JPH0368806A - Defect extracting apparatus - Google Patents

Abstract

PURPOSE:To make it possible to extract a defect accurately even if there is difference in brightness of the surface of a body under inspection by obtaining differential image data wherein the difference in brightness at the central part and the peripheral part of the body under inspection is removed from the reversed image data and the original image data. CONSTITUTION:Light is projected from a lighting device 4. The image of a semiconductor wafer 2 is picked up with an image sensing device 3. The image signal (g) is converted into the digital signal. The signal is stored into an image memory 8 as image data. When the receiving of the image signal (g) is finished, the image data are reversed rightward and leftward with a main control part 6, and the reversed image data are obtained. The differential image data between the reversed image data and the original image data are obtained. In the differential data, the differences in brightness are offset. The image of a semiconductor wafer 2 has the uniform brightness. Defects remain as variable- density levels on the positive side and the negative side. The main control part 6 sends the differential image data to a CRT display 11 so as to display the defects.

Description

DETAILED DESCRIPTION OF THE INVENTION [Object of the Invention] (Industrial Application Field) The present invention relates to a defect extraction device for extracting defects in, for example, semiconductor wafers.

(Prior Art) Defect inspection of semiconductor wafers has been carried out visually by an operator, but due to variations in inspection results, it has been automated using computers and the like. In this case, defects are extracted by illuminating the semiconductor wafer, capturing an image with an imaging device, and extracting defects from the gray level of image data obtained by this imaging.

By the way, in such defect extraction, even if the film thickness of the semiconductor wafer is slightly different, the brightness and color may appear different due to the influence of light interference. In this case, differences in brightness did not cause any problems during visual inspection by an operator, but in automated inspection, differences in brightness affect defect extraction.

To explain this effect, in such inspections, for example, a bypass filter is most often used, and by passing an image signal through this bypass filter, only high frequency components are extracted and defective parts are extracted. However, bypass filters can extract signals that change rapidly, but cannot extract signals that change slowly, and therefore defect extraction is affected by differences in brightness.

(Problems to be Solved by the Invention) As described above, automated defect extraction cannot accurately extract defects due to the influence of differences in brightness on the semiconductor wafer.

SUMMARY OF THE INVENTION Therefore, an object of the present invention is to provide a defect extraction device that can accurately extract defects even if there are differences in brightness on the surface of an object to be inspected.

[Structure of the Invention] (Means for Solving the Problems) The present invention provides an imaging device for imaging an object to be inspected, an image inverting means for horizontally inverting an image of image data obtained by imaging with this imaging device, and A defect aiming at achieving the above object, comprising an extraction means for obtaining difference image data between the inverted image data obtained by the image reversing means and the original image data, and extracting defects in the object to be inspected from this difference image data. It is an extraction device.

(Function) By providing such a means, the image data of the object to be inspected obtained by imaging with the imaging device is horizontally reversed by the image reversing means, and extracted from this reversed image data and the original image data. The means obtains difference image data from which the difference in brightness between the central part and the peripheral part of the object to be inspected is removed, and extracts defects in the object to be inspected from this difference image data.

(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 extraction device. XY table 1
A semiconductor wafer 2 on which no chip pattern is formed is placed thereon. On the other hand, an imaging device 3 is arranged above the XY table 1 to take an image of the semiconductor wafer 2. Note that a lighting device 4 is arranged above the xY table 1 to illuminate the semiconductor wafer 2.

The image signal g output from the imaging device 3 is sent to an image processing device 5.

The image processing device 5 has a function of capturing the image signal g to generate image data and extracting defects in the semiconductor wafer 2 from this image data. Specifically, C
Main control unit 6 consisting of PU (central processing unit) etc.
The main control unit 6 is connected via a bus 7 to an image memory 8, a program memory 9, a data memory 10, and a CR.
T-display is connected. Program memory 9
A defect extraction program according to the defect extraction flowchart shown in FIG. 2 is stored in the memory. However, by executing this defect extraction program, the main control unit 6 uses an image reversing means for horizontally reversing the image of the image data, and the inspection object from the reversed image data obtained by this image reversing means and the original image data. It has the function of extracting means for obtaining difference image data by removing the difference in brightness between the center and peripheral parts of the semiconductor wafer 2 and extracting defects in the semiconductor wafer 2 from this difference image data. Note that an A/D (analog/digital) converter 12 is connected to the image memory 8 so that the image data signal g is digitally converted and input to the image memory 8.

Next, the operation of the apparatus configured as described above will be explained.

While the semiconductor wafer 2 is being irradiated with light by the illumination device 4, the imaging device 3 images the semiconductor wafer 2 and outputs the image signal g. At this time, the main control unit 6 performs step S3.
An image signal g is taken in at. That is, the image signal g
is converted into a digital image signal by the A/D converter 12, sent to the image memory 8, and stored in the image memory 8 as image data. FIG. 3 schematically shows such image data, in which a difference in brightness Q appears in the image 2a of the semiconductor wafer 2, and a defect G is present. Here, the brightness Q varies concentrically from the center to the periphery of the semiconductor wafer 2.

When the capture of the image signal g is completed, the main control section 6 moves to step S3 and inverts the image data horizontally to obtain the inverted image data shown in FIG. 4. Next, in step S4, the main control section 6 obtains difference image data between the original image data before inversion shown in FIG. 3 and the inverted image data shown in FIG.

FIG. 5 shows difference image data, in which the difference in brightness Q is canceled out, and the image 2a of the semiconductor wafer 2 has uniform brightness, and the defect G remains as a gray level on the positive side and negative m1.

Thus, the main control unit 6 extracts the defect G from the difference image data in step 7s5. In this case, the main controller 6 sends the difference image data to the CRT display 11 to display the defect G.

In this way, in the first embodiment, the semiconductor wafer 2
The image of the image data of is horizontally reversed to obtain the difference image data from the original image data, and the defect G on the semiconductor wafer 2 is extracted from this difference image data. Even if a difference in brightness Q occurs concentrically from the center to the periphery, the defect G can be reliably extracted by eliminating the difference in brightness Q. In this case, since the image data is horizontally inverted, it is possible to eliminate the difference in brightness Q for the entire image 2a of the semiconductor urchin 2, and to extract defects on the entire surface of the semiconductor urchin/%2. As a result, it is optimal for inspecting a thin resist film on a semiconductor substrate \2.

Next, a second embodiment of the present invention will be described. This embodiment is applied to defect extraction for a semiconductor wafer on which a periodic chip pattern is formed, and the image of each chip pattern in the difference image data and the original image data is sent to the apparatus of the first embodiment. It is equipped with a function to obtain difference image data when combined. That is, to explain using the apparatus shown in FIG. 1, the program memory 9 stores a defect inspection program having the defect inspection flowchart shown in FIG. 6 and having the above functions.

With such a configuration, the imaging device 3 images the semiconductor wafer 2 while the semiconductor wafer on which a chip pattern is formed is irradiated with light by the illumination device 4, and outputs the image signal g. This image signal g is converted into a digital image signal by the A/D converter 12, sent to the image memory 8, and stored as image data. FIG. 7 schematically shows such image data, and shows image 2 of a semiconductor wafer.
In 0a, there is a difference in brightness Q and a defect G is present. Here, the brightness Q varies concentrically from the center to the periphery of the semiconductor wafer.

Next, in step S3, the main control section 6 reverses the image data horizontally to obtain the reversed image data shown in FIG. Next, in step slO, the main control unit 6 obtains difference image data when the images of each chip pattern in the original image data shown in FIG. 7 and the inverted image data shown in FIG. 8 are matched. That is, the main control unit 6 sequentially obtains difference image data between the original image data and the inverted image data by one pixel at a time, and calculates the variance value of the gray level of the semiconductor wafer portion in these difference image data. Detect the minimum difference image data. In the difference image data detected in this way, as shown in FIG. 9, the chip pattern is canceled out, the brightness Q is canceled out, and the defect G
will remain.

Thus, the main control unit 6 extracts the defect G from the difference image data in step S5.

In this way, in the second embodiment, the original image data and the inverted image data are shifted one pixel at a time to obtain difference image data in which the chip pattern is eliminated. Even if there is a concentric difference in brightness Q on the surface of the wafer from the center to the periphery, it is possible to eliminate the difference in brightness Q and reliably extract the defect G. Even if a pattern is formed, only the defect G can be extracted without being affected by this chip pattern.

Note that the present invention is not limited to the above embodiments, and may be modified without departing from the spirit thereof. For example, the object to be inspected is not limited to semiconductor wafers, but can also be applied to inspections of CCDs (solid-state imaging devices) and TFTs (liquid crystal display panels).

[Effects of the Invention] As described in detail above, according to the present invention, it is possible to provide a defect extraction device that can accurately extract defects even if there is a difference in brightness on the surface of an object to be inspected.

[Brief explanation of drawings]

1 to 5 show a first diagram of a defect extraction 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 a defect extraction flowchart, Fig. 3 is a schematic diagram of semiconductor wafer image data, Fig. 4 is a schematic diagram of inverted image data, Fig. 5 is a schematic diagram of difference image data, and Figs. 6 to 9 6 is a flowchart of defect extraction, FIG. 7 is a schematic diagram of image data of a semiconductor wafer, and FIG. 8 is a schematic diagram of inverted image data. FIG. 9 is a schematic diagram of difference image data. 2... Semiconductor wafer, 3... Imaging device, 5... Image processing device, 6... Main control unit, 7... Bus, 8...
・Image memory, 9...Program memory, 10...
Data memory, 11...CRT display, 12.
...A/D converter.

Claims (1)

[Claims] An imaging device for imaging an object to be inspected, an image inversion device for horizontally inverting an image of image data obtained by imaging with the imaging device, and a difference between the inverted image data obtained by the image inversion device and the original image data. 1. A defect extraction device comprising: extraction means for obtaining difference image data and extracting defects in the object to be inspected from the difference image data.