these golden regional GL are depicted as separated portions with bright green area GB as different color regions.
Among the step S200 of Fig. 3, bore region acquisition unit 220 is obtained the bore region that expression is arranged on the hole on the printed base plate PCB by extracting the black region BK among the coloured image IM.Black region BK, the brightness value that can be used as each pixel that constitutes coloured image IM extracts for the luminance threshold of regulation or the zone below it.The luminance threshold of the regulation that the extraction of this black region BK is used for example can be set by the histogram analysis to brightness value.
In addition, in the present embodiment, the brightness value of each pixel carries out though the extraction of black region BK is based on, and the extraction of black region BK also can be undertaken by other method.The extraction of black region BK also can be carried out in the following way, promptly in the zone of a plurality of representative colors that are predetermined that comprise black with coloured image IM Region Segmentation, the representative color that extracts in the divided zone is the zone of black.The Region Segmentation of coloured image IM for example can be carried out in the following way: try to achieve the range index value of distance of color space of the regulation of each color of pixel of expression coloured image IM and a plurality of representative colors, each pixel is categorized as the range index value is minimum representative color zone.As this range index value, for example can utilize the RGB color space is considered as the Euclidean distance of 3 dimensions during Euclidean spaces, perhaps can utilize L
*a
*b
*Aberration Δ E in the space.In addition, the region segmentation method of coloured image IM so long as the region segmentation method that each pixel is categorized as a plurality of representative colors gets final product, for example can be undertaken by disclosed method in above-mentioned patent documentation 3 or the patent documentation 4.
In addition, in the present embodiment, though bore region obtain by the black region BK that extracts among the coloured image IM, also can by other method obtain be arranged on printed base plate PCB on the corresponding image-region in hole.For example, obtaining bore region by position that is included in the hole in the design data (cad data) that is used to form through hole and substrate aperture and size also is fine.In addition, to making a video recording, the high zone of brightness of the image that photographs is set at bore region also is fine from the illuminated printed base plate PCB of the opposing face of image pickup part 30.
Fig. 4 B is the key diagram that is illustrated in the configuration of the bore region that obtains among the step S200.Shown in Fig. 4 B,, obtain image-region SR1～SR5 corresponding and the image-region SR6 corresponding with substrate aperture HL with through hole TH1～TH5 by extracting black region BK from coloured image IM.Image-region SR1～the SR6 that obtains like this is respectively a bore region.
Among the step S300 of Fig. 3, adjacent area acquisition unit 230 is obtained the adjacent area adjacent with bore region SR1～SR6.Adjacent area acquisition unit 230 is handled the expansion that the bore region SR1～SR6 enlarging amplitude (for example 5 pixels) according to the rules that obtains at step S200 amplifies.And become adjacent area by enlarge handling the zone that has enlarged.As this enlargement range, setting value that except the setting value that is predetermined, can also use designated value that the user imports, calculates based on cad data etc.In addition, also can set enlarging amplitude respectively by each bore region.
In addition, handle, comprise bore region, the zone processing bigger than bore region of generation got final product, can be suitable for arbitrarily and handle so long as handle the zone that produces by expansion as the expansion of bore region.As such processing, for example each pixel 8 near any one when belonging to bore region, can use this pixel is set at the expansion process that belongs to bore region, or use the expansion process of this expansion process being carried out the n section of n time (n is the integer more than 1).In addition, also can use the profile that obtains bore region and the processing that enlarges this profile.
Fig. 4 C is the key diagram of the situation about obtaining of expression adjacent area.Adjacent area acquisition unit 230 generates area E R1 by bore region SR1 being enlarged handle.Enlarge to handle the adjacent area NR1 that the zone (area E R1-region S R1) that has enlarged becomes bore region SR1 thus.Equally, handle from the expansion by bore region SR2～SR6 and remove regional NR2～NR6 of bore region SR2～SR6 the area E R2～ER6 that generates, promptly become the adjacent area of bore region SR2～SR6 by enlarging regional NR2～NR6 that processing enlarged.
Among the step S400 of Fig. 3, special processing region configuration part 240, corresponding to the color of pixel that is comprised in each adjacent area, whether decision is set and is used to carry out the special processing region that special defective detects processing.Specifically, when in the adjacent area of certain bore region, having the pixel of specified color (being gold in the present embodiment), around this bore region, set special processing region.
Fig. 5 A～Fig. 5 C is the key diagram of situation of the setting of the special processing region of expression.Fig. 5 A represents printed base plate PCB (Fig. 2) is made a video recording and the coloured image IM that obtains.In addition, Fig. 5 A is identical with Fig. 4 A.Fig. 5 b represents the configuration of adjacent area NR1～NR6 and the color region under these regional NR1～NR6.In addition, Fig. 5 C represents the special processing region PR2 that sets among the step S400, the configuration of PR3, PR5.In addition, the border of the color region shown in the dotted line presentation graphs 5A that describes among Fig. 5 B and Fig. 5 C.
Shown in Fig. 5 B, because the adjacent area NR1 of bore region SR1 is included among the GD of dirty-green zone, so there is not golden pixel in adjacent area NR1.Therefore, in bore region SR1, do not set special processing region.On the other hand, the adjacent area NR2 of bore region SR2 is comprised among the golden regional GL.Therefore, in adjacent area NR2, there is golden pixel, in bore region SR2, sets special processing region PR2 (Fig. 5 C).Equally, in adjacent area NR3, NR5, exist among bore region SR3, the SR5 of golden pixel, be set with special processing region PR3, PR5.On the other hand, because in adjacent area NR4, NR6, there is not golden pixel, so in bore region SR4, SR6, do not set special processing region.
Identical with adjacent area, special processing region is to handle by the expansion of bore region to set.In the example of Fig. 5 C, special processing region configuration part 240 is carried out the expansion of bore region SR2 and is handled.And, be set to special processing region corresponding to bore region SR2 by enlarge handling the regional PR2 that has enlarged.Other bore region SR3, SR5 are too.
In addition, among Fig. 5 C,, can set the magnitude relationship of special processing region and adjacent area arbitrarily though special processing region PR2, PR3, PR5 describe greatlyyer than adjacent area NR2, NR3, NR5.
Among the step S500 of Fig. 3, special processing region inspection portion 250 detects the defective in each special processing region.Specifically, judge whether the color of pixel in the special processing region is included in the allowed band of color.And all color of pixel in special processing region are all in allowed band the time, and judging does not have defective in this special processing region, when the outer color pixel of allowed band is arranged in the processing region especially, judge defectiveness in this special processing region.
Fig. 6 A, Fig. 6 B are illustrated in the key diagram that the defective that does not have in the special processing region on the printed base plate of the defective PCB detects the situation of processing.Fig. 6 A represents the situation of the special processing region of printed base plate PCB.In addition, Fig. 6 B represents that each color of pixel distributes among the special processing region PR5 of printed base plate PCB.In addition, among Fig. 6 B, in order to illustrate conveniently, expression is described with bullet by 2 points of tieing up each color of pixel of color spaces (hereinafter referred to as " RB color space ") that R composition and 2 color components of B composition constitute.
As shown in Figure 6A, in printed base plate PCB, be included among the golden regional GL with 2 bore region SR2, SR3 corresponding 2 special processing region PR2, PR3.On the other hand, the special processing region PR5 corresponding with bore region SR5 crosses over golden regional GL and 2 color regions of bright green area GB.As mentioned above, the border of these two color region GL, GB, since also indeterminate in the coloured image that obtains reality printed base plate PCB is made a video recording, so the pixel color among the special processing region PR5 presents continuous variation from gold to bright green.Therefore, represent the point of pixel color among the special processing region PR5, shown in Fig. 6 B, be distributed in a part of scope XR in the RB color space of crossing over the golden scope of the green scope of expression and expression (below be also referred to as " physical presence color gamut XR ").
Shown in Fig. 6 A, Fig. 6 B, when all color of pixel in special processing region PR5 all belonged to physical presence color gamut XR, special processing region inspection portion 250 judged in special processing region PR5 there is not defective.For other special processing region PR2, PR3, also be whether to be included in the physical presence color gamut of setting by each special processing region by the color of pixel in the special processing region to judge having or not of defective.
In addition, in the present embodiment, whether each color of pixel among the special processing region PR5 belongs to physical presence color gamut XR, by judging with reference to that generate, that be kept in the external memory 50 in advance question blank LUT before checking each substrate.Here, so-called question blank LUT, be when having imported the rgb value (being also referred to as " input point ") of the individual colors in the expression RGB color space, whether output expression individual colors belongs to the table of the value (when for example belonging to physical presence color gamut XR is 1, is 0 when not belonging to) of physical presence color gamut XR.
Question blank LUT uses the printed base plate that does not have defective is made a video recording and the master image of the coloured image that obtains, generates based on each pixel color in the special processing region of master image (following also become " physical presence color ").Specifically, question blank LUT exports 0 question blank by generating corresponding to all input points, and will change to 1 and generate corresponding to the output valve of each color of pixel of the special processing region of this master image.Like this, by the output valve of rewritten query table, output valve is that 1 color gamut becomes physical presence color gamut XR.In addition, after preferred question blank LUT generates, at each input point of question blank LUT, input point 8 near any one of output valve be 1 o'clock, carry out the output valve corresponding to input point changed to 1 expansion process.By carrying out this expansion process, be included among the physical presence color gamut XR as the color color that exists in the printed base plate that does not have defective, that in the special processing region of master image, do not have to occur and be fine.
In addition, question blank LUT in the present embodiment, though whether output expression individual colors belongs to the value (physical presence is color-coded) of physical presence color gamut, the output valve of question blank is so long as it is just passable to judge whether individual colors belongs to the value of physical presence color gamut.For example, also can be by representing that individual colors is to belong to the output valve of the value of the value (color number) in zone of which color of expression in the color space and the color-coded generation of physical presence as question blank.
In addition, in the present embodiment,, also be fine with other method though be to judge by reference look-up tables LUT whether the pixel color in the special processing region is included in the color gamut of permission.For example, by each RGB composition capping value and lower limit, the RGB of the pixel in the special processing region respectively becomes score value, in the time of respectively between higher limit and lower limit, just can judge that this pixel color is in the color gamut that allows.
Fig. 7 A, Fig. 7 B are illustrated in the key diagram that the defective on the defective printed base plate PCB in the special processing region detects the situation of processing.Identical with Fig. 6, Fig. 7 A represents the situation of the special processing region of printed base plate PCB, and Fig. 7 B represents that each color of pixel among the special processing region PR5 of printed base plate PCB distributes.
In the example that Fig. 7 A represents, in special processing region PR5, there are 2 image-region DF1, the DF2 (being called " defect image zone ") of 2 defectives of expression.These defect image region D F1, DF2 as the black triangle of Fig. 7 B and black quadrilateral are represented, have and the different color of color that shows in the special processing region PR5 of the printed base plate PCB that does not have defective respectively.Like this, in special processing region among the defective printed base plate PCB, special processing region PR5, except the colored pixels of representing with bullet that belongs to physical presence color gamut XR, the color pixel that does not belong to physical presence color gamut XR that also useful black triangle and black quadrilateral are represented.Therefore, defectiveness among the special processing region PR5 is judged by special processing region inspection portion 250.
Among the step S600 of Fig. 3, mask exterior domain inspection portion 260 detects the defective in the image-region (mask exterior domain) of having removed special processing region and bore region (these zones are referred to as " masks area ") from coloured image IM.Specifically, Region Segmentation mask exterior domain for example based on position, the shape in the representative color zone on this Region Segmentation result, is judged the having or not of defective of printed base plate PCB.The mask exterior domain because removed the special processing region of color allowed band broadness from coloured image IM, carries out the high defective of reliability so the accessible region territory is cut apart and detects.
Fig. 8 A, Fig. 8 B are that expression is carried out the key diagram that defective detects the result of processing for the printed base plate PCB that does not have defective.In the comparative example of not setting special processing region, shown in Fig. 8 A, by Region Segmentation, the color of thin resist region R TR becomes golden regional GL.Therefore, judge that the zone became green coated resist originally becomes gold, thin resist region R TR is detected as defective.On the other hand, in the present embodiment,, thin resist region R TR detects processing, so thin resist region R TR is not detected as defective because having been carried out special defective.
Like this, in the present embodiment, around bore region, adjacent area is set, because when in adjacent area, comprising specific color, enlarge and to comprise the masks area of bore region, thus can lower will be adjacent with bore region thin resist region R TR erroneous judgement be the possibility of defective.In addition,, detect the defective of special processing region, so can detect the defective that around bore region, exists because around bore region, special processing region is set.
B. second embodiment:
Fig. 9 is the process flow diagram of the order of the expression defective that detects printed base plate PCB among second embodiment.The process flow diagram of Fig. 9, the process flow diagram of representing with Fig. 3 on 2 step S310, the S320 this point is different having increased between step S200, the S300.Other are identical with Fig. 3.
In step S310, adjacent area acquisition unit 230 (Fig. 1) is obtained borderline region by the expansion processing of bore region.And in step S320, corresponding to the color of borderline region, whether decision needs to have used the setting of the special processing region of adjacent area to judge.
Figure 10 A, Figure 10 B are the key diagrams of situation of the setting of the expression adjacent area that used borderline region.Figure 10 A is illustrated near the configuration of the color region of 2 bore region SR1, SR3.Adjacent area acquisition unit 230 (Fig. 1) is obtained borderline region TR1, TR3 by the expansion processing of bore region SR1, SR3.And when having the pixel of color (for example golden) of regulation among borderline region TR1, the TR3, judge the setting that needs special processing region, around bore region SR1, SR3, set adjacent area.In addition, for other bore region, also carry out same processing.
Figure 10 B represents to use borderline region TR1, TR3 and the configuration of the adjacent area set.In the example of Figure 10, because in borderline region TR1 (being included in dirty-green zone GD), do not have golden pixel, so do not set adjacent area at bore region SR1.On the other hand, owing in borderline region TR3 (comprising golden regional GL), there is golden pixel, set adjacent area NR3 at bore region SR3.In addition, shown in Figure 10 B, adjacent area NR3 sets greatlyyer than borderline region TR3.
Like this, in a second embodiment, whether the color decision corresponding to borderline region needs the setting of special processing region to judge, by the bore region of only judging for the setting of the special processing region of needs, having or not of the setting of the special processing region of decision use adjacent area, can reduce the Flame Image Process amount of adjacent area.
C. variation:
In addition, the present invention is not limited to the foregoing description and form of implementation, in the scope that does not break away from its aim, can implement in various forms, and for example Xia Mian distortion also is fine.
C1. variation 1:
In the various embodiments described above,, also can extract the specific zone beyond the bore region and set special processing region on every side at it though around bore region, set special processing region.At this moment, the extraction in the zone of the specific zone particular color that can generate or by obtaining to wait and obtain from the position of the specific region of cad data and shape by Region Segmentation according to coloured image IM.In addition, as the specific region, preferably in coloured image, select to have the zone of specified color (for example black).
C2. variation 2:
In the various embodiments described above, though by what whether adjacent area contained that specific color pixel decides special processing region will deny,, in general, what can decide special processing region corresponding to the color of adjacent area will deny.Specifically, the regulation color space in adjacent area in each color of pixel the regulation the color distribution scope outside the time, also can set special processing region.In addition, when adjacent area is crossed in a plurality of representative colors zone of regulation 2 or representative color zone more than it, also can set special processing region.At this, so-called " representative color zone " is exactly according to the Region Segmentation of having used a plurality of representative colors and by the zone on the image corresponding with each representative color of coloured image generation.At this moment, whether adjacent area crosses over the above representative color zone of 2 or its, the quantity that can be included in the representative color zone of adjacent area by the Region Segmentation of carrying out coloured image in advance whether be 2 or its with on judge.
C3. variation 3:
In the various embodiments described above, among the step S500 of Fig. 3, though the defective that special processing region inspection portion 250 (Fig. 1) carries out in the special processing region detects processing, also the defective that can omit in the special processing region detects processing.That is, the masks area that special processing region and bore region have been added up to does not carry out defective and detects processing, and the outer zone of masks area carries out defective and detect processing.
C4. variation 4:
In the various embodiments described above, though corresponding to the special processing region that determines by each bore region will deny to set masks area (=bore region+special processing region), but in general, as long as can change the size of masks area corresponding to the color of adjacent area.At this moment, set greatly when masks area comprises specific color pixel in adjacent area, set for a short time when in adjacent area, not comprising specific color pixel.
C5. variation 5:
In the various embodiments described above, though use bore region to set special processing region, also can have no relations, and the zone that will satisfy rated condition is set at special processing region with bore region.As the zone of satisfying rated condition, for example can be as the gold-plated region R GP of printed base plate PCB, the zone that becomes the zone of predetermined position and shape or have the specified color on every side of bore region.
C6. variation 6:
Defective in the setting of special processing region of the present invention and the special processing region detects, be not limited to printed base plate, go for the defective of arbitrary objects is detected, this arbitrary objects has makes a video recording and color allowed band on the coloured image that obtains different image-region mutually to this inspection object.Also defectives of these inspection objects of the coloured image that has obtained applicable to for example having used semiconductor wafer that figure is formed or the inspections objects such as mechanical part with complicated shape to make a video recording detect.
Claims (13)
1. a use is made a video recording to the inspection object that comprises through hole and the coloured image that obtains detects the method for the defective of above-mentioned inspection object, it is characterized in that, comprising:
(a) obtain the operation of the specific region suitable in the above-mentioned coloured image with above-mentioned through hole;
(b) obtain above-mentioned specific region around the operation of adjacent area;
(c) setting comprises the operation of the masks area of above-mentioned specific region;
(d) at the zone beyond the aforementioned mask zone in the above-mentioned coloured image, the defective of stipulating detects the operation of processing,
The size in aforementioned mask zone changes corresponding to the color of above-mentioned adjacent area.
2. the method for claim 1 is characterized in that,
The aforementioned mask zone has above-mentioned specific region and comprises the special processing region on every side of above-mentioned specific region,
Said method also comprises:
(e), carry out detecting the operation of handling different special defect processing with the defective of afore mentioned rules at above-mentioned special processing region.
3. method as claimed in claim 2 is characterized in that, also comprises:
(f) in the color space of regulation, the physical presence color gamut of the scope of the color that the above-mentioned special processing region of setting expression obtains and the operation of the non-physical presence color gamut beyond the above-mentioned physical presence color gamut,
Above-mentioned operation (e) comprising:
(e1) by judging which of above-mentioned physical presence color gamut and above-mentioned non-physical presence color gamut be each color of pixel that constitutes above-mentioned special processing region belong to, and detects the operation of the defective in the above-mentioned special processing region.
4. method as claimed in claim 3 is characterized in that,
Above-mentioned operation (f) comprising:
Preparation is as the operation of the master image of the coloured image of the standard of above-mentioned inspection object;
The scope of the part of the color space of the afore mentioned rules that comprises the color of pixel that constitutes above-mentioned special processing region in the above-mentioned master image is set at the operation of above-mentioned physical presence color gamut.
5. method as claimed in claim 3 is characterized in that,
Above-mentioned operation (f) comprises the operation of making question blank, this question blank with the color arbitrarily in the color space of afore mentioned rules as input, the above-mentioned color arbitrarily of output expression belongs to which the value in above-mentioned physical presence color gamut and the above-mentioned non-physical presence color gamut
Above-mentioned operation (e1) comprising: with reference to above-mentioned question blank, judge that above-mentioned each color of pixel belongs to which the operation in above-mentioned physical presence color gamut and the above-mentioned non-physical presence color gamut.
6. method as claimed in claim 5 is characterized in that,
Above-mentioned operation (f) comprising:
Preparation is as the operation of the master image of the coloured image of the standard of above-mentioned inspection object;
The scope of the part of the color space of the afore mentioned rules that comprises the color of pixel that constitutes above-mentioned special processing region in the above-mentioned master image is set at the operation of above-mentioned physical presence color gamut.
7. as each described method in the claim 1 to 6, it is characterized in that,
Above-mentioned operation (b) comprising:
(b1) obtain the operation of borderline region of the periphery of above-mentioned specific region;
When (b2) color of above-mentioned borderline region satisfies defined terms, comprise above-mentioned borderline region and obtain the operation of the above-mentioned adjacent area bigger than above-mentioned borderline region;
When (b3) color of above-mentioned borderline region does not satisfy the condition of afore mentioned rules, the size in aforementioned mask zone is set at the operation of prescribed level.
8. method as claimed in claim 7 is characterized in that,
The condition of afore mentioned rules is the such condition of pixel that has first color gamut of regulation in above-mentioned borderline region,
Above-mentioned operation (c) comprising: the color of pixel in the above-mentioned adjacent area is compared the operation that aforementioned mask zone is increased outside second color gamut of regulation the time with the situation in above-mentioned second color gamut.
9. as each described method in the claim 1 to 6, it is characterized in that,
Above-mentioned inspection object is a printed base plate.
10. a use is made a video recording to the inspection object that comprises through hole and the coloured image that obtains detects the device of the defective of above-mentioned inspection object, it is characterized in that, comprising:
Obtain the specific region acquisition unit of the specific region suitable in the above-mentioned coloured image with above-mentioned through hole;
Obtain the adjacent area acquisition unit of the adjacent area on every side of above-mentioned specific region;
Setting comprises the masks area configuration part of the masks area of above-mentioned specific region;
At the zone beyond the aforementioned mask zone in the above-mentioned coloured image, the defective of stipulating detects the outer inspection of the masks area portion of processing,
The size in aforementioned mask zone changes corresponding to the color of above-mentioned adjacent area.
11. device as claimed in claim 10 is characterized in that,
The aforementioned mask zone has above-mentioned specific region and comprises the special processing region on every side of above-mentioned specific region,
Said apparatus also possesses: at above-mentioned special processing region, carry out detecting the special processing region inspection portion that handles different special defect processing with the defective of afore mentioned rules.
12. device as claimed in claim 11 is characterized in that, also possesses:
Physical presence color gamut configuration part, it sets the physical presence color gamut and the above-mentioned physical presence color gamut non-physical presence color gamut in addition of the scope of the color of representing that above-mentioned special processing region obtains in the color space of regulation,
Above-mentioned special processing region inspection portion has special processing region color detection unit, which of above-mentioned physical presence color gamut and above-mentioned non-physical presence color gamut this special processing region color detection unit belong to by judging each color of pixel that constitutes above-mentioned special processing region, detects the defective in the above-mentioned special processing region.
13. device as claimed in claim 12 is characterized in that,
Above-mentioned physical presence color gamut configuration part has the question blank preparing department that makes question blank, this question blank with the color arbitrarily in the color space of afore mentioned rules as input, the above-mentioned color arbitrarily of output expression belongs to which the value in above-mentioned physical presence color gamut and the above-mentioned non-physical presence color gamut
Above-mentioned special processing region color detection unit has question blank with reference to portion, and this question blank with reference to above-mentioned question blank, judges which in above-mentioned physical presence color gamut and the above-mentioned non-physical presence color gamut be above-mentioned each color of pixel belong to reference to portion.
Applications Claiming Priority (3)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP2004261789A JP4841819B2 (en) | 2004-09-09 | 2004-09-09 | Defect detection by color image of an object |
| JP2004261789 | 2004-09-09 | ||
| JP2004261792 | 2004-09-09 |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN1746666A true CN1746666A (en) | 2006-03-15 |
| CN100587480C CN100587480C (en) | 2010-02-03 |
Family
ID=36157890
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN200510092185A Active CN100587480C (en) | 2004-09-09 | 2005-08-22 | Object based color image defect detection method and device |
Country Status (2)
| Country | Link |
|---|---|
| JP (1) | JP4841819B2 (en) |
| CN (1) | CN100587480C (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101025398B (en) * | 2006-02-20 | 2010-06-02 | 牧德科技股份有限公司 | Hollow-bulge analyzing method for micro-pore after filled by copper |
| CN104362111A (en) * | 2014-11-27 | 2015-02-18 | 阳光硅峰电子科技有限公司 | Silicon wafer edge breakage automatic detection method |
| CN104848808A (en) * | 2015-06-03 | 2015-08-19 | 湖南大学 | Surface roughness detection method and equipment |
| CN105223139A (en) * | 2014-06-10 | 2016-01-06 | 何卫 | Spatial multi-region color synchronous detection method and device applying same |
| CN105279757A (en) * | 2015-10-19 | 2016-01-27 | 广州视源电子科技股份有限公司 | Welding spot locating method and device |
| CN104520698B (en) * | 2012-05-28 | 2017-02-22 | Msc&Sgcc公司 | Optical method for inspecting transparent or translucent containers bearing visual motifs |
| CN109789697A (en) * | 2016-09-21 | 2019-05-21 | 小森公司 | Press quality check device and press quality inspection method |
| CN111242896A (en) * | 2019-12-31 | 2020-06-05 | 电子科技大学 | Color printing label defect detection and quality rating method |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP4745633B2 (en) * | 2004-09-09 | 2011-08-10 | 大日本スクリーン製造株式会社 | Defect detection method and apparatus using color image of object |
| WO2012164654A1 (en) * | 2011-05-28 | 2012-12-06 | 株式会社メガトレード | Automatic inspection device and automatic inspection method for inspecting silk backing |
| EP3822619A4 (en) * | 2018-07-13 | 2021-07-21 | Fuji Corporation | Foreign matter detection method and electronic component mounting device |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP3179874B2 (en) * | 1992-07-23 | 2001-06-25 | 松下電器産業株式会社 | Teaching method and teaching device for image inspection device |
| JP3220526B2 (en) * | 1992-08-07 | 2001-10-22 | 松下電器産業株式会社 | Teaching method and teaching device for image inspection device |
| JP2790037B2 (en) * | 1994-05-31 | 1998-08-27 | 日本電気株式会社 | Appearance inspection method and appearance inspection apparatus |
| JP3560473B2 (en) * | 1998-05-26 | 2004-09-02 | 大日本スクリーン製造株式会社 | Printed circuit board inspection apparatus and printed circuit board inspection method |
| JP2003172711A (en) * | 2001-09-26 | 2003-06-20 | Dainippon Screen Mfg Co Ltd | Surface inspection of inspected object using image processing |
-
2004
- 2004-09-09 JP JP2004261789A patent/JP4841819B2/en active Active
-
2005
- 2005-08-22 CN CN200510092185A patent/CN100587480C/en active Active
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101025398B (en) * | 2006-02-20 | 2010-06-02 | 牧德科技股份有限公司 | Hollow-bulge analyzing method for micro-pore after filled by copper |
| CN104520698B (en) * | 2012-05-28 | 2017-02-22 | Msc&Sgcc公司 | Optical method for inspecting transparent or translucent containers bearing visual motifs |
| CN105223139A (en) * | 2014-06-10 | 2016-01-06 | 何卫 | Spatial multi-region color synchronous detection method and device applying same |
| CN104362111A (en) * | 2014-11-27 | 2015-02-18 | 阳光硅峰电子科技有限公司 | Silicon wafer edge breakage automatic detection method |
| CN104362111B (en) * | 2014-11-27 | 2017-02-01 | 阳光硅峰电子科技有限公司 | Silicon wafer edge breakage automatic detection method |
| CN104848808A (en) * | 2015-06-03 | 2015-08-19 | 湖南大学 | Surface roughness detection method and equipment |
| CN104848808B (en) * | 2015-06-03 | 2017-08-25 | 湖南大学 | A kind of Surface Roughness Detecting Method and equipment |
| CN105279757A (en) * | 2015-10-19 | 2016-01-27 | 广州视源电子科技股份有限公司 | Welding spot locating method and device |
| CN109789697A (en) * | 2016-09-21 | 2019-05-21 | 小森公司 | Press quality check device and press quality inspection method |
| CN109789697B (en) * | 2016-09-21 | 2021-04-02 | 小森公司 | Printing quality inspection device and printing quality inspection method |
| CN111242896A (en) * | 2019-12-31 | 2020-06-05 | 电子科技大学 | Color printing label defect detection and quality rating method |
Also Published As
| Publication number | Publication date |
|---|---|
| JP4841819B2 (en) | 2011-12-21 |
| CN100587480C (en) | 2010-02-03 |
| JP2006078300A (en) | 2006-03-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN1746666A (en) | Object based color image defect detection | |
| CN1728160A (en) | Technique for detecting a defect of an object by area segmentation of a color image of the object | |
| CN1725231A (en) | Surface region identification by image area segmentation | |
| US6947151B2 (en) | Surface state inspecting method and substrate inspecting apparatus | |
| CN1289901C (en) | Reduction of error alarm in PCB detection | |
| US7720274B2 (en) | Board inspection apparatus and method and apparatus for setting inspection logic thereof | |
| CN109215026B (en) | High-speed accurate LED defect detection method based on machine vision | |
| EP1736758B1 (en) | Board inspecting apparatus, its parameter setting method and parameter setting apparatus | |
| JP4736764B2 (en) | Substrate inspection device, inspection logic setting method and inspection logic setting device | |
| TWI618925B (en) | Defect inspection method and defect inspection system | |
| US8379964B2 (en) | Detecting semiconductor substrate anomalies | |
| JP4539355B2 (en) | Substrate inspection device, parameter setting method and parameter setting device | |
| CN1719455A (en) | Segmentation technique of a color image according to colors | |
| CN1725232A (en) | Segmentation technique of an image | |
| CN1276577A (en) | Pattern defect testing method | |
| CN1619790A (en) | Method for inspecting defect | |
| CN107507130A (en) | A kind of quickly QFN chip pins image obtains and amplification method | |
| JP3660936B1 (en) | Bubble measuring method and bubble measuring device for hardened concrete | |
| KR20120086410A (en) | Teaching data auto-generation apparatus of automated inspection machine and method for teaching data auto-generation the same | |
| JP2007033126A (en) | Substrate inspection device, parameter adjusting method thereof and parameter adjusting device | |
| JP5615076B2 (en) | Component presence / absence determination apparatus and component presence / absence determination method | |
| Bhardwaj | Machine vision algorithm for PCB parameters inspection | |
| CN111273921A (en) | Chip burning method, system and computer readable storage medium | |
| JP4745633B2 (en) | Defect detection method and apparatus using color image of object | |
| JP2004085543A (en) | System and method for visual inspection |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| C56 | Change in the name or address of the patentee |
Owner name: SCREEN GROUP CO., LTD. Free format text: FORMER NAME: DAINIPPON SCREEN MFG. CO., LTD. Owner name: DAINIPPON SCREEN MFG. CO., LTD. Free format text: FORMER NAME: DAINIPPON MESH PLATE MFR. CO., LTD. |
|
| CP01 | Change in the name or title of a patent holder |
Address after: Kyoto Japan Patentee after: Skilling Group Address before: Kyoto Japan Patentee before: DAINIPPON SCREEN MFG Co.,Ltd. Address after: Kyoto Japan Patentee after: DAINIPPON SCREEN MFG Co.,Ltd. Address before: Kyoto Japan Patentee before: Dainippon Screen Mfg. Co.,Ltd. |