JP7775688B2 - Analysis device, analysis method, and program - Google Patents

Description

The present invention relates to an analysis device, an analysis method, and a program.

In an "inspection device for detecting concave and/or convex defects on a mold surface," a technology for "capturing an image of a partial area of the mold surface" is known (see Patent Document 1, claim 1).
Patent Document 1: Japanese Patent Application Laid-Open No. 2011-47681

Semiconductor devices may also be inspected using images. It is desirable to inspect semiconductor devices with high accuracy.

A first aspect of the present invention provides an analysis device. The analysis device may include an image acquisition unit that acquires a photographed image of a semiconductor device. The analysis device may include a defect information acquisition unit that acquires defect information based on defect portions contained in the image. The analysis device may include a brightness information acquisition unit that acquires brightness information of at least a partial region of the image. The analysis device may include an output unit that outputs the defect information and brightness information in association with each other.

The defect information acquisition unit may acquire defect information including classification results obtained by classifying each defect portion by defect mode. The output unit may display the defect information and brightness information together.

The brightness information acquisition unit may acquire brightness information for areas other than defective areas.

The brightness information acquisition unit may acquire brightness information for the defective area.

The defect information acquisition unit may acquire defect information including classification results obtained by classifying each defect portion by defect mode. The brightness information acquisition unit may acquire brightness information of the defect portion area for each defect mode.

The brightness information acquisition unit may acquire brightness information according to the comparison result between the first brightness and the second brightness in the image of the semiconductor device being analyzed.

The brightness information acquisition unit may acquire the brightness in an image of a semiconductor device other than the one being analyzed as the second brightness.

The brightness information acquisition unit may acquire the variation between the first brightness and the second brightness for multiple semiconductor devices from different lots.

The brightness information acquisition unit may determine whether the amount of variation of the first brightness relative to the second brightness is equal to or greater than a reference value, and generate brightness information including the determination result.

The defect information acquisition unit may correct the image based on the brightness information and acquire defect information based on the corrected image.

The analysis device may include a light source control unit that adjusts the brightness of a light source that irradiates light onto the semiconductor device based on the brightness information. The image acquisition unit may acquire an adjusted image captured using the light source with the adjusted brightness. The defect information acquisition unit may acquire defect information based on defective portions included in the adjusted image.

The analysis device may include an instruction generation unit that generates instructions for the user based on the brightness information.

A second aspect of the present invention provides an analysis method. The analysis method may include an image acquisition step of acquiring an image of a semiconductor device. The analysis method may include an information acquisition step of acquiring defect information based on defect portions contained in the image and brightness information of at least a partial region of the image. The analysis method may include an output step of outputting the defect information and brightness information in association with each other.

A third aspect of the present invention provides a program for causing a computer to execute the analysis method according to the second aspect.

Note that the above summary of the invention does not list all of the features of the present invention. Subcombinations of these features may also constitute inventions.

FIG. 1 is a diagram showing an example of the configuration of an inspection system 300 according to an embodiment of the present invention. 2 is a diagram showing an example of an image 201 of the semiconductor device 10 acquired by the image acquisition unit 110. FIG. FIG. 10 is a diagram illustrating an example of a method for acquiring brightness information. 10A and 10B are diagrams showing examples of defect information 142 and brightness information 144 displayed by an output unit 140. FIG. 1 is a diagram showing the number of defects contained in the semiconductor device 10 for each manufacturing lot. FIG. 10 is a diagram showing the brightness of an image of the semiconductor device 10 for each manufacturing lot. FIG. 10 is a diagram illustrating an example of an allowable range for brightness. FIG. 10 is a diagram illustrating another example of the allowable range for brightness. FIG. 10 is a diagram illustrating another example of the allowable range for brightness. FIG. 10 is a diagram illustrating another example of the allowable range for brightness. FIG. 10 is a diagram illustrating another example of the analysis device 100. FIG. 10 is a diagram illustrating another example of the analysis device 100. FIG. 10 is a diagram illustrating another example of the analysis device 100. 10A and 10B are diagrams illustrating another example of a method for acquiring brightness information. 10 is a diagram showing an example of the distribution of brightness information in a brightness acquisition area 240 including a defective portion 210. FIG. FIG. 10 is a diagram showing an example of classification of a defect portion 210. FIG. 10 is a diagram illustrating an example of fluctuations in brightness information. FIG. 1 is a diagram illustrating an example of an analysis method using the analysis device 100. 22 illustrates an example computer 2200 in which aspects of the present invention may be embodied, in whole or in part.

The present invention will be described below through embodiments of the invention, but the following embodiments do not limit the scope of the invention as claimed. Furthermore, not all of the combinations of features described in the embodiments are necessarily essential to the solution of the invention.

Figure 1 is a diagram showing an example configuration of an inspection system 300 according to one embodiment of the present invention. The inspection system 300 detects defects present in the semiconductor device 10 based on an image of the semiconductor device 10. The semiconductor device 10 may be a semiconductor wafer or a semiconductor chip. The semiconductor wafer or semiconductor chip is formed from a semiconductor material such as silicon or a compound semiconductor.

The inspection system 300 includes an imaging device 20 and an analysis device 100. The imaging device 20 captures an image of the surface of the semiconductor device 10. For example, the imaging device 20 captures an image of the surface of a semiconductor wafer on which multiple semiconductor chips are formed. The imaging device 20 may capture a single image of the entire surface of the semiconductor device 10, or may capture multiple images by changing the imaging position on the surface of the semiconductor device 10.

The analysis device 100 extracts defects present in the semiconductor device 10 based on the images captured by the imaging device 20. The analysis device 100 in this example includes an image acquisition unit 110, a defect information acquisition unit 120, a brightness information acquisition unit 130, and an output unit 140. The analysis device 100 in this example is a computer that processes information. The computer may be provided with a program for causing the computer to operate as the image acquisition unit 110, the defect information acquisition unit 120, the brightness information acquisition unit 130, and the output unit 140.

The image acquisition unit 110 acquires an image of the surface of the semiconductor device 10 captured by the imaging device 20. The image acquisition unit 110 may communicate with the imaging device 20 via a wired or wireless connection. The image acquisition unit 110 may acquire images from the imaging device 20 via a network such as the Internet or a local network. The analysis device 100 and the imaging device 20 may also be located in the same housing.

The defect information acquisition unit 120 acquires defect information based on defect portions contained in the image acquired by the image acquisition unit 110. In this example, the defect information acquisition unit 120 extracts multiple defect portions from the image acquired by the image acquisition unit 110. The defect information acquisition unit 120 may extract defect portions by extracting areas in the image where the pixel brightness or color tone is different from other areas. The defect information acquisition unit 120 may extract defect portions based on preset rules, or may extract defect portions using an extraction module trained by machine learning. A known method can be used to extract defect portions from an image. The defect information acquisition unit 120 may perform a predetermined process on the image before extracting defect portions from the image. For example, the defect information acquisition unit 120 may perform at least one of a process to normalize the brightness of each pixel in the image and a process to enhance the contrast in the image.

The defect information may include, for example, information indicating at least one of the position, size, and shape of the defective portion in the semiconductor device 10. The defect information may also include classification results obtained by classifying the defective portion by defect mode. The defect mode is classified, for example, by the cause of the defect. The defect mode may be classified by cause, such as defects caused by remaining resist, mask, etc. without being removed, defects caused by uneven application of resist, etc., or defects caused by the adhesion of foreign matter. The shape of the defective portion has a tendency for each defect cause, such as being particulate, linear, or curved. The defect information acquisition unit 120 may determine the defect mode of the extracted defective portion based on the image acquired by the image acquisition unit 110.

The brightness information acquisition unit 130 acquires brightness information for at least a portion of the area of the image acquired by the image acquisition unit 110. In this specification, this area may be referred to as the brightness acquisition area. The brightness information in this example is information related to the brightness values of the image. The brightness information may be information based on the brightness values of multiple pixels in the brightness acquisition area. For example, the brightness information may be the average value, standard deviation value, maximum value, minimum value, median, etc. of the brightness values of each pixel in the brightness acquisition area. Furthermore, if the image includes intensity information for multiple colors (for example, brightness values for each of the colors red (R), green (G), and blue (B)), the brightness information may include information for each color.

The output unit 140 outputs the defect information acquired by the defect information acquisition unit 120 and the brightness information acquired by the brightness information acquisition unit 130 in association with each other. The output unit 140 may have a display device that displays the defect information and brightness information, a storage device that stores the information, an information processing device that performs predetermined processing on the information, or a transmission device that transmits the information to another device. The output unit 140 may display defect information including classification results obtained by classifying each defect portion by defect mode, together with the brightness information.

When inspecting multiple semiconductor devices 10 to extract defective portions based on images, it is preferable that the brightness of the images be as uniform as possible. However, image brightness also fluctuates depending on fluctuations in the light source intensity, the sensitivity of the image sensor, etc. Variation in brightness between images may result in variations in the accuracy of extracting defective portions or the accuracy of classifying defect modes.

In this example, the output unit 140 outputs the defect information and brightness information in association with each other. This allows the user of the inspection system 300 to determine the brightness of the image from which each piece of defect information was extracted. Furthermore, if the brightness information is not within a predetermined tolerance range, it can be determined that the reliability of the inspection results indicated by the corresponding defect information is low.

Based on the brightness information, the user of the inspection system 300 can adjust at least one of the light source that irradiates the semiconductor device 10 with light for imaging and the imaging device 20. This reduces the variation in brightness between images, making it possible to inspect multiple semiconductor devices 10. If the brightness information in the image of the semiconductor device 10 being analyzed varies by more than a predetermined reference value relative to the brightness information in the images of other semiconductor devices 10, the output unit 140 may adjust at least one of the light source and the imaging device 20 and cause the imaging device 20 to re-acquire an image of the semiconductor device 10 being analyzed.

Figure 2 shows an example of an image 201 of the semiconductor device 10 acquired by the image acquisition unit 110. The semiconductor device 10 has multiple regions formed thereon that will become semiconductor chips 12 when cut out. Image 201 may be an image of the entire surface of the semiconductor device 10 captured in one go, or may be an image of multiple partial images acquired by scanning the surface of the semiconductor device 10 and capturing portions of the entire surface of the semiconductor device 10, combined together.

As shown in Figure 2, image 201 includes images of multiple defect portions 210. As described in Figure 1, the defect information acquisition unit 120 extracts the defect portions 210 based on brightness contrast in the image, etc.

The brightness information acquisition unit 130 acquires brightness information in a brightness acquisition area included in the image 201. The brightness acquisition area may include an area that will become the semiconductor chip 12. The brightness acquisition area may or may not include an area that will not become the semiconductor chip 12.

Figure 3 shows an example of a method for acquiring brightness information. In this example, the brightness information acquisition unit 130 acquires brightness information for a brightness acquisition area 240 of the image 201 other than the defective portion 210. Figure 3 shows a portion of the image 201. First, the brightness information acquisition unit 130 acquires information indicating the position, size, shape, etc. of the defective portion 210 in the image 201 from the defect information acquisition unit 120 (S301).

Next, the brightness information acquisition unit 130 extracts an exclusion area 230 that includes the entire defect area 210 (S302). The exclusion area 230 may be the same size as the defect area 210, or may be larger than the defect area 210.

Next, the brightness information acquisition unit 130 acquires brightness information for the brightness acquisition area 240 in the image 201 other than the exclusion area 230. This processing makes it possible to suppress fluctuations in the average brightness, etc. of the brightness acquisition area 240 depending on the amount of defective portions 210, and enables accurate comparison of brightness information between each image 201.

Figure 4 is a diagram showing an example of defect information 142 and brightness information 144 displayed by the output unit 140. In this example, the output unit 140 displays the location of the defect portion in the semiconductor device 10 as the defect information 142. The output unit 140 also displays the defect mode of each defect portion as the defect information 142. In Figure 4, defect portions of defect mode A are indicated by circles, and defect portions of defect mode B are indicated by crosses.

The output unit 140 displays brightness information 144, which indicates the brightness of the image. As described above, brightness information 144 is, for example, the average brightness value of the image in areas other than the defective portion. This makes it possible to present the brightness of the image from which the defect information 142 was generated.

Figure 5 is a diagram showing the number of defects contained in semiconductor devices 10 for each production lot. The defect information acquisition unit 120 may generate at least a portion of the content shown in Figure 5 as defect information. The output unit 140 may display at least a portion of the content shown in Figure 5 as defect information. The horizontal axis in Figure 5 represents production lots arranged in chronological order. In another example, time, such as the production date or month, may be used instead of production lots. Furthermore, instead of classifying semiconductor devices 10 by production lot, semiconductor devices 10 may be classified by a predetermined number of inspections (e.g., 1,000 units). In this example, the number of defects is shown for each defect mode. One production lot may include one semiconductor device 10, or may include multiple semiconductor devices 10. Figure 5 shows the average number of defects contained in semiconductor devices 10 in each lot for each defect mode.

The defect information acquisition unit 120 may compare the number of defects with a predetermined threshold. If the number of defects exceeds the threshold, the defect information acquisition unit 120 may generate defect information including an alarm indicating which defect mode's number of defects exceeds the threshold. The threshold may be set in advance by a user, or may be generated by the defect information acquisition unit 120 based on the number of defects in past production lots. For example, the defect information acquisition unit 120 may set the threshold to a value obtained by multiplying the average number of defects in past production lots by a predetermined coefficient. Each defect mode is associated with a manufacturing process in which the defect is likely to occur, making it possible to perform maintenance on the equipment in that manufacturing process, adjust setting parameters, and so on. In this example, as shown in Figures 1 to 4, defects can be extracted by uniforming the brightness of each image, allowing changes in the number of defects to be detected accurately.

Figure 6 is a diagram showing the brightness of images of the semiconductor device 10 for each production lot. The brightness information acquisition unit 130 may generate at least a portion of the content shown in Figure 6 as brightness information. The output unit 140 may display at least a portion of the content shown in Figure 6 as brightness information. The horizontal axis in Figure 6 represents the production lots arranged in chronological order. In Figure 6, the average brightness of the images for each lot is shown by a solid line for each RGB color. The moving average of these average values is shown by a dashed line.

The brightness information acquisition unit 130 in this example acquires brightness information including a first brightness in an image of the semiconductor device 10 being analyzed and a predetermined second brightness. The second brightness may be a reference value preset by a user or the like, or may be the brightness in an image of another semiconductor device 10. In the example of Figure 6, the first brightness is the brightness of an image of the semiconductor device 10 in the latest lot, and the second brightness is the brightness of an image of the semiconductor device 10 in one or more past lots. In this way, the brightness information acquisition unit 130 acquires fluctuations in the first brightness and second brightness for multiple semiconductor devices 10 from different lots.

The brightness information acquisition unit 130 may generate a comparison result between the first brightness and the second brightness. For example, the brightness information acquisition unit 130 may generate a comparison result indicating how the brightness of the lot being analyzed (first brightness) has changed relative to the brightness of the past lot (second brightness) by generating time-series brightness information such as that shown in FIG. 6. The brightness information acquisition unit 130 may calculate a moving average of each brightness as the comparison result. In FIG. 6, the moving average of brightness for each RGB is shown by a dashed line. This makes it possible to grasp the trend in changes in brightness of the image.

Figure 7 is a diagram showing an example of the tolerance range for brightness. As in Figure 6, the solid line indicates the average brightness of the images in each lot, and the dashed line indicates the moving average of this average. In the example of Figure 7, the tolerance range for the brightness of green (G) is shown, but tolerance ranges may also be set for the brightness of other colors. In this example, the brightness information acquisition unit 130 determines whether each brightness is within a predetermined tolerance range. The brightness information acquisition unit 130 may generate brightness information including the determination result. For example, the brightness information acquisition unit 130 may generate brightness information including an alarm indicating which color's brightness is outside the tolerance range. In this example, an alarm is generated when the moving average of brightness indicated by the dashed line is outside the tolerance range.

The acceptable range may be set in advance by a user, or may be generated by the brightness information acquisition unit 130 from the brightness of images of past lots. The brightness information acquisition unit 130 may set a range with a preset width based on the brightness of the initial lot as the first acceptable range. In this case, the fluctuation in brightness (first brightness) of the lot being analyzed relative to the brightness (second brightness) of the initial lot is determined, making it possible to detect long-term fluctuations in brightness. The lot being analyzed may be the most recent lot, or it may be a lot from any period.

Figure 8 is a diagram showing another example of the tolerance range for brightness. In this example, the brightness information acquisition unit 130 sets a second tolerance range with a preset width based on the average brightness (second brightness) of one or more lots immediately preceding the lot being analyzed. In Figure 8, the average value is shown by a solid line, and the moving average of the average values of each lot is shown by a dashed line. The brightness information acquisition unit 130 determines whether the brightness (first brightness) of the lot being analyzed is within the second tolerance range. In other words, the brightness information acquisition unit 130 determines whether the first brightness relative to the second brightness is equal to or greater than the reference value. In this case, sudden fluctuations in brightness can be detected.

The brightness information acquisition unit 130 may determine whether the brightness of the lot being analyzed is within the first allowable range shown in FIG. 7 and the second allowable range shown in FIG. 8.

The brightness information acquisition unit 130 may generate an alarm indicating whether the brightness of the lot being analyzed does not satisfy the first or second tolerance range. The alarm may be generated for each color. This makes it possible to indicate to the user of the inspection system 300 whether the brightness is fluctuating over the long term or suddenly. The alarm allows the user to determine which part of the light source and image capture device 20 has the problem. For example, if the brightness is fluctuating over the long term, it can be determined that the light-emitting part of the light source has deteriorated over time. Furthermore, if the brightness is fluctuating suddenly, it can be determined that an incorrect setting value for operating the light source or image capture device 20 has been entered.

Figure 9 is a diagram showing another example of the tolerance range for brightness. In this example, the brightness information acquisition unit 130 determines whether the value obtained by dividing the average brightness (first brightness) of each lot by the moving average of the brightness (second brightness) of one or more immediately preceding lots is within the third tolerance range. The graph shown in Figure 9 corresponds to the value obtained by dividing the average brightness of each lot, indicated by the solid line in Figure 8, by the moving average brightness of each lot, indicated by the dashed line. The third tolerance range may be set in advance by the user, etc. The brightness information acquisition unit 130 may use a fixed third tolerance range regardless of the production lot. This example also makes it possible to detect sudden fluctuations in brightness. Furthermore, it is possible to detect sudden fluctuations in brightness by reducing the influence of long-term fluctuations in brightness.

Figure 10 is a diagram showing another example of the tolerance range for brightness. In this example, the brightness information acquisition unit 130 determines whether the value obtained by subtracting the moving average of the brightness (second brightness) for one or more immediately preceding lots from the average brightness (first brightness) for each lot is equal to or less than a predetermined tolerance value. The tolerance value may be set in advance by a user, etc. The brightness information acquisition unit 130 may use a constant tolerance value regardless of the production lot. This example also makes it possible to reduce the impact of long-term fluctuations in brightness and detect sudden fluctuations in brightness.

Figure 11 is a diagram illustrating another example of the analysis device 100. The analysis device 100 of this example differs from the analysis device 100 described in Figures 1 to 10 in that the defect information acquisition unit 120 corrects the image based on the image brightness information and acquires defect information based on the corrected image. Other functions are the same as those of the analysis device 100 described in Figures 1 to 10.

In this example, the defect information acquisition unit 120 acquires image brightness information from the brightness information acquisition unit 130. The brightness information may be the average brightness value for the entire image or the entire brightness acquisition area. A reference value for image brightness may be set in the defect information acquisition unit 120. The reference value for brightness may be set by the user, or may be set by the brightness information acquisition unit 130 based on the brightness of images in past production lots. For example, the brightness information acquisition unit 130 may use the brightness of images from an initial lot as the reference value, or the average brightness value of images from one or more immediately preceding lots as the reference value.

The defect information acquisition unit 120 corrects the brightness of the entire image to be analyzed so that the average brightness of the image to be analyzed approaches a reference value. For example, the defect information acquisition unit 120 calculates a correction coefficient by dividing the reference value by the average brightness of the image before correction. The defect information acquisition unit 120 may correct the brightness of the image by multiplying the brightness of each pixel of the image to be analyzed by the correction coefficient. This makes it possible to analyze defect information with the brightness of the image for each lot roughly aligned with the reference value. This reduces the impact of variations in brightness of the images for each lot.

The output unit 140 outputs at least one of the brightness information of the image before correction and the brightness information of the image after correction, in association with the defect information. The output unit 140 may output both the brightness information of the image before correction and the brightness information of the image after correction.

Figure 12 is a diagram illustrating another example of the analysis device 100. The analysis device 100 of this example differs from the analysis device 100 described in Figures 1 to 10 in that it includes a light source control unit 150. The other structures are similar to the analysis device 100 described in Figures 1 to 10. In addition, the inspection system 300 of this example includes a light source 30 that irradiates light onto the semiconductor device 10 when capturing an image of the semiconductor device 10.

In this example, the light source control unit 150 adjusts the brightness of the light source 30 based on the brightness information acquired by the brightness information acquisition unit 130. The light source control unit 150 controls the light source 30 so that the difference between the average brightness of the brightness acquisition area 240 (see Figure 3) and a predetermined reference brightness is reduced. The image acquisition unit 110 may acquire an adjusted image captured using the light source 30 whose brightness has been adjusted. The defect information acquisition unit 120 and the brightness information acquisition unit 130 may perform the processing described in Figures 1 to 11 on the adjusted image. This reduces brightness variation and enables defect information to be analyzed.

The image acquisition unit 110 may acquire an adjusted image when the difference between the image brightness and the reference brightness exceeds a predetermined threshold. If the difference is less than the threshold, the image acquisition unit 110 may acquire an image of the next semiconductor device 10 to be inspected, without acquiring an adjusted image of the same semiconductor device 10 after adjusting the light source 30. This allows multiple semiconductor devices 10 to be inspected efficiently. Furthermore, if the difference is less than the threshold, there is no need to adjust the light source 30.

Figure 13 is a diagram illustrating another example of the analysis device 100. The analysis device 100 of this example differs from the analysis device 100 described in Figures 1 to 10 in that it includes an instruction generation unit 160. The other structures are similar to the analysis device 100 described in Figures 1 to 10. In addition, the inspection system 300 of this example includes a light source 30 that irradiates light onto the semiconductor device 10 when capturing an image of the semiconductor device 10.

The instruction generation unit 160 generates instructions for the user of the inspection system 300 based on the brightness information acquired by the brightness information acquisition unit 130. The instructions may be instructions for controlling the light source 30. The instruction generation unit 160 may generate instructions to control the light source 30 so as to reduce the difference between the average brightness of the brightness acquisition area 240 (see FIG. 3) and a predetermined reference brightness. The instruction generation unit 160 may also generate an alarm as described in FIGS. 7 to 10. The instruction generation unit 160 may generate an instruction to check the status of at least one of the light source 30 and the imaging device 20, along with the alarm.

The instruction generation unit 160 may change the content of the instructions to the user depending on the nature of the brightness fluctuation. For example, if the brightness fluctuates over a long period of time as shown in FIG. 7, the instruction generation unit 160 may generate an instruction to check for deterioration of the light source 30 over time. Furthermore, if the brightness fluctuates suddenly as shown in FIG. 8, etc., the instruction generation unit 160 may generate an instruction to check for errors in the settings of the light source 30 and the imaging device 20. Furthermore, if the brightness fluctuates suddenly, the instruction generation unit 160 may generate an instruction to check for problems in the manufacturing process. This type of processing allows more appropriate instructions to be generated.

The instruction generation unit 160 may also determine whether the overall brightness, which indicates the magnitude of the overall brightness of the image or brightness acquisition area, is within a predetermined acceptable range. The overall brightness is, for example, the average, median, or mode of brightness in the entire image. If the overall brightness is outside the acceptable range, the instruction generation unit 160 may acquire brightness variation, which indicates the magnitude of brightness variation within the image. The brightness variation may be, for example, the standard deviation of brightness, the difference between the maximum value and the average value, or the difference between the minimum value and the average value.

The instruction generation unit 160 may generate an instruction to the user based on the brightness fluctuation. For example, if the overall brightness is outside the allowable range and the brightness fluctuation is below a predetermined threshold, it can be determined that the brightness has shifted throughout the image. In this case, the instruction generation unit 160 may generate an instruction to check the brightness of the light source 30.

Furthermore, if the overall brightness is outside the allowable range and the brightness fluctuation exceeds the threshold, it can be determined that the brightness is fluctuating locally. In this case, the instruction generation unit 160 may generate an instruction to check whether there are any problems in the manufacturing process.

Figure 14 is a diagram showing another example of a method for acquiring brightness information. In this example, the brightness information acquisition unit 130 acquires brightness information of a brightness acquisition area 240 in the image 201 that includes a defective portion 210. The processing performed by the inspection system 300 in this example may be similar to the examples in Figures 1 to 13, except that the range of the brightness acquisition area 240 is different.

First, the brightness information acquisition unit 130 acquires information indicating the position, size, shape, etc. of the defect portion 210 in the image 201 from the defect information acquisition unit 120 (S501). The processing of S501 is the same as the processing of S301 described in FIG. 3.

Next, the brightness information acquisition unit 130 extracts a brightness acquisition area 240 that includes the entire defective portion 210 (S502). The brightness acquisition area 240 may include the entire area of the defective portion 210. The brightness acquisition area 240 may coincide with the area of the defective portion 210, or may be larger than the area of the defective portion 210. For example, the brightness acquisition area 240 may be an area obtained by extending the area of the defective portion 210 outward by a predetermined number of pixels. The predetermined number of pixels is, for example, 10 pixels or less.

Next, the brightness information acquisition unit 130 acquires brightness information of the brightness acquisition area 240 (S503). The brightness information acquisition unit 130 may perform the same processing as that performed on the brightness acquisition area 240 by the brightness information acquisition unit 130 described in Figures 1 to 13. For example, the brightness information acquisition unit 130 may acquire the average value, standard deviation value, maximum value, minimum value, median value, etc. of the brightness values of each pixel in the brightness acquisition area 240. Furthermore, if the image includes intensity information for multiple colors (for example, brightness values for each of the colors red (R), green (G), and blue (B)), the brightness information acquisition unit 130 may acquire the above information for each color in the brightness acquisition area 240.

Also in this example, the defect information acquisition unit 120 may acquire defect information including classification results obtained by classifying each defect portion 210 by defect mode. The brightness information acquisition unit 130 may acquire brightness information of the brightness acquisition area 240 including the defect portion 210 for each defect mode.

According to this example, brightness information for the defective portion 210 can be easily acquired. By outputting the brightness information for the defective portion 210 in association with the defect information, analysis of the defective portion 210 can be supported. For example, even for the same defect mode, if the manufacturing process or parameters are changed, or if the state of the manufacturing equipment changes, the appearance of the defective portion 210 in the image 201 (e.g., brightness, chromaticity, and their distribution) may change over time. In such cases, when classifying the defective portion 210 into a defect mode, there is an increased possibility that the defect portion 210 will be classified into an incorrect defect mode. In response to this, by outputting the brightness information for the defective portion 210 in association with the defect information as needed, changes in the appearance of the defective portion 210 for each defect mode can be detected early.

It is also conceivable that a defect part 210 of a new defect mode not included in the preset defect modes may occur. Even in such cases, the occurrence of a defect part 210 due to a new defect mode can be detected early by correlating the brightness information of the defect part 210 with the defect information and outputting it as needed. Users can change the criteria or rules for classifying defect modes in response to changes in the appearance of the defect part 210 or the occurrence of a new defect mode, and can also create or update machine learning extraction modules. This allows the accuracy of defect mode classification to be maintained.

Figure 15 shows an example distribution of brightness information for a brightness acquisition area 240 that includes a defective portion 210. In this example, the brightness information is the average brightness of one or more brightness acquisition areas 240, but the brightness information may also be other information for the brightness acquisition area 240. The brightness information may be the average brightness of the brightness acquisition area 240 included in a predetermined unit. The predetermined unit may be one or a predetermined number of semiconductor chips, one or a predetermined number of semiconductor wafers, one or a predetermined number of lots, or semiconductor chips, semiconductor wafers, or lots manufactured within a predetermined period. The horizontal axis of each graph in Figure 15 represents the average brightness of the brightness acquisition area 240, and the vertical axis represents the probability of occurrence. The probability of occurrence indicates the percentage of cases (e.g., defective portion 210, semiconductor chip, semiconductor wafer, or lot) that have the corresponding average brightness for the brightness acquisition area 240.

Distributions 250, 252, 254, and 256 shown in FIG. 15 are distributions of brightness information for the defective portion 210. Distribution 250 is a distribution of brightness information that serves as a reference for a specified defect mode. Distribution 250 may be a distribution acquired over a specified period of time in the past. As an example, distribution 250 is a Gaussian distribution. Distribution 250 is a distribution when there is no change in the appearance of the defective portion 210 and no new defect mode has occurred. In contrast, if the appearance of the defective portion 210 changes or a defective portion 210 with a new defect mode occurs, the distribution of brightness information will fluctuate.

Distributions 252 and 254 show the distribution of brightness information when the appearance characteristics of the defective portion 210 in the defect mode change. Distribution 252 is an example in which the variance of the average brightness has increased. Distribution 254 shows a state in which the brightness of the defective portion 210 in the defect mode has decreased overall. In such a case, if the defective portion 210 is classified using the same criteria as distribution 250, the defective portion 210 will be classified into the wrong defect mode. In contrast, by adjusting the criteria for classifying the defective portion 210 based on the new distribution 252 or distribution 254, the defective portion 210 can be classified with high accuracy.

Distribution 256 shows the distribution of brightness information when a new defect mode occurs. In this example, the brightness information of defect portion 210 of the new defect mode is distributed as shown in distribution 258. In this case, distribution 256 is a distribution that combines distributions 250 and 258 of the two defect modes. When a new defect mode occurs, defect portion 210 of the new defect mode is classified into one of the existing defect modes. In contrast, by detecting the occurrence of a new defect mode based on distribution 256, it is possible to generate criteria for classifying defect portion 210 into the new defect mode, and it is possible to classify defect portion 210 with high accuracy.

The brightness information acquisition unit 130 may detect whether the criteria for classifying the defective portion 210 should be adjusted based on the distribution of the newly acquired brightness information. The output unit 140 may notify the user or the like of the detection result. For example, the brightness information acquisition unit 130 may detect whether the criteria for classifying the defective portion 210 should be adjusted based on the difference between the reference distribution 250 and the distribution of the newly acquired brightness information. The brightness information acquisition unit 130 may detect changes in the mean value, variance, number of peaks, etc. of the distribution.

Figure 16 is a diagram showing an example of classifying defect portions 210. In this example, defect portions 210 contained in multiple images 201-1 to 201-12 are classified into their respective defect modes. In this example, images 201-1 to 201-4 contain defect portion 210-1 of defect mode A, images 201-5 to 201-8 contain defect portion 210-2 of defect mode B, and images 201-9 to 201-12 contain defect portion 210-3 of defect mode C. Defect mode A and defect mode B are modes for which classification criteria have already been set in the defect information acquisition unit 120, while defect mode C is a mode for which classification criteria have not been set in the defect information acquisition unit 120. In other words, defect mode C is a newly emerged defect mode.

Even if a defect portion 210-3 of a new defect mode occurs, the defect information acquisition unit 120 classifies this defect portion 210-3 into an existing defect mode (A or B). In the example of Figure 16, the defect portion 210-3 is classified into defect mode A. As a result, the defect portion 210-3 is misclassified.

In contrast, the occurrence of a new defect mode can be detected from distribution 256 shown in Figure 15. When the occurrence of a new defect mode is detected, misclassification of defective portion 210-3 can be eliminated by setting criteria for classifying that defect mode.

Figure 17 shows an example of fluctuations in brightness information. The brightness information in this example is the average value, variance, etc. of the average brightness distribution of the defective portion 210 shown in Figure 14. The horizontal axis in Figure 17 represents the semiconductor wafer lot, and the vertical axis represents brightness information.

As described above, when the appearance of the defective portion 210 changes or a new defect mode occurs, the brightness information gradually changes. The brightness information acquisition unit 130 may determine whether the brightness information is within a preset tolerance range. The tolerance range may be set in advance according to the reference distribution 250 shown in FIG. 15. If the brightness information falls outside the tolerance range, the brightness information acquisition unit 130 may generate an alarm indicating this. This allows the user to discover that fluctuations in the brightness information have fallen outside the tolerance range without having to constantly check the brightness information of all lots.

If brightness information for the entire image 201 is used without focusing on the defective portion 210, the influence of the defective portion 210 on the brightness information will be small. This makes it difficult to detect changes in the appearance of the defective portion 210 or the occurrence of a new defect mode. In this example, by focusing on the defective portion 210 and monitoring changes in brightness information, changes in the appearance of the defective portion 210 or the occurrence of a new defect mode can be detected early.

In this example, the fluctuation in the average brightness of the defective portion 210 has been described as an example, but other indicators such as the standard deviation, variance, maximum value, minimum value, and mode of the distribution mentioned above may also be used as brightness information. A combination of multiple indicators may also be used. The same applies to grayscale images, color images, images obtained by converting a color image to grayscale, and images with an alpha channel as the image 201.

Figure 18 is a diagram showing an example of an analysis method using the analysis device 100. Figure 18 shows some of the steps of the analysis method, but the analysis method may include the steps described in Figures 1 to 13.

The analysis method includes an image acquisition step S402, an information acquisition step S404, and an output step S406. The processing in the image acquisition step S402 is the same as the processing in the image acquisition unit 110 described in FIGS. 1 to 13. The processing in the information acquisition step S404 is the same as the processing in the defect information acquisition unit 120 and the brightness information acquisition unit 130 described in FIGS. 1 to 13. The processing in the output step S406 is the same as the processing in the output unit 140 described in FIGS. 1 to 13.

Figure 19 shows an example of a computer 2200 in which aspects of the present invention may be embodied, in whole or in part. Programs installed on the computer 2200 may cause the computer 2200 to function as or perform operations associated with an apparatus or one or more sections of the apparatus according to embodiments of the present invention, and/or to perform methods or steps of methods according to embodiments of the present invention. Such programs may be executed by the CPU 2212 to cause the computer 2200 to perform specific operations associated with some or all of the blocks in the flowcharts and block diagrams described herein.

The computer 2200 according to this embodiment includes a CPU 2212, RAM 2214, a graphics controller 2216, and a display device 2218, which are interconnected by a host controller 2210. The computer 2200 also includes input/output units such as a communications interface 2222, a hard disk drive 2224, a DVD-ROM drive 2226, and an IC card drive, which are connected to the host controller 2210 via an input/output controller 2220. The computer also includes legacy input/output units such as a ROM 2230 and a keyboard 2242, which are connected to the input/output controller 2220 via an input/output chip 2240.

The CPU 2212 operates according to programs stored in the ROM 2230 and RAM 2214, thereby controlling each unit. The graphics controller 2216 retrieves image data generated by the CPU 2212 into a frame buffer or the like provided in the RAM 2214 or into the graphics controller itself, and causes the image data to be displayed on the display device 2218.

The communication interface 2222 communicates with other electronic devices via a network. The hard disk drive 2224 stores programs and data used by the CPU 2212 in the computer 2200. The DVD-ROM drive 2226 reads programs or data from the DVD-ROM 2201 and provides the programs or data to the hard disk drive 2224 via the RAM 2214. The IC card drive reads programs and data from an IC card and/or writes programs and data to an IC card.

ROM 2230 stores therein boot programs and the like that are executed by computer 2200 upon activation, and/or programs that depend on the hardware of computer 2200. I/O chip 2240 may also connect various I/O units to I/O controller 2220 via parallel ports, serial ports, keyboard ports, mouse ports, etc.

The programs are provided on a computer-readable medium such as a DVD-ROM 2201 or an IC card. The programs are read from the computer-readable medium, installed on the hard disk drive 2224, RAM 2214, or ROM 2230, which are also examples of computer-readable media, and executed by the CPU 2212. The information processing described in these programs is read by the computer 2200, resulting in cooperation between the programs and the various types of hardware resources described above. An apparatus or method may be configured by implementing information manipulation or processing in accordance with the use of the computer 2200.

For example, when communication is performed between computer 2200 and an external device, CPU 2212 may execute a communication program loaded into RAM 2214 and instruct communication interface 2222 to perform communication processing based on the processing described in the communication program. Under the control of CPU 2212, communication interface 2222 reads transmission data stored in a transmission buffer processing area provided in RAM 2214, hard disk drive 2224, DVD-ROM 2201, or a recording medium such as an IC card, and transmits the read transmission data to the network, or writes received data received from the network to a reception buffer processing area, etc., provided on the recording medium.

The CPU 2212 may also cause all or necessary portions of a file or database stored on an external recording medium such as the hard disk drive 2224, DVD-ROM drive 2226 (DVD-ROM 2201), IC card, etc. to be read into the RAM 2214, and perform various types of processing on the data on the RAM 2214. The CPU 2212 then writes the processed data back to the external recording medium.

Various types of information, such as various types of programs, data, tables, and databases, may be stored on the recording medium and may undergo information processing. CPU 2212 may perform various types of processing on data read from RAM 2214, including various types of operations, information processing, conditional judgment, conditional branching, unconditional branching, information search/replacement, etc., as described throughout this disclosure and specified by the program's instruction sequence, and write the results back to RAM 2214. CPU 2212 may also search for information in files, databases, etc. on the recording medium. For example, if multiple entries, each having an attribute value of a first attribute associated with an attribute value of a second attribute, are stored on the recording medium, CPU 2212 may search for an entry that matches a condition specified by the attribute value of the first attribute from among the multiple entries, read the attribute value of the second attribute stored in the entry, and thereby obtain the attribute value of the second attribute associated with the first attribute that satisfies a predetermined condition.

The programs or software modules described above may be stored on computer-readable media on or near computer 2200. Recording media such as a hard disk or RAM provided within a server system connected to a dedicated communications network or the Internet can also be used as computer-readable media, thereby providing the programs to computer 2200 via the network.

The present invention has been described above using embodiments, but the technical scope of the present invention is not limited to the scope described in the above embodiments. It will be clear to those skilled in the art that various modifications and improvements can be made to the above embodiments. It is clear from the claims that such modifications and improvements can also be included within the technical scope of the present invention.

The order of execution of each process, such as operations, procedures, steps, and stages, in the devices, systems, programs, and methods shown in the claims, specifications, and drawings is not specifically stated as "before," "prior to," or the like, and it should be noted that processes can be performed in any order, unless the output of a previous process is used in a subsequent process. Even if the operational flow in the claims, specifications, and drawings is described using "first," "next," etc. for convenience, this does not mean that it is necessary to perform the processes in that order.

10: Semiconductor device, 12: Semiconductor chip, 20: Imaging device, 30: Light source, 100: Analysis device, 110: Image acquisition unit, 120: Defect information acquisition unit, 130: Brightness information acquisition unit, 140: Output unit, 142: Defect information, 144: Brightness information, 150: Light source control unit, 160: Instruction generation unit, 201: Image, 210: Defective portion, 230: Exclusion area, 240: Brightness acquisition area, 250, 252, 254, 256, 258: Distribution, 300: Inspection system

Claims (12)

an image acquisition unit that acquires an image of the semiconductor device;
a defect information acquisition unit that acquires defect information based on a defect portion included in the image;
a brightness information acquisition unit that acquires brightness information of an area other than the defective portion of the image;
an output unit that outputs the defect information and the brightness information in association with each other ,
The defect information acquisition unit corrects the overall brightness of the image to be analyzed based on the brightness information, and acquires the defect information based on the corrected image.
Analysis device. an image acquisition unit that acquires an image of the semiconductor device;
a defect information acquisition unit that acquires defect information based on a defect portion included in the image;
a brightness information acquisition unit that acquires brightness information of an area other than the defective portion of the image;
an output unit that outputs the defect information and the brightness information in association with each other ;
a light source control unit that adjusts the brightness of a light source that irradiates the semiconductor device with light based on the brightness information;
Equipped with
the image acquisition unit acquires an adjusted image captured using the light source whose brightness has been adjusted;
The defect information acquisition unit acquires the defect information based on the defect portion included in the adjusted image.
Analysis device. an image acquisition unit that acquires an image of the semiconductor device;
a defect information acquisition unit that acquires defect information based on a defect portion included in the image;
a brightness information acquisition unit that acquires brightness information of an area other than the defective portion of the image;
an output unit that outputs the defect information and the brightness information in association with each other ;
The image capturing device further includes an instruction generating unit that generates, based on the brightness information, an instruction to a user to control a light source that irradiates the semiconductor device with light when capturing the image, or an instruction to check for deterioration of the light source over time.
Analysis device. the defect information acquisition unit acquires the defect information including a classification result obtained by classifying each of the defect portions by defect mode;
The analysis device according to claim 1 , wherein the output unit displays the defect information and the brightness information together. The analysis device according to claim 1 , wherein the brightness information acquisition unit acquires the brightness information according to a result of comparison between a first brightness and a second brightness in the image of the semiconductor device to be analyzed. The analysis device according to claim 5 , wherein the brightness information acquisition unit acquires brightness in the image of the semiconductor device other than the semiconductor device being analyzed as the second brightness. The analysis device according to claim 6 , wherein the brightness information acquisition unit acquires fluctuations in the first brightness and the second brightness for a plurality of semiconductor devices from different lots. The analysis device according to claim 5 , wherein the brightness information acquisition unit determines whether a variation amount of the first brightness with respect to the second brightness is equal to or greater than a reference value, and generates the brightness information including a determination result. an image acquisition step of acquiring an image of the semiconductor device;
an information acquisition step of acquiring defect information based on a defect portion included in the image and brightness information of an area of the image other than the defect portion ;
an output step of outputting the defect information and the brightness information in association with each other ,
In the information acquisition step, the overall brightness of the image to be analyzed is corrected based on the brightness information, and the defect information is acquired based on the corrected image.
Analysis method. an image acquisition step of acquiring an image of the semiconductor device;
an information acquisition step of acquiring defect information based on a defect portion included in the image and brightness information of an area of the image other than the defect portion ;
a light source control step of adjusting brightness of a light source that irradiates light onto the semiconductor device based on the brightness information;
an adjusted image acquisition step of acquiring an adjusted image captured using the light source whose brightness has been adjusted;
a defect information acquisition step of acquiring the defect information based on the defect portion included in the adjusted image;
an output step of outputting the defect information and the brightness information in association with each other. an image acquisition step of acquiring an image of the semiconductor device;
an information acquisition step of acquiring defect information based on a defect portion included in the image and brightness information of an area of the image other than the defect portion ;
an output step of outputting the defect information and the brightness information in association with each other ;
an instruction generating step of generating, based on the brightness information, an instruction to a user to control a light source that irradiates light onto the semiconductor device when capturing the image, or an instruction to check for deterioration of the light source over time;
An analysis method comprising: A program for causing a computer to execute the analysis method according to any one of claims 9 to 11 .