JP2004085543A - System and method for visual inspection - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a visual inspection system being employed in the inspection process of a semiconductor manufacturing system for inspecting the micro electronic circuit pattern in a semiconductor chip formed on a wafer. <P>SOLUTION: A moving section mounts an article to be inspected having a pattern, an imaging section images the article to be inspected mounted on the moving section, and an operating means processes an image picked up at the imaging section. The operating means can extract the pattern of the article to be inspected from the image picked up at the imaging section and set an area based on the shape of the extracted pattern. <P>COPYRIGHT: (C)2004,JPO

Description

[0001]
[Industrial applications]
The present invention relates to a visual inspection apparatus used for an inspection process of a semiconductor manufacturing apparatus and inspecting a fine electronic circuit pattern on a semiconductor chip formed on a wafer.
[0002]
[Prior art]
Conventionally, when setting inspection conditions for a fine electronic circuit pattern formed on a wafer, as shown in FIG. 8, an image obtained by imaging a good electronic circuit pattern is displayed, and an operator uses an input device such as a mouse. Is used to set a rectangular measurement area within the inspection range, and input various conditions (for example, the size and type of foreign matter) as parameters.
[0003]
[Problems to be solved by the invention]
However, when specifying a simple shape or a pattern combining a relatively small number of rectangles, there is no particular problem, but a pattern having a complicated shape for securing the line width, or a vertical or horizontal When conditions need to be input for a large number of finely shaped patterns having the same shape, such as a pattern inclined with respect to a pattern and a constituent part of a semiconductor memory, there is a problem that an operator's burden becomes excessive. .
[0004]
Further, when the pattern to be inspected is formed obliquely with respect to the vertical and horizontal directions of the image, the measurement area must be set by a combination of small rectangles, and the setting operation is extremely complicated.
[0005]
If the boundary of the area occurs on the pattern in this way, if the foreign matter or abnormality is located on the boundary of the area, the size or state of the foreign matter or abnormality cannot be accurately detected, and the detection accuracy Could be reduced.
[0006]
Further, since the burden on the operator becomes excessive, input errors also increase, and there is a concern that the inspection accuracy may decrease.
[0007]
[Means for Solving the Problems]
The present invention has been devised in view of the above problems, a moving unit for mounting an inspection object having a pattern, and an imaging unit for imaging the inspection object mounted on the moving unit, The image processing unit includes an arithmetic processing unit for performing image processing and the like on the image captured by the imaging unit. The arithmetic processing unit extracts a pattern of the inspection object from the image captured by the imaging unit. It is characterized in that an area is set based on a pattern shape.
[0008]
Further, when the extracted pattern shape of the present invention is closed, the inside of the closed area may be set as an area. When the pattern extends to the outside of the image, the outer peripheral portion of the image can be recognized as a pattern boundary, and can be recognized as a closed pattern. The discrimination between the pattern portion and the non-pattern portion can be set by a change in contrast in an image processing process, visual setting by an operator, or the like.
[0009]
Further, according to the present invention, an inspection condition can be set for each area.
[0010]
The inspection conditions of the present invention can be line width, foreign matter size, threshold value for each pixel, sensitivity, defect area, number of defects, and the like.
[0011]
Further, the pattern of the inspection object of the present invention may be a fine electronic circuit pattern on a chip formed on a wafer.
[0012]
Further, the appearance inspection method of the present invention is an appearance inspection method for inspecting an inspection object having a pattern formed on a surface, wherein the imaging unit captures an image of the inspection object and the arithmetic processing unit captures an image of the inspection object by the imaging unit. The image processing is performed on the extracted image, and the arithmetic processing unit can extract a pattern of the inspection object from the image captured by the imaging unit, and set an area based on the extracted pattern shape.
[0013]
BEST MODE FOR CARRYING OUT THE INVENTION
In the present invention configured as described above, the moving unit mounts the inspection object having the pattern, the imaging unit captures an image of the inspection object mounted on the moving unit, and the arithmetic processing unit includes the imaging unit. The image processing device extracts the pattern of the inspection object from the image captured by the imaging section, and sets an area based on the extracted pattern shape. can do.
[0014]
Further, when the extracted pattern shape of the present invention is closed, the inside of the closed area can be set as the area.
[0015]
Further, according to the present invention, inspection conditions can be set for each area.
[0016]
The inspection conditions of the present invention can be line width, foreign matter size, threshold value for each pixel, sensitivity, defect area, number of defects, and the like.
[0017]
Further, the pattern of the inspection object of the present invention may be a fine electronic circuit pattern on a chip formed on a wafer.
[0018]
【Example】
[0019]
An embodiment of the present invention will be described with reference to the drawings.
[0020]
As shown in FIG. 1, the appearance inspection apparatus 10000 of the present embodiment includes an illumination unit 100, an imaging unit 200, a moving unit 300, an arithmetic processing unit 400, a monitor 500, and an input unit 600. I have.
[0021]
The appearance inspection device 10000 inspects the wafer 20000 shown in FIG.
[0022]
The illumination unit 100 is for illuminating the wafer placed on the moving unit 300.
[0023]
The imaging unit 200 is for capturing an image of a wafer placed on the moving unit 300, and includes a CCD sensor and an imaging optical system in this embodiment.
[0024]
The moving unit 300 is for moving the wafer placed on the stage 310, and can move in parallel in the X-Y direction or rotate by a desired angle. For these movements, the control unit 410 connected to the arithmetic processing unit 400 controls and drives the moving unit 300 via the driver 310.
[0025]
The arithmetic processing unit 400 performs image processing and the like and controls the appearance inspection apparatus 10000 and the like. The arithmetic processing unit 400 includes a CPU, and is connected to a storage unit 420 and the like.
[0026]
The monitor 500 can display and observe an image captured by the image capturing unit 200, a result of image processing performed by the arithmetic processing unit 400, and the like.
[0027]
The input unit 600 includes a keyboard, a mouse, and the like, and allows an operator to issue a command to the visual inspection apparatus 10000.
[0028]
Next, the operation of the visual inspection device 10000 of this embodiment will be described with reference to FIG.
[0029]
First, in step 1 (hereinafter abbreviated as S1), a fine electronic circuit pattern formed on a wafer to be inspected mounted on the moving unit 300 is selected from non-defective chips having no foreign matter or defects. , With the imaging unit 200. The image captured by the imaging unit 200 is, for example, as shown in FIG.
[0030]
Next, in S2, the arithmetic processing means 400 performs image processing such as shading as image processing (1) on the image of the good chip photographed in S1, and removes illumination unevenness and the like. Accordingly, it is possible to eliminate an optical influence at the time of shooting with a captured image and improve the accuracy of the next image processing.
[0031]
Further, in S3, the edge of the fine electronic circuit pattern of the chip formed on the inspection target wafer 20000 is detected on the inspection target wafer on the image data subjected to the image processing in S2, and the pattern is detected. The image processing (2) for extracting the shape is performed by the arithmetic processing means 400. For example, when the binarization processing is performed, an image as shown in FIG. 5 is obtained.
[0032]
Here, for example, the following processing can be considered to separate the fine electronic circuit pattern formed on the wafer from the background (substrate).
[0033]
(A) First, a differentiation process is performed on the image data, and a portion where the contrast changes most is defined as an edge portion. The image subjected to the differentiation processing is, for example, as shown in FIG.
[0034]
(B) Next, a change in the frequency distribution of the image data is obtained, and the edge portion is recognized from the change.
[0035]
(C) If the captured image is a color image, a change in hue is obtained, and an edge portion is recognized from the change.
[0036]
(D) When the captured image is a color image, a change in saturation is obtained, and an edge portion is recognized from the change.
[0037]
The processing is not limited to the above-described method, and any processing can be selected as long as the fine electronic circuit pattern formed on the wafer can be separated from the background (substrate).
[0038]
In S4, an area is set based on the result in S3. For example, the arithmetic processing unit 400 automatically sets the area based on the fine electronic circuit pattern shape formed on the wafer detected in S3.
[0039]
Here, when the fine electronic circuit pattern formed on the wafer is closed, it is set inside the closed area. That is, the outline of the fine electronic circuit pattern formed on the wafer is detected, and when the outline is closed, it is recognized as an area. An image in which the area is automatically set is, for example, as shown in FIG.
[0040]
On the other hand, in the same pattern, there are many cases where the line width and the surface state are different depending on the portion. In this case, the operator sets and edits an area using the input unit 600 while referring to the area setting image displayed on the monitor 500. In this case, the area is divided by setting the boundaries of the areas having different detection conditions. In addition, areas can be connected to each other.
[0041]
On the other hand, if there is recognition at the time of detecting the edge of the pattern in the process of S3, this can be corrected.
[0042]
Next, in S5, inspection conditions are set for each area. The inspection condition is a condition of a foreign substance and a defect to be detected, and is set with respect to a predetermined size, arrangement, and the like in the area. For example, in the same pattern, there are many cases where the line width and the surface state are different depending on the portion. As the inspection conditions, those set in stages with respect to the state of the inspection target are applied. For example, the step size is set stepwise according to the size of a foreign substance with respect to the line width of a fine electronic circuit pattern formed on a wafer.
[0043]
An example will be given with respect to the above-described inspection target conditions.
[0044]
(A) Brightness
The size of a foreign substance or the like determined to be defective is set for each brightness level.
[0046]
(B) Shape
Depending on the shape of the pattern, different judgments may be made as to whether or not a foreign substance having the same size and shape is determined to be defective. For example, while FIG. 9A is defective, FIG. 9B is determined to be good.
[0048]
(C) Frequency
The surface state is determined as a pattern, and good or bad is determined according to the pattern.
[0050]
These settings can be configured, for example, to send data from the input unit 600 to the arithmetic processing unit 400.
[0051]
Since the inspection conditions can be set according to the characteristics of the pattern in each area, there is an excellent effect that a stable inspection can be performed.
[0052]
Further, it is also possible to select the setting of the inspection method according to the frequency characteristics in the area.
[0053]
Then, in S6, pattern inspection is performed by setting the set area information as a template. The parameters set in each area include, for example, a threshold value serving as a lower limit of the size of a foreign substance to be detected, a number in quality judgment, a threshold value for each pixel, sensitivity, a size of a defect or a foreign substance, a defect area, a surface state, and the like. is there. These settings can be configured, for example, to send data from the input unit 600 to the arithmetic processing unit 400.
[0054]
Note that the wafer 20000 corresponds to an inspection object having a pattern.
[0055]
Alternatively, a boundary may be provided inside the area so that the area can be set as a different area.
[0056]
Further, when the pattern extends to the outside of the image, the outer peripheral portion of the image can be recognized as a pattern boundary, and can be recognized as a closed pattern. The discrimination between the pattern portion and the non-pattern portion can also be set by a change in contrast in the image processing process, visual setting by an operator, or the like.
[0057]
As described above, the present embodiment has an effect that the trouble of inputting the area setting by the operator can be reduced, and the input error of the condition setting can be prevented.
[0058]
Next, a procedure for creating (teaching) the inspection template will be described with reference to FIG.
[0059]
First, in S1, a fine electronic circuit pattern formed on the wafer 20000 is photographed by the imaging unit 200 by selecting non-defective chips having no foreign matter or defects. Next, in S2, the position of the captured image is adjusted. That is, the operator visually checks the chip photographed image on the monitor 500 for the presence or absence of a defect and selects a non-defective non-defective image.
[0060]
Next, in S3, processing before area setting is performed. S3 is composed of S31 and S32.
[0061]
In S31, processes such as multi-value conversion, gradation conversion, chromaticity conversion, expansion / contraction processing, and the like are performed. Here, in the multi-value processing, the edge of the pattern formed on the chip can be clarified by performing the processing of multi-value processing (for example, binarization) on the good-quality image. In the gradation conversion, by performing gradation conversion filter processing, the influence of the density of the pattern and the density of the image (contrast) can be reduced, and the pattern can be easily recognized. Further, in the chromaticity conversion, an image is processed by changing to a color tone that facilitates pattern recognition according to the color tone of a good-quality image. The expansion / shrinkage processing and the like are the same as those described previously, and thus description thereof will be omitted.
[0062]
In S32, the arithmetic processing means 400 performs a differentiation process on the image data obtained in S31, and sets a portion where the contrast changes most as an edge portion and executes recognition of each pattern.
[0063]
Then, in S4, area setting is performed on the recognized pattern of the non-defective image. Further, in S5, an inspection condition is set. Since the inspection condition at the time of the inspection in S5 is set for each set area, it is set according to the density and state of the pattern on the wafer and the request of the operator.
[0064]
The inspection condition in S5 is a condition for recognizing a defect or a defective product, for example, a threshold value for each pixel, sensitivity, defect size, defect area, the number of defects, and the like.
[0065]
Next, in S6, the layer defined in S5 is stored. These settings include the above-described area setting pre-processing, area setting, area-specific inspection conditions, and the like. These settings can be configured, for example, to send data from the input unit 600 to the arithmetic processing unit 400.
[0066]
Thus, the teaching step is completed.
[0067]
Next, a case where the inspection process is performed based on the teaching setting will be described.
[0068]
The compatibility between the chip to be inspected and the set template is confirmed. If the chip to be inspected is different from the set template, a suitable template is searched for or a teaching step is performed to create a template.
[0069]
In the inspection process, an inspection is started in S10, and after an inspection target is recognized in S11, an inspection target (chip) image is photographed in S12. Then, the position of the photographed image is adjusted in S13.
[0070]
Next, in S14, the procedure of the preprocessing performed in the teaching stored in S6 is read from the chip image to be inspected captured in S12, and the same processing is performed. The reading of the procedure of the preprocessing is not limited to the described timing, and may be already read or set at the time of starting the inspection. Further, in S14, similarly, the stored area setting and inspection condition are read.
[0071]
In S15, inspection is performed under the same conditions, and in S16, a non-defective or defective chip to be inspected is determined. Further, in S17, the judgment of good or defective chip is stored in association with the chip to be inspected.
[0072]
In S18, it is determined whether or not the inspection target has been completed, and if completed, the process proceeds to S19,
End the inspection. If it is determined in S18 that the inspection target has not been completed, the process returns to S11. That is, if there is another chip to be inspected, the inspection process is repeated again, and if there is no other chip to be inspected, the inspection process is terminated.
[0073]
In addition, it is also possible to output the result of the inspection process arbitrarily stored as a report.
[0074]
Other configurations, operations, and the like are the same as those of the previous embodiment, and thus description thereof is omitted.
[0075]
【effect】
The present invention configured as described above provides a moving unit for mounting an inspection object having a pattern, an imaging unit for imaging the inspection object mounted on the moving unit, and an imaging unit And an arithmetic processing unit for performing image processing and the like on the image captured by the imaging unit. The arithmetic processing unit extracts a pattern of the inspection object from the image captured by the imaging unit, and converts the pattern into the extracted pattern shape. Since the area is set on the basis of this, there is an excellent effect that the burden on the operator does not become excessive, input errors are reduced, and the inspection accuracy is improved by setting an area having an appropriate shape.
[0076]
[Brief description of the drawings]
FIG. 1 is a diagram illustrating a configuration of a visual inspection apparatus 10000 according to an embodiment of the present invention.
FIG. 2 is a diagram illustrating a wafer.
FIG. 3 is a diagram for explaining the operation of the visual inspection device 1000 of the present embodiment.
FIG. 4 is a diagram illustrating an image captured by an imaging unit.
FIG. 5 is a diagram illustrating an image that has been subjected to a binarization process.
FIG. 6 is a diagram illustrating an image on which a differentiation process has been performed.
FIG. 7 is a diagram illustrating area setting according to the present embodiment.
FIG. 8 is a diagram illustrating a conventional technique.
FIG. 9A is a view for explaining an image regarded as a defective product.
FIG. 9 (b) is a diagram illustrating an image regarded as a non-defective product.
FIG. 10 is a diagram illustrating a procedure such as teaching.
[Explanation of symbols]
10,000 Appearance inspection device 20000 Wafer 100 Illumination unit 200 Imaging unit 300 Moving unit 310 Driver 400 Arithmetic processing unit 500 Monitor 600 Input unit

Claims (9)

A moving unit for mounting an inspection object having a pattern, an imaging unit for capturing an image of the inspection object mounted on the moving unit, and an image processing unit for processing an image captured by the imaging unit; The arithmetic processing means extracts a pattern of the inspection object from the image captured by the imaging unit, and sets an area based on the extracted pattern shape. Appearance inspection device. The visual inspection device according to claim 1, wherein when the extracted pattern shape is closed, the inside of the closed area is set as an area. 3. The visual inspection device according to claim 1, wherein inspection conditions are set for each area. 4. The visual inspection apparatus according to claim 3, wherein the inspection conditions include a line width, a foreign matter size, and the like. The visual inspection apparatus according to claim 1, wherein the pattern of the inspection object is a fine electronic circuit pattern formed on a wafer. 4. The visual inspection apparatus according to claim 3, wherein the inspection conditions include a threshold value for each pixel, sensitivity, a defect area, and the number of defects. An appearance inspection method for inspecting an inspection object having a pattern formed on a surface, wherein an imaging unit captures an image of the inspection object, and an arithmetic processing unit performs image processing of an image captured by the imaging unit. A visual inspection method in which an arithmetic processing unit extracts a pattern of an inspection object from an image captured by an imaging unit and sets an area based on the extracted pattern shape. 8. The appearance inspection method according to claim 7, wherein when the extracted pattern shape is closed, the inside of the closed area is set as an area. The appearance inspection method according to claim 7, wherein a different area can be set by providing a boundary inside the area.

Images