JPH02253109A - Defect discriminating device - Google Patents

Abstract

PURPOSE:To discriminate a defect of an object to be inspected having a periodical pattern with a high precision in a short time by dividing difference image data into small areas and obtaining the average value of brightness of each small area and deciding the defect in accordance with the difference between maximum and minimum values of average values. CONSTITUTION:The image of an object 10 to be inspected like a semiconductor wafer having a periodical pattern is picked up by an image pickup device 11 and is converted by an A/D converter 19 to store a digital image in a memory 14. In an image difference arithmetic part 15, this image data is shifted by integer-fold pattern period of the object 10 to be inspected to obtain difference image data having the difference of brightness between original image data and shifted image. An area dividing part 17 divides the difference image into respective small areas, and a defect discriminating part 16 obtains the average value of brightness of each divided small area and discriminates a defect in accordance with the difference between maximum and minimum values of these average values. Thus, the defect is quickly and precisely discriminated because of the simple processing.

Description

DETAILED DESCRIPTION OF THE INVENTION [Object of the Invention] (Industrial Application Field) The present invention relates to a defect determination apparatus for determining defects in an inspected object having a periodic pattern, such as a semiconductor wafer.

(Prior art) Defect inspection of objects with periodic patterns such as semiconductor wafers was initially carried out visually by an operator, or as inspection automation has progressed in recent years, inspections have been carried out by computer processing. became. Therefore, such an inspection apparatus uses FFT (fast Fourier transform) processing, or as shown in FIG. There is a technique for comparing images taken by the microscopes 1 and 2 in the processing device 4.

However, FFT processing is for removing periodic components, and may even process defective parts. The FFT process removes periodic components to facilitate the detection of aperiodic components, and in order to identify defects, it is necessary to add a separate defect determination device. Furthermore, when classifying inspected objects with defects and inspected objects without defects, it is necessary to add a device for this classification. Another problem is that the FFT processing device is expensive and its processing takes time.

Furthermore, in the technique of comparing two images, binarization processing is required in order to perform this comparison at high speed. However, in the case of binarization processing, it is difficult to identify defects unless the object to be inspected is a bright black and white semiconductor wafer 3. Moreover, when two images are compared to determine the difference between them, the sensitivity becomes low for small areas. or,
Since the semiconductor wafer 3'' is scanned by two microscopes, the driving mechanism becomes complicated and the scanning takes time.

(Problems to be Solved by the Invention) As described above, in FFT processing, there is a possibility that defective portions are also processed, and defects cannot be determined with high accuracy by comparing two images.

SUMMARY OF THE INVENTION Therefore, an object of the present invention is to provide a defect determination device that can identify defects in an object to be inspected having a periodic pattern in a short time and with high accuracy.

[Structure of the Invention (Means for Solving the Problems) The present invention provides a defect discrimination device that discriminates defects of an object to be inspected having a periodic pattern. an imaging means for obtaining data; an image difference calculation means for shifting the image data obtained by the imaging means by an integral multiple of the period of the pattern and calculating a difference in brightness from the original image data; A region dividing means divides the obtained difference image data into each small region according to a pattern, and each average of brightness in each small region divided by the region dividing means is calculated and the maximum and minimum values of these average values are calculated. The present invention is a defect discriminating device that attempts to achieve the above object by including a defect discriminating means for discriminating defects based on differences.

The present invention also provides a defect discrimination device for determining defects of an inspected object having a periodic pattern, including: an imaging means for imaging an inspected object to obtain image data of the inspected object; image difference calculation means for calculating the difference in brightness from the original image data by shifting the image data by an integral multiple of the period of the pattern; The above-mentioned object is provided with a region dividing means for dividing the region, and a defect determining means for determining the difference between the maximum value and the minimum value of brightness in each small region divided by the region dividing means and determining the defect based on the value of this difference. This is a defect discrimination device that aims to achieve the following.

Furthermore, the present invention provides a defect discrimination device for determining defects of an inspected object having a periodic pattern, including an imaging means for capturing an image of the inspected object to obtain image data of the inspected object, and an imaging means for obtaining image data of the inspected object. image difference calculation means for calculating the difference in brightness from the original image data by shifting the image data by an integral multiple of the period of the pattern; The apparatus includes an area dividing means for dividing the area, and a defect determining means for calculating the standard deviation of brightness in each small area divided by the area dividing means and determining the defect from the difference between the maximum value and the minimum value of these standard deviations. This is a defect discriminating device that attempts to achieve the above objectives.

Further, the present invention provides a defect discrimination device for determining defects of an inspected object having a periodic pattern, including an imaging means for capturing an image of the inspected object to obtain image data of the inspected object, and an imaging means for obtaining image data of the inspected object. image difference calculation means for calculating the difference in brightness from the original image data by shifting the image data by an integral multiple of the period of the pattern; The standard deviation of each small area divided by this area dividing means is calculated, and the difference between the maximum and minimum values of the standard deviation values for each small area is calculated. The above object is achieved by providing a defect determination means for determining a second standard deviation for the entire inspection area from a group of standard deviation values, and determining defects based on the value of the quotient of this difference and the second standard deviation. This is a defect discrimination device.

(Function) By having such a means, the image data of the object to be inspected obtained by the imaging means is shifted by an integral multiple of the period of the pattern by the image difference calculation means, and the difference in brightness from the original image data is calculated. The difference image data obtained by the image difference calculating means is divided into small regions according to the pattern by the region dividing means, each average value of the brightness in these small regions is found by the defect discriminating means, and Defects can be determined from the difference between the maximum and minimum values.

Furthermore, by providing the above means, the difference in brightness of each small area divided by the above area dividing means can be determined by the defect determining means, and the maximum standard deviation in the difference image data determined by the above image difference calculating means. The difference between the value and the minimum value is determined, and defect determination is performed based on these difference values.

Furthermore, by providing the above means, the standard deviation of brightness in each small area divided by the above area dividing means is determined by the defect determining means, and the defect is determined from the difference between the maximum value and the minimum value of these standard deviations. A determination is made.

Furthermore, by providing the above means, the standard deviation of the brightness of each small region divided by the above region dividing means is determined;
The difference between the maximum and minimum standard deviations for each small area and the second standard deviation for the entire inspection area are calculated from the group of standard deviation values for each small area, and this difference and the second standard deviation are calculated. Defect discrimination is performed with higher sensitivity for defects in small areas than the value of the quotient of .

(Example) Hereinafter, a first example of the present invention will be described with reference to the drawings.

FIG. 1 is a block diagram of a defect discrimination device. In the same figure, 1
0 is an object to be inspected, and a periodic pattern is formed on this object 10 as shown in FIG. In addition,
The object to be inspected 10 shown in FIG. 2 has patterns "■" arranged periodically, and each of 10a and 10b is defective. An imaging device] 1 is placed above the object to be inspected 10 to take an image of the object to be inspected. The image signal output from this imaging device]1 is sent to an image processing device 12.

This image processing device 12 has a function of determining defects in the object to be inspected]0 from image data obtained by imaging with the imaging device 11, and has the following configuration. That is, a main control unit consisting of a CPU (central processing unit), etc.]3
The main control unit]3 includes an image memory 14, an image difference calculation unit 15, a defect determination unit]6, an area dividing unit 17, and a
A display device 1-8 such as RT is connected. The image memory 14 has an A/D (analog/digital) converter 19
is connected, and the image signal from the imaging device 11 is converted into a digital image signal by the A/D converter and stored in the image memory 1.
4 is stored as image data. On the other hand, the image difference calculation unit 15 has a function of shifting the image data stored in the image memory 4 by an integral multiple of the period of the pattern of the object to be inspected 10 and calculating the difference in brightness from the original image data. ] It is. The area dividing unit 17 has a function of dividing the difference image data obtained by the image difference calculation unit 1-5 into each small area according to the pattern of the inspected object. The unit 17 has a function of determining each average value of brightness in each divided small area and determining defects from the difference between the maximum value and the minimum value of these average values.

Next, the operation of the apparatus configured as described above will be explained. When the inspection object 10 with defects 10a10b as shown in FIG. 2 is placed below the imaging device 11, the imaging device 1
1 takes an image of the object to be inspected 10 and outputs the image signal.

This image signal +;i A/D converter converts it into a digital image signal and stores it in the image memory 14 as image data. When the image data is stored in the image memory 14 in this way, the image difference calculation section 15 stores the image data in the image memory 14.
The image data stored in 4 is read out and this image data G1 is used as one of the patterns of the object to be inspected 10 as shown in FIG.
Image data G2 is obtained by shifting by the period. Then, the image difference calculation unit 15 calculates the difference between these image data G1 and 62 to obtain difference image data G3. In other words, by finding the difference between each image data G1 and G2, the pattern "■" with no defects is canceled out, and the pattern with defects 1. , Oa, 10b are parts with high density levels], Oa-1, 1, Ob -1
and low density level], Oa-2,1-Ob2
As a result, the defective parts are emphasized.

Next, the image difference calculation section]5 obtains difference image data G3 as shown in FIG. When the difference image data G3 is obtained in this way, the region dividing unit 17 divides the difference image data G3 into each small region (+1. q2...qn) according to the periodic pattern of the inspected object 10 as shown in the figure. Next, the defect discriminating unit]6 calculates the average brightness value for each small area Q+, (+2...qn).Then, the defect discriminating unit 1
Step 6 determines the maximum value and minimum value from among the average values of each small region QL, q2, and qn, and determines the difference between these maximum and minimum values. However, the defect determining section KORO determines whether the inspected object 10 is a good product or a defective product based on the value of the difference between the maximum value and the minimum value.

In this case, if there is a defect in the object to be inspected]0, then The value of the difference between the maximum value and the minimum value becomes large.

In this way, in the first embodiment, the difference image data G
3 is divided into small areas Ql-+ q2...qll, each average value of brightness in these small areas ql, q2...qn is determined, and defects are determined from the difference between the maximum and minimum values. Since this is done, defects in the object to be inspected 10 can be determined with high accuracy through simple processing. Moreover, since it is a simple process, defects in the inspected object 10 can be determined at high speed.

Next, a second embodiment of the present invention will be described. Here, the second embodiment will be described with reference to FIG. 1 above, but the difference from the first embodiment is the image processing device]2. That is, the image processing device 12 is connected to the area dividing section 17 and the defect determining section 16 or the main control section 13 . Area division section]7
has a function of dividing the difference image data obtained by the image difference calculating section 15 into each small region according to the pattern of the inspected object 10, and the defect discriminating section 16 has a function of dividing the difference image data obtained by the image difference calculating section 15 into each small region divided by the region dividing section 17. It has a function of determining the difference between the maximum and minimum brightness values in a region and determining defects based on this difference.

Next, the effect will be explained. As explained in the above embodiment, the difference image data G3 as shown in FIG. 4 is obtained by the image difference calculating section 15. When the difference image data G3 is obtained in this manner, the region dividing unit 17 divides the difference image data G3 into small regions q1, q2, . Next, the defect determination unit 16 identifies each small area ql, Q2. ...The maximum value and minimum value of brightness are determined for each qn, and the difference between the maximum value and the minimum value - is determined. However, the defect determination unit 16 determines whether the inspected object 10 is good or not based on the value of the difference between the maximum value and the minimum value.
Determines defects in defective products. In this case, if the inspection object 10 has a defect, the value of the difference between the maximum value and the minimum value increases.

In this way, in the second embodiment, the difference image data G
3 is divided into small areas ql, q2...qn, and defects are determined based on the difference between the maximum and minimum brightness values in these small areas ql, q2...qn.
This embodiment can also provide the same effects as the first embodiment.

Next, a third embodiment of the present invention will be described with reference to a block diagram of a defect discriminating apparatus shown in FIG. Note that the same parts as in FIG. 1 are given the same reference numerals, and detailed explanation thereof will be omitted.

The image processing device 30 is equipped with a defect determination section 31. This defect determination unit 31 determines the standard deviation of brightness in each of the small areas ql, q2...qn divided by the area dividing unit 17, and performs defect determination based on the difference between the maximum value and the minimum value of these standard deviations. It has a function.

Next, the effect will be explained. As in the second embodiment, the difference image data G3 is obtained by the image difference calculating section 15, and then the difference image data G3 is divided into each small region ql, q2, . . . by the region dividing section 17 as shown in FIG. It is divided into qn. Here, the defect determination unit 31 identifies each small area ql, q2...
- Find the standard deviation value of brightness for each qn. 6th
Figures (a) and (b) show the frequency distribution of such standard deviation values. This is obtained for the inspected object 10 that has a defect. In addition, el,
e2 indicates the standard deviation value of each small area having a defect. Next, the defect determination section 31 determines the difference between the maximum value and the minimum value of the standard deviation value for each small region determined as described above, and performs defect determination based on the value of this difference.

In this way, in the third embodiment, each small region ql,
The standard deviation of brightness at q2...qn is determined, and defects are determined based on the difference between the maximum and minimum values of these standard deviations, so it is possible to achieve the same effect as in the first embodiment. can.

Next, a fourth embodiment of the present invention will be described with reference to a block diagram of a defect discriminating apparatus shown in FIG. Note that the same parts as in FIG. 1 are given the same reference numerals, and detailed explanation thereof will be omitted.

The image processing device 30 is equipped with a defect determination section 31. This defect determination unit 31 determines the standard deviation of brightness in each of the small areas ql, q2...qn divided by the area dividing unit 17, and performs defect determination based on the difference between the maximum value and the minimum value of these standard deviations. It has a function.

Next, the effect will be explained. As in the second embodiment, the difference image data G3 is obtained by the image difference calculating section 15, and then the difference image data G3 is divided into each small region ql, q2, . . . by the region dividing section 17 as shown in FIG. It is divided into qn. Here, the defect determination unit 31 identifies each small area ql, q2...
- Find the standard deviation value of brightness for each qn. 6th
Figures (a) and (b) show the frequency distribution of such standard deviation values. This is obtained for the inspected object 10 that has a defect. In addition, cl,
c2 indicates the standard deviation value of each small area having a defect. Next, the defect determination unit 31 determines the difference A (A2+ in FIG. 6(b)) between the maximum value and minimum value of the distribution of standard deviation values for each small region obtained as described above. The second standard deviation value σ1 (σ2 in Fig. 6(b)) for the entire image is calculated from the distribution of standard deviation values for each area.In other words, the standard deviation σ1 increases as the defective area increases. Then, the defect determination unit 3]
The difference A is divided by the standard deviation σ1, and defects are determined from this divided value. In addition, in FIG. 6(b), defects are determined based on the value of A2/σ2. In this case, the smaller the size of the defective area, the larger the division value becomes. By the way, when the division value An/σn for the above-mentioned difference An is represented in a diagram, it is as shown in FIG. Note that n is 1,2°3... In the figure, an area α0 larger than the value α for the difference An indicates a small area with a large defect, and an area β0 larger than the value β for the division value An/σn indicates a small area with a small defect. ing. Therefore, the defect discriminating unit 31 uses each of the above values α
.

Large defects and small defects are discriminated using β as the boundary.

In this manner, in the fourth embodiment, the types of defects such as large defects and small defects can be determined.

Note that the present invention is not limited to the above embodiments, and may be modified without departing from the spirit thereof. For example, although the image difference calculation unit 15 shifts the periodic pattern by one period, it is not limited to one period, but may shift by an integral multiple of the period. Furthermore, the patterns "■" may be shifted not only vertically and horizontally but also diagonally so that the patterns "■" overlap.

[Effects of the Invention] As described in detail above, according to the present invention, it is possible to provide a defect discrimination device that can discriminate defects of an object to be inspected having periodic turns in a short time and with high precision.

[Brief explanation of drawings]

1 to 3 show a first diagram of a defect discriminating device according to the present invention.
FIG. 1 is a diagram for explaining an embodiment, and FIG. 1 is a configuration diagram;
FIG. 2 is an external view of the object to be inspected, FIG. 3 is a schematic diagram of difference image data, and FIG. 4 is a schematic diagram showing each small area of difference image data.
FIG. 5 is a configuration diagram of the third embodiment of the present invention, FIG. 6 is a distribution diagram of standard deviation values, FIG. 7 is a diagram for explaining the operation of defect discrimination, and FIG. 4 Example (h diagram, FIG. 9 is a diagram for explaining the prior art.) 0. Inspection object, 11. Imaging device, 1.2.3040
...Image processing device, 13...Main control section, 14...
Image memory, 15... Image difference calculation unit, 1.6, 31.
. 41... Defect determination section, 17... Area division section, 18.
...Display device, 19...A/D converter. Figure 4 Figure 5 (a) (b) Figure 6 Figure

Claims (4)

[Claims] (1) In a defect discrimination device that discriminates defects of an object to be inspected having a periodic pattern, an imaging means for capturing an image of the object to be inspected to obtain image data of the object to be inspected; image difference calculation means for calculating a difference in brightness from the original image data by shifting the image data by an integral multiple of the cycle of the pattern; and a defect determining means that calculates each average of brightness in each small area divided by the area dividing means and performs defect determination based on the difference between the maximum value and the minimum value of these average values. A defect determination device characterized by: (2) In a defect discrimination device that discriminates defects of an object to be inspected having a periodic pattern, an imaging means for capturing an image of the object to be inspected to obtain image data of the object to be inspected; image difference calculation means for calculating a difference in brightness from the original image data by shifting the image data by an integral multiple of the cycle of the pattern; and a defect determining means for determining the difference between the maximum value and the minimum value of brightness in each small area divided by the area dividing means and determining the defect based on the value of this difference. A defect discrimination device characterized by: (3) In a defect discrimination device that discriminates defects of an object to be inspected having a periodic pattern, an imaging means for capturing an image of the object to be inspected to obtain image data of the object to be inspected; image difference calculation means for calculating a difference in brightness from the original image data by shifting the image data by an integral multiple of the cycle of the pattern; and a defect determining means that determines the standard deviation of brightness in each small area divided by the area dividing means and determines the defect based on the difference between the maximum value and the minimum value of these standard deviations. A defect determination device characterized by comprising: (4) In a defect discrimination device that discriminates defects of an object to be inspected having a periodic pattern, an imaging means for capturing an image of the object to be inspected to obtain image data of the object to be inspected; image difference calculation means for calculating a difference in brightness from the original image data by shifting the image data by an integral multiple of the cycle of the pattern; The standard deviation of each small area divided by this area dividing means is calculated, and the difference between the maximum value and the minimum value of the blood group of the standard deviation for each of these small areas, and the standard deviation of each small area divided by this area dividing means are calculated. A second standard deviation for the entire inspection area is determined from a group of standard deviation values, and a defect determination means is provided for determining defects based on the value of the quotient of this difference and the second standard deviation. Defect identification device.