CN103018260A - Defect detection method - Google Patents
Defect detection method Download PDFInfo
- Publication number
- CN103018260A CN103018260A CN2011102982449A CN201110298244A CN103018260A CN 103018260 A CN103018260 A CN 103018260A CN 2011102982449 A CN2011102982449 A CN 2011102982449A CN 201110298244 A CN201110298244 A CN 201110298244A CN 103018260 A CN103018260 A CN 103018260A
- Authority
- CN
- China
- Prior art keywords
- regional
- value
- maternal
- checked
- zone
- 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
Abstract
A defect detection method comprises the following steps: providing a regional female threshold for a plurality of regions on a target unit sheet of a female substrate; providing a substrate to be detected, and selecting a to-be-detected unit sheet and a neighboring unit sheet in the to-be-detected substrate; partitioning the to-be-detected unit sheet and the adjacent unit sheet to a plurality of regions, and acquiring the regional to-be-detected pixel of each region on the to-be-detection unit sheet, and acquiring the regional reference pixel in each region on the neighboring region; calculating the regional to-be-detection value and the regional reference value of each region on the to-be-detected unit sheet and the neighboring unit sheet, and acquiring the difference between the regional to-be-detection value and the regional reference value as the regional difference, which is an absolute value; and comparing the regional difference with the regional female threshold, and determining that a defect exists if the regional difference is larger than the regional female threshold. The defect detection method provided by the invention can improve the defect detection accuracy and efficiency.
Description
Technical field
The present invention relates to semiconductor applications, particularly a kind of defect inspection method to substrate.
Background technology
Develop rapidly along with VLSI (very large scale integrated circuit) ULSI (Ultra Large Scale Integration), the integrated circuit technology manufacture craft becomes and becomes increasingly complex with meticulous, this just requires to form semiconductor devices with higher precision and better homogeneity, and at actual integrated circuit production process, because the impact of the different factors such as technology and equipment, often can form the defective that to expect at substrate surface, for example: grain defect, finally cause the reduction of product yield, therefore, the defects detection of Semiconductor substrate also becomes most important in integrated circuit production.
Fig. 1 is the schematic flow sheet of prior art defect inspection method, comprises step: S101 provides threshold value; S102, regional reference pixel on the pixel to be checked in a zone and the adjacent dice on the acquisition unit sheet; S103, gray-scale value according to a regional reference pixel on the gray-scale value of the pixel to be checked in a zone on the unit sheet and the adjacent dice, obtain the value to be checked of whole unit sheet and the reference value of adjacent dice, and obtain described value to be checked and reference value between difference, described difference is absolute value; S104, the size of more described difference and threshold value when difference during greater than threshold value, is judged defective existence.Gray-scale value represents that the technology of pixel is that known technology is not described in detail in this.The value to be checked of described whole unit sheet is the weighted mean value of the gray-scale value of the pixel to be checked in a zone on the described unit sheet.
More methods about defects detection please refer to the United States Patent (USP) that the patent No. is US5995218A.
The value to be checked of the sheet of unit described in the prior art is the gray-scale value acquisition according to a zone on the unit sheet, although can improve sweep speed, but value to be checked can not accurately reflect the actual shape characteristic of unit sheet, during Defect Scanning, very easily cause the incorrect detection of defective, reduced the accuracy and efficiency of Defect Scanning.
Summary of the invention
The problem that the present invention solves has provided a kind of defect inspection method, has improved accuracy and the efficient of defects detection.
For addressing the above problem, the invention provides a kind of defect inspection method, comprise step:
The regional maternal threshold value in a plurality of zones on the maternal substrate one object element sheet is provided;
Provide substrate to be checked, unit to be checked sheet and an adjacent dice on the selected described substrate to be checked;
Described unit to be checked sheet and adjacent dice are divided into a plurality of zones, and obtain the pixel regional to be checked in each zone on the described unit to be checked sheet and the regional reference pixel in each zone on the adjacent dice;
According to described zone pixel to be checked and regional reference pixel, calculate value regional to be checked and the regional reference value in each zone on unit to be checked sheet and the adjacent dice, obtain described zone value to be checked and regional reference value between difference and with as regional difference, described regional difference is absolute value; The size of more described regional difference and regional maternal threshold value when described regional difference during greater than the maternal threshold value in zone, is judged defective existence.
Optionally, on described maternal substrate and the substrate to be checked a plurality of unit sheets with same patterned are arranged.
Optionally, the value regional to be checked of unit to be checked sheet is on the described substrate to be checked: the mean value of the gray-scale value of regional pixel to be checked, the maximum gradation value of regional pixel to be checked and the difference of minimum gradation value, regional grey scale pixel value to be checked or the variance yields of regional grey scale pixel value to be checked.
Optionally, the regional reference value of adjacent dice is on the described substrate to be checked: the variance yields of the difference of the gray-scale value of regional reference pixel, the maximum gradation value of regional reference pixel and minimum gradation value, the mean value of regional reference pixel gray-scale value or regional reference pixel gray-scale value.
Optionally, the quantity of described adjacent dice is equal to or greater than 1.
Optionally, the preparation method of the maternal threshold value in described zone is: maternal substrate is provided; Object element sheet and an adjacent dice on the selected described maternal substrate; Described object element sheet and adjacent dice are divided into the quantity N identical with the position zone, obtain the regional reference pixel in corresponding the 1st～N zone on the 1st～N regional regional maternal seized pixel of described object element sheet and the adjacent dice; According to the gray-scale value of the maternal seized pixel in described zone and regional reference pixel, calculate the regional reference value in regional maternal seized value 1st～N corresponding to adjacent dice the zone in the 1st～N zone of described object element sheet; Obtain on the described object element sheet corresponding the 1st～N regional regional reference value on the regional maternal seized value in the 1st～N zone and the adjacent dice between difference, as the maternal threshold value in zone, the maternal threshold value in described zone is absolute value with described difference.
Optionally, the regional maternal seized value of described n-quadrant is the weighted mean value of the regional maternal seized value in the 1st～N-1 zone.
Optionally, the reference value of described n-quadrant is the weighted mean value of the reference value in the 1st～N-1 zone.
Optionally, the user that is divided in zone realizes by human and machine interface unit on the described maternal substrate target unit sheet.
Optionally, the division numbers N in zone is equal to or greater than 3 on described maternal substrate target unit sheet and the adjacent dice.
Optionally, also comprise, store the regional dividing line positional information of object element sheet on the maternal seized value in described zone, regional maternal threshold value and the maternal substrate.
Optionally, the regional maternal seized value of object element sheet is on the described maternal substrate: the variance yields of the difference of the gray-scale value of regional maternal seized pixel, the maximum gradation value of regional maternal seized pixel and minimum gradation value, the mean value of regional maternal seized pixel gray-scale value or regional maternal seized pixel gray-scale value.
Optionally, the regional reference value of adjacent dice is on the described maternal substrate: the variance yields of the difference of the gray-scale value of regional reference pixel, the maximum gradation value of regional reference pixel and minimum gradation value, the mean value of regional reference pixel gray-scale value or regional reference pixel gray-scale value.
Optionally, the maternal threshold value in described zone is that the user is by the value of human and machine interface unit input.
Optionally, quantity and the position of zone division are identical on the quantity that the zone is divided on unit to be checked sheet and the adjacent dice on the described substrate to be checked and position and the maternal substrate one object element sheet.
Optionally, also comprise, provide the alerting signal that defective exists.
Compared with prior art, technical solution of the present invention has the following advantages:
Unit sheet on the described substrate to be checked is carried out the division of multizone, and obtain the value to be checked of respective regions according to the grey scale pixel value to be checked in each zone, the value to be checked in described a plurality of zones consists of the value to be checked of whole unit sheet, obtained the value to be checked of whole unit sheet by the gray-scale value in a zone than prior art, the actual shape characteristic that more can accurately reflect described unit to be checked sheet, therefore, when Defect Scanning, accuracy rate and efficient when improving Defect Scanning.
Further, the user that is divided in described zone realizes by human and machine interface unit, the user is according to the repeatability of figure on the unit sheet, closeness, quantity and position that the material of figure and the characteristics of technique come definite area to divide, improved the selectivity of user's operation, and the maternal threshold value in described zone is that the user is by the value of human and machine interface unit input, the user can come the maternal threshold value of setting regions with reference to the difference between the regional reference value of the regional maternal seized value of object element sheet on the maternal substrate and adjacent dice, the user thinks that the important corresponding regional maternal Threshold in zone is the value than described difference little 5%～20% in the technological process, the corresponding regional maternal Threshold in unessential zone is the value than described difference large 5%～20%, the problem of avoiding real defect can't detect has improved accuracy rate and the efficient of Defect Scanning.
Description of drawings
Fig. 1 is the schematic flow sheet of prior art defect inspection method;
Fig. 2 is the schematic flow sheet of defect inspection method of the present invention;
Fig. 3 is the structure simplified diagram of defect detection tool of the present invention;
Fig. 4～Fig. 8 is the maternal substrate of the present invention and substrat structure schematic diagram to be checked.
Embodiment
Find in the process of inventor's defective on detecting Semiconductor substrate, the value to be checked of unit sheet is the gray-scale value acquisition according to a zone on the unit sheet in the prior art, and the differentiation in zone is apparent in view on the product unit sheet, described value to be checked can not accurately reflect the actual shape characteristic of unit sheet, during Defect Scanning, very easily cause the incorrect detection of defective, defects detection efficient and accuracy rate are low, for overcoming the above problems, the inventor proposes a kind of defect inspection method.
With reference to figure 2, be the schematic flow sheet of defect inspection method embodiment of the present invention.
Step S201 provides the regional maternal threshold value in a plurality of zones on the maternal substrate one object element sheet;
Step S202 provides substrate to be checked, unit to be checked sheet and an adjacent dice on the selected described substrate to be checked;
Step S203 is divided into a plurality of zones with described unit to be checked sheet and adjacent dice, and obtains the pixel regional to be checked in each zone on the described unit to be checked sheet and the regional reference pixel in each zone on the adjacent dice;
Step S204, according to described zone pixel to be checked and regional reference pixel, calculate value regional to be checked and the regional reference value in each zone on unit to be checked sheet and the adjacent dice, obtain described zone value to be checked and regional reference value between difference and with as regional difference, described regional difference is absolute value;
Step S205, the size of more described regional difference and regional maternal threshold value when described regional difference during greater than the maternal threshold value in zone, is judged defective existence.
With reference to figure 3, the structure simplified diagram for Defect Scanning instrument used in the present invention comprises: image acquisition units 310, for the surface of scanning substrate, Information Monitoring; CPU (central processing unit) 311, for the information that communicates with described image acquisition units 310 and Treatment Analysis image acquisition units 310 gathers, described CPU (central processing unit) 311 also has the function of storage relevant information; Human and machine interface unit 312 is used for showing the analysis result of described CPU (central processing unit) 311 and the Information Monitoring of image acquisition units 310, and communicates as user terminal and CPU (central processing unit) 311.
Step S201 provides the regional maternal threshold value in a plurality of zones on the maternal substrate one object element sheet.
With reference to figure 4, maternal substrate 30 is provided, be formed with many unit sheets (Die) with same patterned on the described maternal substrate 30, for example unit sheet 301, unit sheet 302, unit sheet 303.Be formed with similar figure on the described unit sheet (Die), for example the figure of L-type shown in Figure 5 and I type.Note can comprising a large amount of figures in each unit sheet, these figures are divided into again many layers, as the part of groove, through hole or grid etc.
Object element sheet and an adjacent dice on the selected described maternal substrate 30, described object element sheet generally are positioned in the middle of the described maternal substrate 30, also can be positioned at other positions; Described adjacent dice is positioned at described object element sheet left side or right side.The selected object element sheet of the present embodiment is 301, and adjacent dice is 302.
With reference to figure 6, the user is divided into a plurality of zones by human and machine interface unit 312 (shown in Figure 3) with described object element sheet 301, the division numbers N in described zone is equal to or greater than 3, described regional dividing line be shaped as rectangle, or other suitable figures, the quantity that the sheet of object element described in the present embodiment 301 zones are divided is 3, comprising first area 3030, second area 3040, the 3rd zone 3050, described first area 3030, second area 3040 is determined to divide by described human and machine interface unit 312 for the user, described the 3rd zone 3050 is confirmed voluntarily for described Defect Scanning instrument, for removing first area 3030 on the described object element sheet 301, other zones beyond the second area 3040; Preserve described first area 3030, second area 3040, the positional information of the 3rd zone 3050 regional dividing lines, described regional dividing line position acquiring method is: take described second area 3040 dividing line positional informations as example, select a point on the described object element sheet 301 as initial point, be initial point such as 20, the coordinate of initial point is (0,0), calculate four summits 50 of the rectangle of described second area 3040 dividing lines formation, 60,70,80 coordinate position, the coordinate position on described four summits are second area 3040 regional dividing line positional informations; Obtain and store the regional maternal seized pixel in described first area 3030, second area 3040, the 3rd zone 3050, the maternal seized pixel in described zone is gray-scale value, namely represent pixel with gray-scale value, represent that with gray-scale value the technology of pixel is that known technology is not described in detail in this, all pixels all are to represent with gray-scale value among the present invention; Selected adjacent dice 302, and described adjacent dice 302 is divided into 3 identical zones according to the regional dividing line positional information of described object element sheet 301 of storage, comprise first area 3060, second area 3070, the 3rd zone 3080, and obtain simultaneously the regional reference pixel of respective regions.The selection of region quantity and position is determined according to material and the importance of figure in processing step of repeatability, closeness or the figure of figure on the unit sheet by the user on the described object element sheet 301.
According to regional reference pixel on regional maternal seized pixel and the adjacent dice 302 on the described object element sheet 301, calculate the regional reference value in first area 3060, second area 3070, the 3rd zone 3080 on the regional maternal seized value in first area 3030 on the described object element sheet 301, second area 3040, the 3rd zone 3050 and the adjacent dice 302; Obtain the difference between the corresponding regional reference value on the regional maternal seized value of described object element sheet 301 and the described adjacent dice 302, as the maternal threshold value in zone, the maternal threshold value in described zone is absolute value with described difference, stores the maternal threshold value in described zone.
The maternal threshold value in described zone also can be inputted by human and machine interface unit by the user, the user thinks that according to process characteristic the regional maternal Threshold in important zone is a little bit smaller, for example: be mainly memory block (zone that comprises a large amount of repetitive patterns) in the first area on object element sheet 301 and the adjacent dice 302, then the regional maternal threshold value setting of corresponding first area is little 5%～20% value than the difference between the 3060 regional reference values of first area on the 3030 regional maternal seized values of first area on the described object element sheet 301 and the adjacent dice 302; The regional maternal Threshold in unessential zone more greatly, for example: the 3rd zone is mainly background area (being generally very smooth zone) on object element sheet 301 and the adjacent dice 302, then the regional maternal threshold value setting in corresponding the 3rd zone is for than the 3rd zone 3050 regional maternal seized values on the described object element sheet 301 and adjacent dice being large 5%～20% value of the difference between the 3rd regional 3080 regional reference values on 302, adopt such setting, avoid real defects detection less than problem, improved the accuracy and efficiency of Defect Scanning.
The regional maternal seized value of object element sheet is on the maternal substrate: the variance yields of the difference of the gray-scale value of regional maternal seized pixel, the maximum gradation value of regional maternal seized pixel and minimum gradation value, the mean value of regional maternal seized pixel gray-scale value or regional maternal seized pixel gray-scale value.For example the regional maternal seized value of first area 3030 is on the described object element sheet 301: the mean value of the maximum gradation value of the gray-scale value of first area 3030 maternal seized pixels, first area 3030 maternal seized pixels and the difference of minimum gradation value, first area 3030 maternal seized pixel gray-scale values or the variance yields of first area 3030 maternal seized pixel gray-scale values.
The variance yields that regional maternal seized value is regional maternal seized pixel gray-scale value described in the present embodiment.
The regional reference value of adjacent dice is on the maternal substrate: the variance yields of the difference of the gray-scale value of regional reference pixel, the maximum gradation value of regional reference pixel and minimum gradation value, the mean value of regional reference pixel gray-scale value or regional reference pixel gray-scale value.
The selection of the regional reference value of adjacent dice is corresponding with the selection of the regional maternal seized value of object element sheet on the described maternal substrate, and the regional reference value of adjacent dice described in the present embodiment is the variance yields of regional reference pixel gray-scale value.
In other embodiments, the regional maternal seized value in the 3rd zone 3050 is the weighted mean value of regional maternal seized value and the second area 3040 regional maternal seized values of first area 3030 on the described object element sheet 301.In actual production process, the 3rd zone 3050 is regarded as requiring relatively on the low side to defective, the 3rd zone 3050 regional maternal seized values that regional maternal seized value by first area 3030 and the weighted mean value of second area 3040 regional maternal seized values obtain can reflect the shape characteristic in the 3rd zone 3050 substantially, and do not need to obtain by the 3rd zone 3050 maternal seized pixels, save computing time, improved efficient.The regional reference value in the 3rd zone 3080 is the weighted mean value of first area 3060 and second area 3070 regional reference values on the described adjacent dice 302, and the flexible strategy that the 3rd zone 3050 maternal seized values are calculated on the flexible strategy of described weighted mean value and the object element sheet 301 are identical.
Step S202 provides substrate to be checked, unit to be checked sheet and an adjacent dice on the selected described substrate to be checked.
With reference to figure 7 and Fig. 8, substrate 40 to be checked is provided, described substrate 40 to be checked is formed with the unit sheet (Die) with described maternal substrate 30 (shown in Figure 4) same patterned.
The quantity of described adjacent dice is equal to or greater than 1, and the quantity of adjacent dice described in the present embodiment is 2.
Step S203 and step S204 are divided into a plurality of zones with described unit to be checked sheet and adjacent dice, and obtain the pixel regional to be checked in each zone on the described unit to be checked sheet and the regional reference pixel in each zone on the adjacent dice; According to described zone pixel to be checked and regional reference pixel, calculate value regional to be checked and the regional reference value in each zone on unit to be checked sheet and the adjacent dice, obtain described zone value to be checked and regional reference value between difference and with as regional difference, described regional difference is absolute value.
With reference to figure 8, selected described unit to be checked sheet 401, the Defect Scanning instrument is divided into Three regions according to the regional dividing line positional information of object element sheet 301 on the maternal substrate of storage with described unit to be checked sheet 401, comprise: first area 4030, second area 4040, the 3rd zone 4050, obtain simultaneously described first area 4030, second area 4040, the pixel regional to be checked in the 3rd zone 4050 is according to first area 4030, second area 4040, the pixel regional to be checked in the 3rd zone 4050 is calculated described first area 4030, second area 4040, the value regional to be checked in the 3rd zone 4050; Selected described adjacent dice 402, the Defect Scanning instrument is divided into Three regions according to the regional dividing line positional information of object element sheet 301 on the maternal substrate of storage with described adjacent dice 402, comprise: first area 4060, second area 4070, the 3rd zone 4080, obtain simultaneously first area 4060, second area 4070, the 3rd zone 4080 regional reference pixels are according to first area 4060, second area 4070, the regional reference pixel in the 3rd zone 4080 calculates described first area 4060, second area 4070, the regional reference value in the 3rd zone 4080; Selected described adjacent dice 403, the Defect Scanning instrument is divided into Three regions according to the regional dividing line positional information of object element sheet 301 on the maternal substrate of storage with adjacent dice 403, comprise: first area 4090, second area 4010, the 3rd zone 4020, obtain simultaneously first area 4090, second area 4010, the regional reference pixel in the 3rd zone 4020 is according to first area 4090, second area 4010, the regional reference pixel in the 3rd zone 4020 obtains described first area 4090, second area 4010, the regional reference value in the 3rd zone 4020; Obtain the value regional to be checked of described unit to be checked sheet 401 and described adjacent dice 402 respective regions and the difference between the regional reference value as regional difference 1, obtain the value regional to be checked of described unit to be checked sheet 401 and described adjacent dice 403 respective regions and the difference between the regional reference value as regional difference 2.Described regional difference 1, regional difference 2 are absolute value.Unit sheet on the described substrate to be checked is carried out the division of multizone, and obtain the value to be checked of respective regions according to the grey scale pixel value to be checked in each zone, the value to be checked in described a plurality of zones consists of the value to be checked of whole unit sheet, obtained the value to be checked of whole unit sheet by the gray-scale value in a zone than prior art, the actual shape characteristic that more can accurately reflect described unit to be checked sheet, therefore, when Defect Scanning, accuracy rate and efficient when improving Defect Scanning.
The value regional to be checked of unit to be checked sheet is on the substrate to be checked: the mean value of the gray-scale value of regional pixel to be checked, the maximum gradation value of regional pixel to be checked and the difference of minimum gradation value, regional grey scale pixel value to be checked or the variance yields of regional grey scale pixel value to be checked.For example: the value regional to be checked of first area 4030 is on the described unit to be checked sheet 401: the mean value of the maximum gradation value of the gray-scale value of first area 4030 pixels to be checked, first area 4030 pixels to be checked and the difference of minimum gradation value, first area 4030 grey scale pixel values to be checked or the variance yields of first area 4030 grey scale pixel values to be checked.
On the substrate to be checked on the selection of the regional value to be checked of unit to be checked sheet and the maternal substrate selection of the regional maternal seized value of object element sheet corresponding, described in the present embodiment on the substrate to be checked the value regional to be checked of unit to be checked sheet be the variance yields of regional grey scale pixel value to be checked.
The regional reference value of adjacent dice is on the substrate to be checked: the variance yields of the difference of the gray-scale value of regional reference pixel, the maximum gradation value of regional reference pixel and minimum gradation value, the mean value of regional reference pixel gray-scale value or regional reference pixel gray-scale value.
On the described substrate to be checked on the selection of the regional reference value of adjacent dice and the maternal substrate selection of the regional maternal seized value of object element sheet corresponding, described in this example on the substrate to be checked the regional reference value of adjacent dice be the variance yields of regional reference pixel gray-scale value.
Execution in step S205, the size of more described regional difference and regional maternal threshold value when described regional difference during greater than the maternal threshold value in zone, is judged defective existence.
The detailed process of judging defective is: when described regional difference 1 during less than the maternal threshold value in described zone, then described to-be-measured cell sheet 401 does not have defective existence; When described regional difference 1 during greater than the maternal threshold value in described zone, and described regional difference 2 is greater than the maternal threshold value in zone, the then defective existence of described unit to be checked sheet 401 respective regions; When described regional difference 1 during greater than the maternal threshold value in described zone, and described regional difference 2 is less than the maternal threshold value in described zone, and then described unit to be checked sheet 401 does not have defective existence.
When defective existed, described defect detection tool provided the alerting signal that defective exists.
To sum up, adopt defect inspection method provided by the invention, unit sheet on the described substrate to be checked is carried out the division of multizone, and obtain the value to be checked of respective regions according to the grey scale pixel value to be checked in each zone, the value to be checked in described a plurality of zones consists of the value to be checked of whole unit sheet, obtained the value to be checked of whole unit sheet by the gray-scale value in a zone than prior art, the actual shape characteristic that more can accurately reflect described unit to be checked sheet, therefore, when Defect Scanning, accuracy rate and efficient when improving Defect Scanning.
Further, the user that is divided in described zone realizes by human and machine interface unit, the user is according to the repeatability of figure on the unit sheet, closeness, quantity and position that the material of figure and the characteristics of technique come definite area to divide, improved the selectivity of user's operation, and the maternal threshold value in described zone is that the user is by the value of human and machine interface unit input, the user can reference target unit sheet on regional maternal seized value and the adjacent dice difference between the reference value in zone come the maternal threshold value of setting regions, the user thinks that the important corresponding regional maternal Threshold in zone is the value than described difference little 5%～20% in the technological process, the corresponding regional maternal Threshold in unessential zone is the value than described difference large 5%～20%, the problem of avoiding real defect can't detect has improved accuracy rate and the efficient of Defect Scanning.
Although the present invention with preferred embodiment openly as above; but it is not to limit the present invention; any those skilled in the art without departing from the spirit and scope of the present invention; can utilize method and the technology contents of above-mentioned announcement that technical solution of the present invention is made possible change and modification; therefore; every content that does not break away from technical solution of the present invention; to any simple modification, equivalent variations and modification that above embodiment does, all belong to the protection domain of technical solution of the present invention according to technical spirit of the present invention.
Claims (16)
1. a defect inspection method is characterized in that, comprises step:
The regional maternal threshold value in a plurality of zones on the maternal substrate one object element sheet is provided;
Provide substrate to be checked, unit to be checked sheet and an adjacent dice on the selected described substrate to be checked;
Described unit to be checked sheet and adjacent dice are divided into a plurality of zones, and obtain the pixel regional to be checked in each zone on the described unit to be checked sheet and the regional reference pixel in each zone on the adjacent dice;
According to described zone pixel to be checked and regional reference pixel, calculate value regional to be checked and the regional reference value in each zone on unit to be checked sheet and the adjacent dice, obtain described zone value to be checked and regional reference value between difference and with as regional difference, described regional difference is absolute value; The size of more described regional difference and regional maternal threshold value when described regional difference during greater than the maternal threshold value in zone, is judged defective existence.
2. defect inspection method as claimed in claim 1 is characterized in that, a plurality of unit sheets with same patterned are arranged on described maternal substrate and the substrate to be checked.
3. defect inspection method as claimed in claim 1, it is characterized in that, the value regional to be checked of unit to be checked sheet is on the described substrate to be checked: the mean value of the gray-scale value of regional pixel to be checked, the maximum gradation value of regional pixel to be checked and the difference of minimum gradation value, regional grey scale pixel value to be checked or the variance yields of regional grey scale pixel value to be checked.
4. defect inspection method as claimed in claim 1, it is characterized in that, the regional reference value of adjacent dice is on the described substrate to be checked: the variance yields of the difference of the gray-scale value of regional reference pixel, the maximum gradation value of regional reference pixel and minimum gradation value, the mean value of regional reference pixel gray-scale value or regional reference pixel gray-scale value.
5. defect inspection method as claimed in claim 1 is characterized in that, the quantity of described adjacent dice is equal to or greater than 1.
6. defect inspection method as claimed in claim 1 is characterized in that, the preparation method of the maternal threshold value in described zone is: maternal substrate is provided; Object element sheet and an adjacent dice on the selected described maternal substrate; Described object element sheet and adjacent dice are divided into the quantity N identical with the position zone, obtain the regional reference pixel in corresponding the 1st～N zone on the 1st～N regional regional maternal seized pixel of described object element sheet and the adjacent dice; According to the gray-scale value of the maternal seized pixel in described zone and regional reference pixel, calculate the regional reference value in regional maternal seized value 1st～N corresponding to adjacent dice the zone in the 1st～N zone of described object element sheet; Obtain on the described object element sheet corresponding the 1st～N regional regional reference value on the regional maternal seized value in the 1st～N zone and the adjacent dice between difference, as the maternal threshold value in zone, the maternal threshold value in described zone is absolute value with described difference.
7. defect inspection method as claimed in claim 6 is characterized in that, the regional maternal seized value of described n-quadrant is the weighted mean value of the regional maternal seized value in the 1st～N-1 zone.
8. defect inspection method as claimed in claim 6 is characterized in that, the reference value of described n-quadrant is the weighted mean value of the reference value in the 1st～N-1 zone.
9. defect inspection method as claimed in claim 6 is characterized in that, the user that is divided in zone realizes by human and machine interface unit on the described maternal substrate target unit sheet.
10. defect inspection method as claimed in claim 6 is characterized in that, the division numbers N in zone is equal to or greater than 3 on described maternal substrate target unit sheet and the adjacent dice.
11. defect inspection method as claimed in claim 6 is characterized in that, also comprises, stores the regional dividing line positional information of object element sheet on the maternal seized value in described zone, regional maternal threshold value and the maternal substrate.
12. defect inspection method as claimed in claim 6, it is characterized in that, the regional maternal seized value of object element sheet is on the described maternal substrate: the variance yields of the difference of the gray-scale value of regional maternal seized pixel, the maximum gradation value of regional maternal seized pixel and minimum gradation value, the mean value of regional maternal seized pixel gray-scale value or regional maternal seized pixel gray-scale value.
13. defect inspection method as claimed in claim 6, it is characterized in that, the regional reference value of adjacent dice is on the described maternal substrate: the variance yields of the difference of the gray-scale value of regional reference pixel, the maximum gradation value of regional reference pixel and minimum gradation value, the mean value of regional reference pixel gray-scale value or regional reference pixel gray-scale value.
14. defect inspection method as claimed in claim 1 is characterized in that, the maternal threshold value in described zone is that the user is by the value of human and machine interface unit input.
15. defect inspection method as claimed in claim 1 is characterized in that, quantity and the position of zone division are identical on the quantity that the zone is divided on unit to be checked sheet and the adjacent dice on the described substrate to be checked and position and the maternal substrate one object element sheet.
16. defect inspection method as claimed in claim 1 is characterized in that, also comprises, provides the alerting signal that defective exists.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110298244.9A CN103018260B (en) | 2011-09-27 | 2011-09-27 | Defect detection method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110298244.9A CN103018260B (en) | 2011-09-27 | 2011-09-27 | Defect detection method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103018260A true CN103018260A (en) | 2013-04-03 |
| CN103018260B CN103018260B (en) | 2015-01-21 |
Family
ID=47967107
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201110298244.9A Active CN103018260B (en) | 2011-09-27 | 2011-09-27 | Defect detection method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103018260B (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108956616A (en) * | 2018-05-31 | 2018-12-07 | 南京航空航天大学 | A kind of workpiece, defect detection method based on image |
| CN110913682A (en) * | 2019-11-29 | 2020-03-24 | 深圳市智微智能软件开发有限公司 | SMT (surface Mount technology) reloading method and system |
| CN111289526A (en) * | 2020-02-28 | 2020-06-16 | 北京伟杰东博信息科技有限公司 | Method and system for detecting defects on inner surface of gas conveying pipeline |
| CN112345544A (en) * | 2020-09-24 | 2021-02-09 | 海克斯康制造智能技术(青岛)有限公司 | Multifunctional connecting rod automatic detection line |
| CN113012120A (en) * | 2021-03-09 | 2021-06-22 | 普迪飞半导体技术(上海)有限公司 | Gray value validity analysis method and device, electronic equipment and storage medium |
| CN116754578A (en) * | 2023-08-18 | 2023-09-15 | 国镓芯科(成都)半导体科技有限公司 | Detection system for wafer scratch and detection method of system |
| CN117705822A (en) * | 2024-02-06 | 2024-03-15 | 中国科学院长春光学精密机械与物理研究所 | Cylinder surface detection method and device based on CIS and storage medium |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107402218A (en) * | 2017-09-25 | 2017-11-28 | 武汉华星光电技术有限公司 | Microdefect detection method, device and the equipment of CF substrates |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6178257B1 (en) * | 1990-11-16 | 2001-01-23 | Applied Materials, Inc. | Substrate inspection method and apparatus |
| US20030099392A1 (en) * | 2001-11-28 | 2003-05-29 | Evgeni Levin | Method for detecting defects |
| US20080239319A1 (en) * | 2007-03-28 | 2008-10-02 | Hiroyuki Yamashita | Inspection apparatus and inspection method |
| US7440605B2 (en) * | 2002-10-08 | 2008-10-21 | Dainippon Screen Mfg. Co., Ltd. | Defect inspection apparatus, defect inspection method and program |
-
2011
- 2011-09-27 CN CN201110298244.9A patent/CN103018260B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6178257B1 (en) * | 1990-11-16 | 2001-01-23 | Applied Materials, Inc. | Substrate inspection method and apparatus |
| US20030099392A1 (en) * | 2001-11-28 | 2003-05-29 | Evgeni Levin | Method for detecting defects |
| US7440605B2 (en) * | 2002-10-08 | 2008-10-21 | Dainippon Screen Mfg. Co., Ltd. | Defect inspection apparatus, defect inspection method and program |
| US20080239319A1 (en) * | 2007-03-28 | 2008-10-02 | Hiroyuki Yamashita | Inspection apparatus and inspection method |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108956616A (en) * | 2018-05-31 | 2018-12-07 | 南京航空航天大学 | A kind of workpiece, defect detection method based on image |
| CN110913682A (en) * | 2019-11-29 | 2020-03-24 | 深圳市智微智能软件开发有限公司 | SMT (surface Mount technology) reloading method and system |
| CN111289526A (en) * | 2020-02-28 | 2020-06-16 | 北京伟杰东博信息科技有限公司 | Method and system for detecting defects on inner surface of gas conveying pipeline |
| CN111289526B (en) * | 2020-02-28 | 2021-08-17 | 涿州滨海燃气有限公司 | Method and system for detecting defects on inner surface of gas conveying pipeline |
| CN112345544A (en) * | 2020-09-24 | 2021-02-09 | 海克斯康制造智能技术(青岛)有限公司 | Multifunctional connecting rod automatic detection line |
| CN113012120A (en) * | 2021-03-09 | 2021-06-22 | 普迪飞半导体技术(上海)有限公司 | Gray value validity analysis method and device, electronic equipment and storage medium |
| CN116754578A (en) * | 2023-08-18 | 2023-09-15 | 国镓芯科(成都)半导体科技有限公司 | Detection system for wafer scratch and detection method of system |
| CN116754578B (en) * | 2023-08-18 | 2023-11-03 | 国镓芯科(成都)半导体科技有限公司 | Detection system for wafer scratch and detection method of system |
| CN117705822A (en) * | 2024-02-06 | 2024-03-15 | 中国科学院长春光学精密机械与物理研究所 | Cylinder surface detection method and device based on CIS and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103018260B (en) | 2015-01-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103018260B (en) | Defect detection method | |
| CN100499057C (en) | Chip detection method | |
| KR101860506B1 (en) | A system, a method and a computer program product for cadbased registration | |
| CN113688807B (en) | Self-adaptive defect detection method, device, recognition system and storage medium | |
| TWI614721B (en) | Detection of defects embedded in noise for inspection in semiconductor manufacturing | |
| JP4095860B2 (en) | Defect inspection method and apparatus | |
| KR101860393B1 (en) | A system, a method and a computer program product for cadbased registration | |
| CN109741324B (en) | Detection method, detection device and terminal equipment | |
| US20140355873A1 (en) | Method and apparatus for detecting repetitive pattern in image | |
| JP2008175549A (en) | Defect detection device and method | |
| US20190347505A1 (en) | Descriptor guided fast marching method for analyzing images and systems using the same | |
| CN111724335A (en) | Detection method and detection system | |
| CN110766095A (en) | Defect detection method based on image gray level features | |
| CN101236914A (en) | Wafer appearance detection device | |
| CN114764770A (en) | Wafer detection method, device, equipment and storage medium | |
| JP2016183916A (en) | Spot validity determination device of dna chip image, spot validity determination method of dna chip image, and spot validity determination program of dna chip image | |
| CN114937037B (en) | Product defect detection method, device and equipment and readable storage medium | |
| CN115690104A (en) | Wafer crack detection method and device and storage medium | |
| CN112414943B (en) | Method and module for locating defects of semiconductor chip | |
| CN107615333B (en) | Image speckle processing method | |
| CN109919909B (en) | Method and device for testing light-emitting diode chip | |
| CN111414909B (en) | Target detection method and device | |
| WO2008096211A4 (en) | Measurement of critical dimensions of semiconductor wafers | |
| CN103593667A (en) | Rapid image foreign matter identification method based on set connectivity principle | |
| US20090067722A1 (en) | Memory cell and page break 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 |