Method for distinguishing abnormity of test piece in material / fault analysis in digital mode

Abstract

The invention provides a method for distinguishing abnormality of a test piece in material / fault analysis in a digital mode, which comprises the following steps of: S1, detecting a semiconductor test piece in a passive VC mode of a scanning electron microscope (SEM) to obtain an SEM PVC image; s2, preparing a Layout diagram, originally designed on the metal contact layer, of a semiconductor test piece; s3, performing data processing on the SEM PVC image and the Layout image; s4, fault analysis; according to the invention, a digital processing mode is utilized, an SEM PVC image is obtained, a gray-scale value is analyzed, the SEM PVC image and the gray-scale value are automatically compared with an original design Layout graph, different positions are marked by a remarkable red star number, and the position of a fault transistor is easily identified. The judgment method is visual, the accuracy and the accuracy are far better than those of human eye identification images, errors of human eye judgment are avoided, the brightness of different transistors is accurately identified, the yield of material / fault analysis is improved, and the quality and the success rate of material / fault analysis are effectively improved.

Description

Technical Field

[0001] This invention relates to the field of semiconductor test wafer analysis methods, and more particularly to a method for digitally identifying test wafer anomalies in material / fault analysis. Background Technology

[0002] Passive voltage contrast (PVC) in semiconductor imaging (SEM) is a crucial detection method in semiconductor materials and failure analysis. Its principle lies in the fact that when an electron beam (SEM) or ion beam (FIB) irradiates materials with different potentials, different secondary electron yields occur, resulting in varying image brightness. Different types of transistors exhibit different brightness levels; for example, NMOS transistors appear dark, PMOS transistors appear bright, and gate transistors, being floating, appear dark. Current technology relies on the human eye to identify SEM PVCs to determine if semiconductor wafers are abnormal. Engineers can judge the normality or abnormality of transistors based on the brightness of the SEM image; for example, a bright NMOS or gate transistor indicates an abnormality. However, with the continuous evolution of advanced manufacturing processes, even abnormal transistors show increasingly smaller differences in brightness on their PVCs compared to normal transistors, exceeding the level discernible to the human eye. This increases the risk of misjudgment and analysis failure. Summary of the Invention

[0003] The purpose of this invention is to address the shortcomings of existing technologies by proposing a method for digitally identifying specimen anomalies in material / fault analysis.

[0004] To achieve the above objectives, the present invention adopts the following technical solution: A method for digitally identifying specimen anomalies in material / failure analysis includes the following steps: S1: The semiconductor test piece was inspected using the passive VC mode of a scanning electron microscope (SEM) to obtain SEM PVC images; S11: Prepare a semiconductor test wafer, which includes PMOS, NMOS, and Gate. The semiconductor test wafer is processed to the metal contact layer, and the metal contacts of different transistors can be seen on the surface using SEM. In this application, SEM refers to a scanning electron microscope.

[0005] S12: Place the processed semiconductor test piece in the SEM chamber and analyze it in SEM PVC mode to obtain a SEM PVC image. In SEM PVC analysis, PMOS will appear bright, NMOS will appear gray, and Gate will appear completely dark or black. If the transistor has no faults, the type of transistor can be identified based on the brightness of the SEM PVC image, distinguishing between PMOS, NMOS, and Gate.

[0006] SEM PVC mode is a passive virtual slicing mode, a technique for enhancing imaging that uses an electron beam to scan and image the surface of a sample. This mode does not damage the structure of the semiconductor sample, avoiding damage to the sample, offers high resolution, reduces the complexity and time required for semiconductor sample processing, and is convenient and easy to implement.

[0007] S13: Saves SEM PVC images as JPG or TIFF format. Supports various image analysis software and can preserve every pixel and detail of the image, facilitating high-precision image analysis.

[0008] S2: Prepare the layout diagram of the original semiconductor test piece design on the metal contact layer; The layout diagram displays the coordinates of all metal contacts and transistor type information, including PMOS, NMOS, or Gate. The coordinates of the metal contacts are location coordinate regions, representing the location region of each transistor type. There is a one-to-one relationship between the transistors in the layout diagram and the transistors in the actual semiconductor test piece in the SEM PVC image; that is, the coordinates of each transistor in the actual semiconductor test piece in the SEM PVC image are the same as the coordinates of the corresponding metal contacts in the layout diagram. The layout diagram is saved as a JPG or TIFF file.

[0009] Both the layout diagram and the SEM PVC image are represented using a Cartesian coordinate system. The layout diagram uses its top-left corner as the origin (0,0), with the X and Y axes pointing right and downwards, in pixels. The SEM PVC image uses its top-left corner as the origin (0,0), with the X and Y axes pointing right and downwards, in pixels. Using image data simplifies the localization of pixel positions of the detected objects, facilitates image data analysis and processing, and enables automated analysis.

[0010] S3: Process the SEM PVC image and layout diagram; Data processing of SEM PVC images and layout diagrams was performed using data analysis and numerical calculation software, such as, but not limited to, Mathlab.

[0011] S31: Process the layout diagram; Load and display the layout map to obtain the coordinates of the metal contacts. Set the grayscale value of the metal contacts of all transistors to 255, which is completely white, and reconstruct an image as a reference image. Locate the metal contact coordinates, remove noise, and reduce interference from meaningless data, thereby saving computation time and storage space, reducing false positives and false negatives, enhancing the accuracy of the analysis results, reducing complexity, and allowing the algorithm to focus more on the true target area in the image, making the analysis process more efficient.

[0012] S32: Process SEM PVC images; Load and display the SEM PVC image. Select the metal contact area in the SEM PVC image and obtain the coordinates of the selected area. Average the grayscale values ​​of all pixels within these coordinates to define the grayscale value of this metal contact. To prevent calculation errors caused by discontinuous grayscale values ​​at the metal contact edge in the SEM PVC image and incorrect pixel selection, the grayscale value calculation range is defined as all pixels within 5 nm of the outermost edge of the metal contact. This ensures more accurate analysis and avoids misjudgment.

[0013] S33: Mark grayscale values; The calculated grayscale values ​​are marked above the selected area in the SEM PVC image. After all the markings are completed, a new image is formed as the processed image.

[0014] Specifically, the calculated grayscale values ​​are obtained, and the Text function is called to annotate the calculated grayscale values ​​above the coordinates of the selected area. This process is repeated for all selected areas, and the grayscale values ​​are annotated. After annotation, the image is saved as the processed image. All metal contacts will be marked with their grayscale values ​​for easy identification. When automatic identification is not required, technicians can determine which transistor in the SEM PVC image is faulty upon reaching this point, allowing for direct identification and further fault analysis. This saves a significant number of steps, is simple and convenient to operate, and is suitable for production scenarios with dedicated technicians. This improves the versatility of this method in actual production processes and facilitates its widespread use.

[0015] S34: Calculate the grayscale value difference to obtain the difference map; The grayscale difference value is obtained by subtracting the grayscale values ​​of the corresponding selected range of metal contacts in the processed image from the grayscale values ​​of all metal contacts in the reference image. The layout map is then reconstructed into a new image, and the resulting grayscale difference value is marked on all corresponding metal contacts. Merging two Cartesian coordinate systems into one, mapping the image to the same coordinate system ensures precise correspondence between each pixel, avoiding deviations or errors caused by different coordinate systems. This facilitates the analysis of grayscale differences at corresponding locations in the image and prevents misjudgments.

[0016] The algorithm iteratively compares the coordinates of the metal contacts in the layout diagram with the region coordinates in the SEM PVC image. It determines whether the region coordinates fall within the range of the metal contact coordinates, establishing a one-to-one correspondence between the region coordinates and the metal contact coordinates. The grayscale difference value is obtained by subtracting the corresponding region coordinate's grayscale value from the grayscale value of all metal contact coordinates in the reference image. A copy of the layout diagram is then created, and the metal contact coordinates of the copy are obtained. The `Text` function is used to annotate the corresponding grayscale difference values ​​obtained from the metal contact coordinates onto the metal contact coordinates. After annotation, the difference map is saved as a difference map. This method ensures strong traceability; the difference map allows for intuitive fault identification, facilitating fault diagnosis verification, enabling timely and accurate automatic judgment of errors, locating error points, and analyzing causes, thereby improving the yield rate of material fault analysis.

[0017] S35: Malfunction indicator; Compare the grayscale difference values ​​of two pairs of transistors of the same type. Transistors whose grayscale difference value exceeds 5 are marked with a red asterisk. Reconstruct an image and store it as a fault identification map.

[0018] First, obtain the layout diagram. Next, based on the transistor type, obtain the coordinates of all metal contacts with the same transistor type. Based on the coordinates of the metal contacts, obtain the corresponding grayscale difference values ​​to form a similar dataset. Iterate through the grayscale difference values ​​in the similar dataset, comparing each pair to determine if the difference exceeds 5. If it exceeds 5, mark the coordinates of the metal contacts with a difference exceeding 5 with a red asterisk. Iterate through all similar datasets for all transistor types, marking all those with differences exceeding 5, and store them as a fault identification diagram.

[0019] Specifically, select any grayscale difference value from the same dataset as the baseline value. Iterate through other grayscale difference values ​​in the same dataset and calculate their differences from the baseline value. Use the `abs()` function to obtain the absolute value of the difference between the two grayscale difference values. Determine if the absolute value of the difference exceeds 5. If it does, record the two current grayscale difference values. Calculate whether the absolute value of the difference between the two grayscale difference values ​​and other grayscale difference values ​​exceeds a threshold of 5. The threshold can be customized to be set between 5 and 10. If the absolute value of the difference between one of the current two grayscale difference values ​​and other grayscale difference values ​​exceeds 5 multiple times, obtain the coordinates of the metal contact corresponding to this grayscale difference value and mark it with a red asterisk on the metal contact coordinates of the layout diagram. Obtain the next similar dataset for the next transistor type and mark it with a red asterisk in the same way until all transistor types are determined, and store the fault identification diagram. Alternatively, multi-threading can be used to calculate anomalies in multiple similar datasets simultaneously. The difference judgment is faster, saving the calculation time of multiple gray level difference values ​​in the same dataset. It is suitable for complex semiconductor judgment, meets market demand, automatically identifies differences, realizes the automation and intelligence of judgment, improves production efficiency, and meets the needs of unmanned factories.

[0020] S4: Fault Analysis; According to the fault identification diagram, the transistor marked with a red asterisk is the faulty transistor. Once the location of the faulty transistor is determined, the next step of fault analysis can be carried out.

[0021] Compared with existing technologies, the beneficial effects of this invention are as follows: This invention utilizes digital processing methods to analyze the grayscale values ​​of PVC images obtained by SEM using data analysis and numerical calculation software, and automatically compares them with the original design layout diagram. Locations with differences are marked with a prominent red asterisk, easily identifying the location of faulty transistors. The judgment method is intuitive, with accuracy and correctness far surpassing human visual image recognition, avoiding errors in human judgment. Automatic digital comparison with the original layout diagram accurately identifies the brightness of different transistors in the PVC, improving the yield of material / fault analysis and effectively enhancing the quality and success rate of material / fault analysis. Attached Figure Description

[0022] Figure 1 This is a flowchart of the steps of the method for digitally identifying specimen anomalies in material / fault analysis according to the present invention; Figure 2 This is a flowchart of step S3 of the method for digitally identifying specimen anomalies in material / fault analysis according to the present invention; Figure 3 SEM PVC image of the method for digitally identifying specimen anomalies in material / fault analysis according to the present invention; Figure 4This is a reference image for the method of digitally identifying specimen anomalies in material / fault analysis according to the present invention; Figure 5 This is a schematic diagram illustrating the steps of the method for digitally identifying the selection range of specimen anomalies in material / fault analysis according to the present invention. Figure 6 This is a structural schematic diagram illustrating the selection range of the method for digitally identifying specimen anomalies in material / fault analysis according to the present invention. Figure 7 This is a processed image of the method for digitally identifying specimen anomalies in material / fault analysis according to the present invention. Figure 8 This is a diagram illustrating the differences in the method of digitally identifying specimen anomalies in material / fault analysis according to the present invention. Figure 9 This is a fault identification diagram of the method for digitally identifying specimen abnormalities in material / fault analysis according to the present invention. Detailed Implementation

[0023] To provide a further understanding of the purpose, structure, features, and functions of the present invention, detailed descriptions are provided below with reference to specific embodiments.

[0024] like Figure 1 and Figure 2 A method for digitally identifying specimen anomalies in material / failure analysis, comprising the following steps: S1: The semiconductor test piece was inspected using the passive VC mode of a scanning electron microscope (SEM) to obtain SEM PVC images; S11: Prepare a semiconductor test piece, which includes PMOS, NMOS, and Gate transistors. Process the semiconductor test piece to the metal contact layer, and the metal contacts of different transistors can be seen on the surface using SEM.

[0025] S12: Place the processed semiconductor test piece in the SEM chamber and analyze it in SEM PVC mode to obtain a SEM PVC image. In SEM PVC analysis, PMOS will show a bright brightness, NMOS will show a gray brightness, and Gate will show a completely dark or black brightness. If the transistor has no faults, the type of transistor can be identified based on the brightness of the SEM PVC image. Figure 3 It can distinguish whether a transistor is PMOS, NMOS, or Gate.

[0026] SEM PVC mode is a passive virtual slicing mode, a technique for enhancing imaging that uses an electron beam to scan and image the surface of a sample. This mode does not damage the structure of the semiconductor sample, avoiding damage to the sample, offers high resolution, reduces the complexity and time required for semiconductor sample processing, and is convenient and easy to implement.

[0027] S13: Saves SEM PVC images as JPG or TIFF format. Supports various image analysis software and can preserve every pixel and detail of the image, facilitating high-precision image analysis.

[0028] S2: Prepare the layout diagram of the original semiconductor test piece design on the metal contact layer; The layout diagram displays the coordinates of all metal contacts and transistor type information, including PMOS, NMOS, or Gate. The coordinates of the metal contacts are location coordinate regions, representing the location region of each transistor type. There is a one-to-one relationship between the transistors in the layout diagram and the transistors in the actual semiconductor test piece in the SEM PVC image; that is, the coordinates of each transistor in the actual semiconductor test piece in the SEM PVC image are the same as the coordinates of the corresponding metal contacts in the layout diagram. The layout diagram is saved as a JPG or TIFF file.

[0029] Both the layout diagram and the SEM PVC image are represented using a Cartesian coordinate system. The layout diagram uses its top-left corner as the origin (0,0), with the X and Y axes pointing right and downwards, in pixels. The SEM PVC image uses its top-left corner as the origin (0,0), with the X and Y axes pointing right and downwards, in pixels. Using image data simplifies the localization of pixel positions of the detected objects, facilitates image data analysis and processing, and enables automated analysis.

[0030] S3: Process the SEM PVC image and layout diagram; Data processing of SEM PVC images and layout diagrams was performed using data analysis and numerical calculation software, such as, but not limited to, Mathlab.

[0031] S31: Process the layout diagram; Load and display the layout image, obtain the coordinates of the metal contacts, set the grayscale value within the metal contacts of all transistors to 255 (completely white), and reconstruct an image as a reference image. For example... Figure 4 As shown, the metal contact coordinates are located, and noise is removed, reducing interference from meaningless data. This saves computation time and storage space, reduces false positives and false negatives, enhances the accuracy of analysis results, reduces complexity, and allows the algorithm to focus more on the true target area in the image, making the analysis process more efficient.

[0032] S32: Process SEM PVC images; like Figure 5Load and display the SEM PVC image, select the metal contact area in the SEM PVC image, obtain the coordinates of the selected area, and define the average grayscale value of all pixels within the selected area as the grayscale value of this metal contact. For example... Figure 6 To prevent calculation errors caused by discontinuous grayscale values ​​and pixel selection at the metal contact edge in SEM PVC, the grayscale value selection range is defined as all pixels within 5 nm of the outermost edge of the metal contact. This results in more accurate analysis and avoids misjudgments.

[0033] S33: Mark grayscale values; The calculated grayscale values ​​are then labeled above the selected area in the SEM PVC image. After all labels are completed, a new image is created for processing, such as... Figure 7 As shown.

[0034] Specifically, the calculated grayscale values ​​are obtained, and the Text function is called to annotate the calculated grayscale values ​​above the coordinates of the selected area. This process is repeated for all selected areas, and the grayscale values ​​are annotated. After annotation, the image is saved as the processed image. All metal contacts will be marked with their grayscale values ​​for easy identification. When automatic identification is not required, technicians can determine which transistor in the SEM PVC image is faulty upon reaching this point, allowing for direct identification and further fault analysis. This saves a significant number of steps, is simple and convenient to operate, and is suitable for production scenarios with dedicated technicians. This improves the versatility of this method in actual production processes and facilitates its widespread use.

[0035] S34: Calculate the grayscale value difference to obtain the difference map; The grayscale difference value is obtained by subtracting the grayscale values ​​of the corresponding selected range of metal contacts in the processed image from the grayscale values ​​of all metal contacts in the reference image. The layout map is then reconstructed into a new image, and the resulting grayscale difference value is marked on all corresponding metal contacts. Figure 8 Merging two Cartesian coordinate systems into one ensures precise correspondence between each pixel, avoiding deviations or errors caused by different coordinate systems. This facilitates the analysis of grayscale differences at corresponding locations in the image and prevents misjudgments.

[0036] The algorithm iteratively compares the coordinates of the metal contacts in the layout diagram with the region coordinates in the SEM PVC image. It determines whether the region coordinates fall within the range of the metal contact coordinates, establishing a one-to-one correspondence between the region coordinates and the metal contact coordinates. The grayscale difference value is obtained by subtracting the corresponding region coordinate's grayscale value from the grayscale value of all metal contact coordinates in the reference image. A copy of the layout diagram is then created, and the metal contact coordinates of the copy are obtained. The `Text` function is used to annotate the corresponding grayscale difference values ​​obtained from the metal contact coordinates onto the metal contact coordinates. After annotation, the difference map is saved as a difference map. This method ensures strong traceability; the difference map allows for intuitive fault identification, facilitating fault diagnosis verification, enabling timely and accurate automatic judgment of errors, locating error points, and analyzing causes, thereby improving the yield rate of material fault analysis.

[0037] S35: Malfunction indicator; Transistors of the same type are compared pairwise for grayscale difference. Transistors with a grayscale difference exceeding 5 are marked with a red asterisk, and a new image is generated. Figure 9 .

[0038] First, obtain the layout diagram. Next, based on the transistor type, obtain the coordinates of all metal contacts with the same transistor type. Based on the coordinates of the metal contacts, obtain the corresponding grayscale difference values ​​to form a similar dataset. Iterate through the grayscale difference values ​​in the similar dataset, comparing each pair to determine if the difference exceeds 5. If it exceeds 5, mark the coordinates of the metal contacts with a difference exceeding 5 with a red asterisk. Iterate through all similar datasets for all transistor types, marking all those with differences exceeding 5, and store them as a fault identification diagram.

[0039] Specifically, select any grayscale difference value from the same dataset as the baseline value. Iterate through other grayscale difference values ​​in the same dataset and calculate their differences from the baseline value. Use the `abs()` function to obtain the absolute value of the difference between the two grayscale difference values. Check if the absolute value of the difference exceeds 5. If it does, record the two current grayscale difference values. Calculate whether the absolute value of the difference between the two grayscale difference values ​​and other grayscale difference values ​​exceeds the threshold of 5. If the absolute value of the difference between one of the current two grayscale difference values ​​and other grayscale difference values ​​exceeds 5 multiple times, obtain the coordinates of the metal contact corresponding to this grayscale difference value and mark it with a red asterisk on the metal contact coordinates of the layout diagram. Obtain the next similar dataset for the next transistor type and mark it with a red asterisk in the same way until all transistor types are determined, and store the fault identification diagram. Alternatively, multi-threading can be used to calculate anomalies in multiple similar datasets simultaneously. The difference judgment is faster, saving the calculation time of multiple gray level difference values ​​in the same dataset. It is suitable for complex semiconductor judgment, meets market demand, automatically identifies differences, realizes the automation and intelligence of judgment, improves production efficiency, and meets the needs of unmanned factories.

[0040] S4: Fault Analysis; According to the fault identification diagram, the transistors marked with a red asterisk are the faulty transistors. Once the location of the faulty transistor is determined, the next step of fault analysis can be performed. Conventional and general methods can be used for fault analysis; this invention does not innovate any fault analysis methods, therefore the detailed steps of the fault analysis method will not be elaborated upon.

[0041] The present invention has been described in the above-described embodiments; however, these embodiments are merely examples for implementing the present invention. It must be noted that the disclosed embodiments do not limit the scope of the present invention. Conversely, any modifications and refinements made without departing from the spirit and scope of the present invention are within the scope of patent protection of the present invention.

Claims

1. A method for digitally identifying specimen anomalies in material / fault analysis, characterized in that: The following steps are involved: S1: The passive VC method of the scanning electron microscope SEM is used to detect the semiconductor test piece and obtain the SEM PVC image; S2: Prepare the layout diagram of the original design of the semiconductor test piece on the metal contact layer; S3: Data processing of SEM PVC images and Layout diagrams; S31: Processing Layout diagram; Load and display the Layout diagram, obtain the coordinates of the metal contacts, set the grayscale value of the metal contacts of all transistors to 255, and re-form an image as a reference image; S32: Processing SEM PVC images; Load and display the SEM PVC image, select the metal contact range in the SEM PVC image, obtain the regional coordinates of the selected range, and define the average grayscale value of all pixels within the regional coordinates as the grayscale value of the metal contact; S33: marking grayscale value; The calculated grayscale value is marked above the selected range corresponding to the SEM PVC image. After all markings are completed, a new image is formed as a processed image; S34: Calculate the grayscale value difference to obtain a difference map; Subtract the grayscale values ​​of the metal contacts in the selected range corresponding to the processed image from the grayscale values ​​of all the metal contacts in the reference image to obtain the grayscale difference value, reshape the Layout image into an image, and mark the obtained grayscale difference values ​​on all the corresponding metal contacts. After marking, store it as a difference map; S35: Marking failure; Compare the grayscale difference values ​​of transistors of the same type in pairs, mark the transistors whose grayscale difference values ​​exceed the threshold with red asterisks, and re-form an image to store it as a fault identification map; S4: Fault analysis; According to the fault identification diagram, the transistor marked with a red asterisk is the faulty transistor. Once the location of the faulty transistor is determined, the next step of fault analysis can be carried out.

2. The method for digitally identifying specimen abnormalities in material / failure analysis as claimed in claim 1, characterized in that: Step S1 includes: S11: Prepare a semiconductor test chip: The semiconductor test chip is processed to the metal contact layer, and the metal contacts of different transistors can be seen on the surface by SEM; the semiconductor test chip includes PMOS, NMOS, and Gate; S12: placing the processed semiconductor test piece in a SEM chamber and analyzing it in a SEM PVC mode to obtain a SEM PVC image; S13: Store the SEM PVC image in JPG or TIFF format.

3. The method for digitally identifying specimen abnormalities in material / failure analysis as claimed in claim 1, characterized in that: The layout diagram shows the coordinates of all metal contacts and information about transistor types; The transistor types include PMOS, NMOS, or Gate; the coordinates of the metal contacts are position coordinate areas, representing the position areas of each transistor type; both the Layout diagram and the SEM PVC image are represented using a Cartesian coordinate system; the coordinates of each transistor in the actual semiconductor test piece in the SEM PVC image are the same as the coordinates of the metal contacts corresponding to the Layout diagram.

4. The method for digitally identifying specimen abnormalities in material / failure analysis as claimed in claim 1, characterized in that: The selection range for calculating the grayscale value is defined as all pixels within 5 nm of the outermost edge of the metal contact.

5. The method for digitally identifying specimen abnormalities in material / failure analysis as claimed in claim 3, characterized in that: Step S34 is specifically as follows: cyclically comparing the coordinates of the metal contacts in the Layout diagram and the regional coordinates in the SEM PVC image, determining whether the regional coordinates are within the range of the metal contact coordinates, making one-to-one correspondence between the regional coordinates and the metal contact coordinates, subtracting the grayscale values ​​of the corresponding regional coordinates from the grayscale values ​​of all metal contact coordinates in the reference image to obtain grayscale difference values, copying the Layout diagram to obtain the coordinates of the metal contacts in the Layout diagram copy, calling the Text function to mark the corresponding grayscale difference values ​​obtained from the coordinates of the metal contacts on the coordinates of the metal contacts, and saving as a difference diagram after marking.

6. The method for digitally identifying specimen anomalies in material / failure analysis as claimed in claim 1, characterized in that: Step S35 includes: first obtaining a Layout diagram; then, according to the transistor type, obtaining the coordinates of all metal contacts of the same transistor type, obtaining the corresponding grayscale difference values ​​according to the coordinates of the metal contacts to form a similar data set; selecting any grayscale difference value in the similar data set as a reference value, looping through other grayscale difference values ​​in the similar data set and making differences with the reference value in turn, using the abs() function to obtain the absolute value of the difference between two grayscale difference values, judging whether the absolute value of the difference between the grayscale difference values ​​exceeds a threshold value, if it exceeds the threshold value, recording the current two grayscale difference values, respectively calculating whether the absolute value of the difference between the two grayscale difference values ​​and other grayscale difference values ​​exceeds the threshold value, if there are multiple situations where the absolute value of the difference between a certain grayscale difference value of the current two grayscale difference values ​​and other grayscale difference values ​​exceeds the threshold value, obtaining the coordinates of the metal contact corresponding to the grayscale difference value, marking the coordinates of the metal contact in the Layout diagram with a red asterisk, obtaining the similar data set of the next transistor type, marking the red asterisk in the same way, until all transistor types are judged, and storing the fault identification diagram.

7. The method for digitally identifying specimen abnormalities in material / failure analysis according to claim 6, characterized in that: The threshold is 5-10.

8. The method of digitally identifying specimen anomalies in material / failure analysis as claimed in claim 1, characterized in that: Step S33 specifically includes obtaining the calculated grayscale value, calling the Text function, marking the calculated grayscale value above the area coordinates of the selected range, looping through all selected ranges, marking the grayscale value; and saving as a processed image after marking.

9. The method for digitally identifying specimen abnormalities in material / failure analysis according to claim 6, wherein: Multithreading is used to simultaneously calculate anomalies of multiple similar data sets.

Images