Critical dimensional changes in a drawing

Abstract

It can receive reference images and captured images of a pattern. It can identify the contour of interest (ROI) of the pattern. For each reference image and captured image, it can determine one or more measurements of the pattern's dimensions for the ROI. Based on one or more measurements of the pattern's dimensions determined for each reference image and captured image, it can classify defects associated with the ROI.

Description

[0001] This application is a divisional application of the invention patent application with application number 201880073469.4. Technical Field

[0002] This disclosure relates to critical dimensions, and more specifically, to determining critical dimension changes in a pattern. Background Technology

[0003] The evolution of integrated circuit (IC) production or manufacturing on semiconductor wafers utilizes increasingly smaller design elements to implement ICs. For example, ICs can be implemented on a semiconductor wafer, with patterns of one or more design elements corresponding to different parts of the IC. Manufacturing processes can cause one or more defects in ICs on the semiconductor wafer. These defects can be identified and / or classified by measuring the dimensions of the patterns of the design elements corresponding to different parts of the IC. Summary of the Invention

[0004] The following is a simplified summary of the disclosure to provide a basic understanding of some aspects of the disclosure. This summary is not an extended overview of the disclosure. This summary is not intended to identify key or critical elements of the disclosure, nor is it intended to define any scope of any particular implementation of the disclosure or any scope of the claims. The sole purpose of this summary is to present some concepts of the disclosure in a simplified form as a preface to the more detailed description that follows.

[0005] In this implementation, a captured image of the pattern can be received. A reference image of the pattern can be received. A contour of interest of the pattern can be identified. For each reference image and captured image, one or more measurements of the pattern's dimensions can be determined for the contour of interest. Defects associated with the contour of interest can be classified based on one or more measurements of the pattern's dimensions determined for each reference image and captured image.

[0006] In some implementations, the profile of interest corresponds to the location of a deformation of a pattern associated with a defect.

[0007] In some embodiments, determining one or more measurements of the size of the pattern further includes determining a first plurality of measurements at a specific point on the pattern along a plurality of points of interest along the contour of interest in the captured image, wherein each of the first plurality of measurements corresponds to a distance from the specific point to a separate point along the contour of interest in the captured image. Furthermore, determining a second plurality of measurements at a specific point on the pattern along a plurality of points of interest along the contour of interest in the reference image, wherein each of the second plurality of measurements corresponds to a distance from the specific point to a separate point along the contour of interest in the reference image.

[0008] In some implementations, multiple points along the contour of interest are associated with different directions from a specific point in the pattern to the contour of interest.

[0009] In some implementations, the method further includes determining a critical dimension parameter value for each individual point along the contour of interest based on a pair of measurements from the first and second plurality of measurements.

[0010] In some implementations, defects are classified based on a critical size parameter value, which has the greatest difference between a first plurality of measurements from a captured image and another plurality of measurements from a second plurality of measurements from a reference image.

[0011] In some implementations, one or more measurement results are determined based on a distance transformation function. Attached Figure Description

[0012] This disclosure will be more fully understood in light of the detailed description below and the accompanying drawings showing various embodiments of the disclosure.

[0013] Figure 1A The illustration shows a block diagram of an example system used to determine changes in the critical dimension (CD) of a drawing.

[0014] Figure 1B The illustration is a flowchart illustrating an exemplary method for determining critical dimensional changes in a drawing, according to some implementation methods.

[0015] Figure 2A The illustrations are exemplary patterns of semiconductor wafers according to some embodiments of the present disclosure.

[0016] Figure 2B The illustration shows another example pattern of a semiconductor wafer according to some embodiments of the present disclosure.

[0017] Figure 2C The illustration shows another example pattern of a semiconductor wafer according to some embodiments of the present disclosure.

[0018] Figure 2D The illustration shows another example pattern of a semiconductor wafer according to some embodiments of the present disclosure.

[0019] Figure 2E The illustration shows another example pattern of a semiconductor wafer according to some embodiments of the present disclosure.

[0020] Figure 2F The illustration shows another example pattern of a semiconductor wafer according to some embodiments of the present disclosure.

[0021] Figure 3A The illustrations depict exemplary patterns of defects associated with critical dimensions according to some implementation methods.

[0022] Figure 3B The illustration shows another example of a defect associated with a critical dimension according to some implementation methods.

[0023] Figure 3C The illustration shows another example of a defect associated with a critical dimension according to some implementation methods.

[0024] Figure 4A The illustrations illustrate the determination of critical dimensional parameters for patterns of arbitrary shapes according to some embodiments of this disclosure.

[0025] Figure 4B The illustration illustrates another example of determining critical dimensional parameters of a line-based pattern according to some embodiments of this disclosure.

[0026] Figure 5A The illustration shows how critical size parameters are determined based on distance measurement results according to some implementation methods.

[0027] Figure 5B The illustration shows another example of determining critical size parameters based on distance measurement results according to some implementation methods.

[0028] Figure 5C The illustration illustrates how, according to some implementation methods, critical size parameters are determined based on ratios associated with a reference image and an inspection image.

[0029] Figure 6 The illustration shows the measurement of critical dimensions according to some implementation methods.

[0030] Figure 7 The illustration shows another example of measuring critical dimensions according to some implementation methods.

[0031] Figure 8 The illustration shows another example of measuring critical dimensions according to some implementation methods.

[0032] Figure 9 The illustrations illustrate exemplary methods for determining critical size changes according to some embodiments of this disclosure.

[0033] Figure 10 This is an example method for determining the critical dimensions of a pattern with an arbitrary shape based on some implementation methods.

[0034] Figure 11 This is another exemplary method for determining the critical dimensions of patterns with arbitrary shapes, based on some implementation methods.

[0035] Figure 12 This is an example method for determining the neighbor-wise critical size parameter of a pattern of arbitrary shape and its neighboring patterns, according to some implementation methods.

[0036] Figure 13 The block diagram is an example of a computer system in which embodiments of this disclosure can be operated. Detailed Implementation

[0037] This disclosure relates to determining critical dimensional variations in a pattern. Generally, semiconductor wafer inspection systems can analyze semiconductor wafers containing patterns corresponding to integrated circuits. For example, the inspection system can identify any defects in the pattern of the semiconductor wafer and can classify the identified defects.

[0038] Semiconductor wafer inspection systems identify patterns corresponding to defects on semiconductor wafers. These patterns can correspond to various shapes and complex structures. During the fabrication of integrated circuits on a semiconductor wafer, the shape or structure of a particular pattern can change, differing from what the particular pattern should have. Such pattern variations can cause defects in the semiconductor wafer. Operators in semiconductor wafer inspection can examine each defect. For example, an operator can inspect a defect and determine whether a change in the pattern impacts the integrated circuit fabricated on the semiconductor wafer. However, patterns fabricated on semiconductor wafers can correspond to different shapes and / or structures. Therefore, operators of semiconductor wafer inspection systems may need to spend a significant amount of time examining many different shapes and / or structures of different patterns.

[0039] This disclosure addresses the aforementioned and other shortcomings by determining critical dimensional changes in a drawing. Various critical dimensions of the drawing can be determined and used to determine whether changes in these critical dimensions would cause a deterioration in the quality of the corresponding drawing (e.g., a defect). For example, a reference image (e.g., a desired drawing) and an inspection image (e.g., a manufactured and / or defect-associated drawing) can be received. Critical dimensional values ​​or parameters for a specific drawing from the reference image, and critical dimensional values ​​or parameters for the corresponding drawing from the inspection image, can be determined.

[0040] In some implementations, a distance transformation (DT) operation can be used to determine critical size values. Critical size values ​​can be determined by identifying the contour of interest (COI) of a pattern and measuring distances based on the identified COI. The COI can be considered a deformed portion (MP) of the pattern. For example, the deformed portion can correspond to a portion of the pattern in the inspection image that differs from the corresponding portion of the corresponding pattern in the reference image. A first critical size value can be determined for the COI of the pattern in the reference image. In some implementations, the first critical size value is the distance between the pattern location and the COI on the reference image. A second critical size value can be the distance between the corresponding location of the pattern and the COI on the inspection image. Thus, two size values ​​of the pattern can be determined. A parameter or change in the critical size can then be determined based on the combination of the first and second critical size values. In some implementations, the change or parameter of the critical size value can be based on the ratio between the first and second critical size values. Therefore, the change or parameter of the critical size can identify a change in the pattern as a result of a deformed portion of the pattern. This change or parameter of the critical size can then be used to determine whether the change or alteration in the pattern is significant. For example, if a critical dimensional change or parameter exceeds a threshold associated with a pattern, the deformed portion can cause a defect that needs to be addressed in the semiconductor process for a semiconductor wafer.

[0041] The advantages of this disclosure include (but are not limited to) improved efficiency of semiconductor wafer inspection systems. For example, critical dimensional changes or parameters can be determined for many different types of patterns without operator input. Therefore, semiconductor wafer inspection can be performed in less time. Furthermore, the determination of critical dimensional changes in patterns by the semiconductor wafer inspection system improves the reliability of semiconductor manufacturing processes because patterns with critical dimensional changes indicating defects can be processed during the manufacturing process, thereby reducing subsequent defects in subsequent semiconductor wafers.

[0042] Figure 1A The illustration shows a block diagram of an example system used to determine changes in the critical dimension (CD) of a pattern. This system could correspond to a semiconductor wafer inspection system.

[0043] like Figure 1AThe first system (system 1) may correspond to a computer system, which includes a processing device (e.g., a processor) that implements a critical size component (SP). In some embodiments, the critical size component may correspond to firmware, software products, or software applications. The critical size component may implement a distance transformation (DT) function to determine critical size values ​​or measurement results of a pattern of the semiconductor wafer W. In some embodiments, the first system S1 may include a memory M, a computing device C, a display D, and an input unit (e.g., a keyboard) K. The computing device C may be operatively coupled to an optical tool or a scanning electron microscope (SEM) via a communication link L. Although the illustration shows an SEM, any other type of optical equipment may be used to obtain an image of the semiconductor wafer W.

[0044] In operation, computing device C receives data from a semiconductor wafer W via a semiconductor imager (SEM). The SEM can generate one or more images of one or more patterns manufactured on the semiconductor wafer W. Computing device C can receive a reference image (e.g., from memory M) corresponding to the one or more patterns to determine variations in critical dimensions of the one or more patterns.

[0045] When a drawing has been deformed or can be deformed, the DT function can be used to determine the critical dimension of the drawing relative to a specific location on the drawing for any arbitrary drawing. The CD change can be determined relative to a reference image, which can be presented as a simulated computer-aided design (CAD) image corresponding to the drawing's design, and / or an image of the corresponding drawing that does not contain deformed portions (e.g., a drawing without defects).

[0046] In some implementations, the DT function allows for reliable and accurate estimation of CD in the presence of defects. The DT function allows semiconductor wafer inspection systems to improve the monitoring of CD and critical dimensional variation (CDV) for all types of patterns with different structures, without requiring the operator to define the expected actual measurement results for each type of pattern.

[0047] Figure 1B The illustration is a flowchart illustrating an example method 150 for determining critical dimensional changes in a drawing. Method 150 can be executed by processing logic, which may include hardware (e.g., processing device, circuit system, dedicated logic, programmable logic, microcode, device hardware, integrated circuit, etc.), software (e.g., instructions running or executed on the processing device), or a combination of the above. In some embodiments, it may be executed by... Figure 1A Method 150 is performed on critical dimension components (SPs).

[0048] As shown in the icon, at box 151, the processing logic may receive a reference image and a captured image (or inspection image). The reference image and the captured image may contain patterns of a semiconductor wafer. The pattern in the reference image may be considered an ideal pattern without any defects or deformations, while the pattern in the captured image may contain defects or deformations. Therefore, the reference image may contain a pattern, and the captured image may contain a pattern manufactured based on the reference image. At box 152, the processing logic may identify contours of interest (ROIs) of the patterns in the reference image and the captured image. For example, a ROI may be a boundary or portion of a pattern that contains deformations in the captured image. A deformation may be a change in the pattern relative to a boundary or portion of the corresponding pattern in the reference image. In some embodiments, the deformation may be a defect in the pattern or a result of a potential defect. Therefore, a portion of the pattern may be identified as a ROI.

[0049] In block 153, the processing logic may determine one or more measurements of the pattern's dimensions for each reference image and captured image relative to the contour of interest. For example, a position on the pattern may be identified. The position may be located in a normal position relative to the contour of interest of the pattern. Therefore, the position may be separated from the contour of interest. Measurements of multiple points or positions along the contour of interest may be obtained. In some embodiments, a first measurement may be obtained for the reference image, and a second measurement may be obtained for the captured image. For example, a first measurement may be obtained to measure the distance of a position on the pattern relative to a point on the contour of interest in the captured image. Subsequently, at the same position on the pattern in the reference image (i.e., at the same position as in the captured image), a second measurement may be obtained for the captured image to measure the distance between this position and the corresponding point on the pattern. In some embodiments, multiple pairs of measurements may be obtained relative to each different point on the contour of interest. Therefore, for each point along the contour of interest, a distance value relative to the reference image and another distance value relative to the captured image may be obtained.

[0050] Reference Figure 1BIn block 154, the processing logic may classify defects associated with a contour of interest based on one or more determined measurements. A defect may cause a defect to affect the contour of interest. In some implementations, defects may be classified as significant or insignificant based on one or more measurements. A defect may be significant if it causes the integrated circuit containing the pattern to fail. Defects may be classified based on a critical size parameter value, selected based on one or more measurements. For example, for each pair of measurements, the critical size parameter may be determined based on the combination of a first measurement from a reference image and a second measurement from a captured image, relative to the same location on the pattern and the same point along the contour of interest. The critical size parameter may be the ratio between the first and second measurements. Subsequently, the critical size parameter with the maximum value (indicating the maximum change along the contour of interest between the reference image pattern and the captured image pattern) may be selected as the critical size parameter. Defects may then be classified based on the selected critical size parameter exceeding a threshold. For example, if the critical size parameter exceeds the threshold, the contour of interest may be considered a significant defect. Alternatively, if the critical size parameter does not exceed the threshold, the contour of interest may not be considered a significant defect. In some implementations, the threshold may be varied based on the type of defect (e.g., shape) associated with the contour of interest.

[0051] The reference patterns mentioned herein (e.g., original, initial, desired, designed, planned patterns) may be compared only with patterns (also referred to as arbitrary, captured, real, deformed, defective, inspected, or deformed patterns). In some implementations, the reference image of a pattern may be, for example, a CAD image, a simulated image of the desired reference pattern, a grayscale image of a reference pattern generated by SEM during the manufacturing process, an image of a reference pattern created by optical tools (e.g., in mask inspection), a die, or a group of dies (e.g., adjacent dies), etc. A reference profile may be the profile of the pattern at the reference image. For a pattern that is a single feature, any profile of the reference image of such a pattern may be understood as the reference profile of such a pattern. For a pattern that includes internal (or sub) profiles, the external and internal profiles on the reference image of such a pattern may be used as reference profiles. In a group of patterns, the external profile of a pattern adjacent to a basic pattern may be considered as the reference profile of the basic pattern.

[0052] The captured image of the pattern may involve information indicating the brightness produced on a photomask or semiconductor wafer when irradiated with a primary electron beam. In this case, the captured image corresponds to a real / arbitrary / deformed pattern. Alternatively, the captured image may be a simulated image of a planned deformation of a reference pattern. The captured image can therefore be any image based on the captured signal (generated by scanning the mask or semiconductor wafer with a beam or primary electron beam), information based on the captured signal (a file of digital measurements of image brightness), the functionality of the image, data obtained during image processing, filtering or enhancement, or simulation of the image. The reference image and the captured image may be in the form of visible images (e.g., grayscale images obtained by SEM). The captured contour may involve the contour of the pattern contained in the captured image.

[0053] The deformed portion mentioned herein may relate to at least one item on a drawing. For example, a deformed portion may be (but is not limited to) an alteration, defect, added feature, or area with characteristics different from other areas of the drawing. The deformed portion may have any shape / contour and may be located on any part of the drawing. The deformed portion may be a virtual deformed portion (such as an area on the drawing) where the mentioned alteration / defect / added feature / area is planned / expected. A contour of interest (COI) may relate to a set of points on the drawing and be affected (or expected to be affected) by an initial (reference) drawing deformation. For example, a COI may be at least a portion of a reference contour adjacent to (actual or virtual) the deformed portion and located where the deformed portion is "attached" (or should be attached) to a reference image / contour. For example, a COI may be understood as a portion of a reference contour adjacent to the deformed portion, or as a boundary between the drawing and the deformed portion of the drawing. The COI may be pre-selected or determined based on at least one drawing image (captured image or reference image) and based on information about the location and shape of the deformed portion (defect). A segment of the deformed profile (not a portion of the reference profile) can be called a deformed profile of interest (COI). A COI can be defined for a drawing that does not have any visible deformed portion (defect). For example, a specific segment of the reference profile (where potential deformation may exist) can be considered a COI and called a hypothetical (or virtual) COI. For a drawing on which a hypothetical COI has a specific location, CD parameters can be determined.

[0054] Distance Transform (DT) functions are operations that can be performed on a derived digital image representation. DT functions can be defined as maps, where each pixel of the image of interest is labeled with its distance to the nearest non-zero pixel. For example, non-zero obstacle pixels can be boundary pixels in a binary image.

[0055] Figure 2AThe illustrations depict exemplary patterns of semiconductor wafers. As shown, a basic pattern P1 may include its own outer contour, denoted RO (refer to its own contour), and its inner / sub-contour, denoted RCh (refer to the sub-contour). A neighboring pattern N1 is located close to pattern P1. A deformed portion MP1 may appear on one side of pattern P1. MP1 may be a protrusion (illustrated by a protruding thick line denoted MP1a). In some embodiments, the deformed portion may be a recess, such as MP2 (e.g., a thick dashed line). This deformed portion is denoted MP1b. The contour of interest may be the boundary between MP1a or MP1b and RO (e.g., the portion of RO between the thick dashed lines).

[0056] As shown in the figure, two types of CD measurements can be determined. For example, the first type (Type I) can be a measurement of the CD of the drawing P1 itself to estimate how the defect MP1 affects the quality of the drawing P1 itself. The second type (Type II) can be a measurement of the proximity CD of the drawing P1 relative to the drawing N1 to estimate how the defect MP1 affects the surrounding area of ​​the drawing P1.

[0057] Figure 2B The illustration shows another example pattern of a semiconductor wafer. As shown, Type III and Type IV measurements can be determined. For Type III measurements, pattern P2 has been reduced in size during the manufacturing process (the diameter / overall size of pattern P2 is consistently reduced). The outer, fine profile is designated as the reference self-profile RO of P2, while the inner, thick profile is the deformed new profile of P2. The difference is designated as the annular deformed portion MP2. The boundary between the deformed portion MP2 and the reference pattern is the reference profile RO of the pattern (considered as the profile of interest COI-2). In Type IV measurements, the overall size / diameter of pattern P3 is increased during manufacturing (the deformed new profile of P3 is illustrated by thick lines). The reference self-profile of pattern P3 is designated as RO. Because the overall RO has changed, RO also represents the boundary of the deformed portion MP3. In this case, the overall RO is the profile of interest COI-3. This defect type is Type IV, which can be handled by the proposed critical size measurement technique to determine the self-CD of patterns similar to P3.

[0058] Figure 2C The illustration shows another example pattern of a semiconductor wafer. For example, Figure 2C Can be regarded as Figure 2AThe changes occur where pattern P4 and its neighboring pattern N2 both have defects (MP4, MP5), which together alter the distance between adjacent patterns and thus change the proximity CD of either of these patterns. This defect type is Type V, where the profile of interest for pattern P4 is COI-4, and CN (the capture profile of neighboring pattern N2) can be used as a reference profile (relative to which the effects of defects MP4 and MP5 are estimated). COI-5 is the profile of interest for pattern N2, if its own CD is estimated.

[0059] Figure 2D The illustration shows another example pattern of a semiconductor wafer. As shown, the pattern can be a circular pattern P6, which contains a profile of interest COI-6 on one of its arcs. Deformations may appear on the arc, as indicated by the profile of interest COI-6. The CD parameters of pattern P6 can be measured based on the position of COI-6.

[0060] Figure 2E The illustration shows another example pattern for a semiconductor wafer. As shown, the pattern can correspond to a linear pattern P7 with a contour of interest COI-7, which can be hypothetical (e.g., virtual or imagined). The CD parameters of pattern P7 can be determined based on the position of COI-7.

[0061] Figure 2F The illustration shows another example pattern for a semiconductor wafer. As shown, pattern P8 may contain complex shapes. Critical dimension parameters for pattern P8 can be determined for the contour of interest COI-8. In some embodiments, the critical dimension parameters may correspond to identifying which dimension of pattern P8 will be most affected when deformed portions appear on COI-8.

[0062] Although Figures 2A to 2F It illustrates various example drawings, but can determine the critical dimensional changes of any type or shape of drawing.

[0063] Figure 3A , Figure 3B and Figure 3C This diagram illustrates how to determine the relative effect of drawing deformation (e.g., defects) on the drawing CD. Any drawing may contain deformed portions (e.g., defects) that alter the outer contour of the drawing (and consequently change the drawing's CD). Deformed portions on a drawing can affect the drawing's CD relative to one or more neighboring drawings.

[0064] Figure 3AThe illustration depicts an example drawing with defects associated with critical dimensions. As shown, the drawing may contain defects within the drawing (i.e., clearly defined target areas). For example, drawing P10 may contain complex shapes. Profiles 120 and 140 represent the planned reference shape of the drawing. The actual drawing P10 contains a defect or deformed portion MP (shaded area 130) that alters the outer profile of the drawing. For example, the deformed drawing now terminates at boundary / edge 160 (instead of 140). Edge 140 can be considered the profile of interest (COI-10) of drawing P10.

[0065] In this case, CD of drawing P10 represents the dimension or measurement distance that differs most (or changes most) relative to the reference drawing. For example, the maximum change distance is the distance most affected by the deformed portion / defect 130. (See also...) Figure 3A The diameter Dr of the reference drawing P10 (measured approximately to, and preferably orthogonally to, the profile of interest 160) is significantly larger than the diameter Dd (captured) of the defect drawing measured in the same direction (measured approximately to, and preferably orthogonally to, the profile of interest 160). For example, the diameter Dd (i.e., the defect diameter) can be the distance between the "o" at profile 120 and profile 160, and the diameter Dr (i.e., the reference diameter) can be the distance between the "o" at profile 120 and the "o" at profile 140. The scale Dd / Dr indicates the maximum effect of the defect, such that Dd (captured) can be considered as the CD of the deformed drawing P10. Because Dr can be considered as the CD of the reference drawing, the scale can be specified as CD. 捕获 / CD 参考 .

[0066] Given the scale and the specific direction of measurement, the critical dimensional change (CDV) of drawing P10 can be determined as follows:

[0067]

[0068] In some implementations, the value Ratio is the ratio as previously described, the value C is the weight coefficient, and the values ​​α, β, and γ are angles formed by the measurement directions of CD selected for the reference profile RC, the profile of interest COI, and the deformed profile of interest MCOI, respectively, as illustrated in the figure.

[0069] Figure 3B The illustration shows another example pattern with defects associated with critical dimensions. For example, Figure 3BThe diagram illustrates a pair of patterns D11 and D12, where the patterns are straight lines. A defect 135 on pattern D12 alters the CD of pattern D12 itself and its proximity CD (e.g., the critical dimension for the neighboring pattern D11). For the CD of the neighboring pattern D11, a reference distance Dr (measured between the unaltered / reference pattern D12 and the neighboring pattern D11) can be compared to a distance Dd (measured between the defective pattern D12 and the neighboring pattern D11). It is understood that this distance is most significantly affected by defect 135.

[0070] To estimate / measure the relative impact of a defect (or any deformed portion of a drawing) on ​​the drawing CD, a Dd / Dr scale can be used. The scale can be based on a distance Dd measured on a captured image of the deformed drawing (e.g., a defect), and a distance Dr measured on a reference image of the undeformed drawing (e.g., without a defect), where both distances are measured close to the deformed portion. For example, the distance can be measured close to the profile of interest 137 (e.g., surrounded by defect 135). The maximum scale can be selected as the CD scale.

[0071] Figure 3C The illustration shows another example drawing with defects associated with critical dimensions. For example, the effect of the deformed portion on the drawing and neighboring drawings can be determined. The practice is illustrated using an example drawing surrounded by several neighboring drawings.

[0072] Figure 3C The illustration shows a set of patterns. In the set of patterns, one of the patterns 122 is deformed by a protrusion 121, thus affecting both the size of pattern 122 and its distance from other adjacent patterns 124 and 126, which are closest to or approximate to the deformed portion 121. The impact of the protrusion 121 can be estimated or determined based on the maximum value of the relative CD changed due to the protrusion (i.e., defect) 121.

[0073] In order to measure the effect of protrusion 21 on the CD value or parameters of drawing 122, Dd1 and Dr1 of drawing 122 should be measured at the position / direction of the greatest difference / change.

[0074] The relative change in CD can be estimated as:

[0075] max|1-Dd (捕获) / Dref)|

[0076] As an example, the ratio Dd1 / Dr1, which can be measured in drawing 122 (relative to the profile of interest at the base of the protrusion), is equal to 1.31. To measure the relative effect of the defect or protrusion 121 on the CD parameters of drawing 122 with respect to drawing 126, Dd2 and Dr2 should be measured at the location / direction of the greatest difference / change (or such direction / location should be automatically determined). For example, the ratio Dd2 / Dr2, which can be measured for the closest neighboring drawing 126 (e.g., represented by the dashed line between drawing 122 and drawing 126), is equal to 0.71.

[0077] Figure 4A The illustration shows how to determine the critical dimensional parameters for a pattern of arbitrary shape. As shown in the diagram, the CD parameters of an arbitrary pattern can be determined by using a distance transformation function (measuring distances in various directions).

[0078] As shown, pattern 131 may contain a circular shape. The outer contour of pattern 131 is illustrated as the boundary of pattern 131. The outer contour can be considered as a reference contour. Arc 132 indicates a set of points S1 of interest, which may be subject to deformation or impact from a defect. A DT function can be applied in various directions near the interest contour (set of S1), i.e., between points on the pattern 131 contour and points on arc 132 containing the set of S1, to measure distances, and the DT values ​​of these multiple distance measurements are plotted on graph 134. The extreme values ​​of graph 134 correspond to the diameter 136 of the pattern, measured at the center of the arc (e.g., at the center of the set of points S1 of the interest). The extreme values ​​may represent the maximum distance between any point on the pattern 131 contour and any point on arc 132. This measurement result may be the CD parameter or value of the pattern for the interest contour S1 (e.g., the local CD parameter of the pattern itself).

[0079] Figure 4B The illustration shows another example pattern 140 with a line pattern, which includes a profile of interest 142 with a set of points S2. Applying a DT function to the pattern profile near the profile of interest 142 produces a graph 144 with three extrema. One of the extrema (e.g., the minimum value of the graph) is associated with a point on the line pattern where the distance between the pattern profile and profile 142 (e.g., the set of points S2) will change most significantly. The distance 146 (CD) measured at this point can be a CD parameter (e.g., a local CD) for the deformed pattern. Furthermore, because distance 146 has the smallest absolute value among the other distances forming the extremum 144 (distance 146 is perpendicular to 142), distance 146 can be selected as a local CD parameter or value for pattern 140.

[0080] Figure 4A and Figure 4BThe location of defects / deformations on a drawing (represented by the profile of interest) affects the drawing's CD scale. CD is called local CD and can be estimated by analyzing changes in DT extrema and scale. Estimation can be provided by considering the profiles of the reference drawing and the deformed drawing (or one of them and the profile of interest of the virtual / planned or real deformation), the proximity of the measurement direction to the perpendicular direction of the profile of interest, and selecting the best candidate among the DT function measurement directions.

[0081] Figure 5A and Figure 5B The illustration shows how to determine critical size parameters based on distance measurement results. For example, the illustration shows... Figure 5A and Figure 5B The icons illustrate two examples, where pattern P13 has a generally circular shape with deformed portions (or profiles of interest) of varying positions and shapes. In this example, the profile of interest indicates the boundary between the reference pattern and the deformed pattern. As shown, candidates for CD parameters or values ​​can be measured from various directions near the deformed portion / COI, and at least one candidate can be selected as a suitable local CD value or parameter. Directions are indicated by black dashed lines (e.g., DT). In some embodiments, directions closer to and perpendicular to the deformed pattern profile (and the profile of interest COI) may be preferred candidates for determining the pattern's CD parameters (the maximum relative change in size, illustrated by the measured Dr and Dd on the right).

[0082] Figure 5C The icon illustrates that critical dimension parameters are determined based on the ratio associated with the reference image and the inspection image. As shown in the icon, a reference image 150 of a pattern and a captured image 151 (e.g., a captured image of a defect) can be received. The profile of interest 153 can be obtained in the combined image 153, and CD parameters or values ​​can be estimated from the resulting combined image 153.

[0083] Figure 5C It was also verified that the maximum CD ratio (indicating CD parameter or local CD) achievable for the circular shape of pattern 150 (with a defect present near profile of interest 153) was determined to be 0.83. Measurements near profile of interest 153 can be performed by applying the DT function.

[0084] Figure 6 , Figure 7 and Figure 8 The illustrations illustrate the measurement of critical dimensions. For example, each figure illustrates different patterns, selecting appropriate directions to measure CD parameters (e.g., DT operations) in different directions (illustrated by black dashed lines) near the contours of the profile of interest and deformed parts (e.g., protrusions and / or recesses). Corresponding DT value charts are constructed for these different patterns using the DT function. Figure 6 , Figure 7 and Figure 8 The central chart is at least related to the corresponding example placed to the left of the chart (e.g., related to the left and right examples, as they have the same COI and symmetrical deformation). Each DT chart plots the CD parameter candidates as the measurements most affected by the deformation (i.e., measurements close to the extreme value of the DT function). Then, the CD parameter is selected as the measurement with the largest CD proportion among the candidates. An exemplary value for the largest CD proportion is illustrated in the corresponding figure.

[0085] Figure 9 This is an exemplary method for determining changes in critical dimensions. Figure 9 The method can be executed by processing logic, which may include hardware (e.g., processing device, circuit system, special-purpose logic, programmable logic, microcode, device hardware, integrated circuit, etc.), software (e.g., instructions that run or execute on the processing device), or a combination of the above. In some implementations, it may be executed by... Figure 1A The critical dimension component (SP) execution method.

[0086] Generally, the CD parameters of a deformed pattern can be determined to estimate its quality. The quality of a deformed pattern corresponds to the measurement results of the influence of deformation or defects on its CD parameters. Furthermore, Figure 9 It includes individual branches for determining CD parameters, and for estimating the impact of the deformed portion on the drawing based on the individual CD parameters. For example, the first branch may be related to determining the CD parameters within the containing drawing (e.g., a left branch for determining its own CD parameters) and the CD parameters for one or more neighboring drawings (e.g., a right branch for determining the adjacent CD parameters).

[0087] In some implementations, local CD is measured within and / or outside the pattern deformed by a change in order or defect, and the extent to which the deformation alters the pattern is subsequently determined. For example, the ratio between the deformed dimension and the original (reference) dimension can be calculated, such that the output is the resulting change. As illustrated in the diagram, in box 10, the processing device can receive an input image. For example, the input image may contain at least one image of the pattern and some location-specific data that identifies whether the pattern has (or should have) a deformed portion MP, and identifies where the pattern has (or should have) the deformed portion MP. In some implementations, the image of the pattern may constitute a SEM image, a captured image of the pattern, or a reference image of the pattern. The data may contain information about the contour of interest (COI), which is the boundary between the pattern and the MP. If the MP is hypothetical or planned, the COI may be considered virtual.

[0088] In block 12, the processing device can form a set of boundaries (contours) based on the input received from block 10. For example, the boundary set may include the profile PC of a drawing, a reference profile or capture profile, the profile of the deformed portion (CMP) (if such a profile exists), the profile of interest COI being provided or created, and so on. Other contours can be obtained based on the input data.

[0089] In block 14, the processing device measures its own CD parameters. For example, block 14 may correspond to a first branch for determining the self-CD measurement results to estimate the impact of the MP on the containing pattern that includes the contour of interest itself. Block 14 corresponds to measuring the self-CD near the COI and / or CMP. Block 16 indicates a second branch for determining the proximity CD parameters or measurement results to estimate the impact of the MP on one or more neighboring patterns (e.g., other adjacent patterns that do not contain the contour of interest). Block 16 corresponds to measuring the proximity CD of the COI and CMP of the base pattern.

[0090] The measurement in block 14 can be performed by at least one of sub-branches a) and b). As illustrated, in block 14, the processing device measures its own CD parameter within a solid receiving pattern. In block 14b, the processing device measures its own CD parameter within a receiving pattern having an inner profile (or sub-profile). Block 15 indicates that for each of options 14a and 14b, its corresponding self-CD parameter can be selected as the measurement result that changes the most due to the presence of MP. However, among all the measurements performed in 14a and 14b, the self-CD parameter that changes the most can be selected.

[0091] In block 15, the processing device can calculate the respective change scale (and / or CDV) for the measurement results and can select the CD parameter characterized by the maximum value of the scale and CDV. Similarly or alternatively, block 16a may correspond to the processing device measuring the proximity CD parameter for a specific neighboring pattern. Block 16b may correspond to the processing device measuring the proximity CD for the deformed portion (MP') of a neighboring pattern, if it is also located between neighboring pattern pairs.

[0092] In block 17, for each of options 16a and / or 16b, the processing device may determine the nearest neighbor CD parameter that changes the most due to the presence of one or more MPs. In block 18, the processing device may collect the CD parameters determined from the two branches of the flowchart into two sets of combinations of itself and the nearest neighbor CD parameters, which change the most due to the presence of deformed MPs at specific locations. Individual scale and / or CDV values ​​may also be recorded.

[0093] Figure 10 The illustrations demonstrate an exemplary method for determining variations in critical dimensions for patterns of arbitrary shapes. Figure 10The method can be executed by processing logic, which may include hardware (e.g., processing device, circuit system, special-purpose logic, programmable logic, microcode, device hardware, integrated circuit, etc.), software (e.g., instructions that run or execute on the processing device), or a combination of the above. In some implementations, it may be executed by... Figure 1A The critical dimension component (SP) execution method.

[0094] Generally, CD parameters can be determined based on a distance transformation (DT) function. In some implementations, the DT function can determine the optimal measurement direction (associated with the extreme value of the DT function) and identify CD candidates in this direction. As shown in the icon, in block 20, the processing device can receive input. The input may include two grayscale scanning electron microscope (SEM) images (e.g., a reference image and a captured or inspected image) of the pattern and the surrounding area, such that information about deformed portions (MP) (e.g., defects) can be obtained as differences. For example, the difference may lie between the pattern in the reference image and the pattern in the inspected image. In the same or alternative implementations, the input may include a reference image and information about a hypothetical contour of interest (COI). In block 22, the processing device can form a set of contours. The contours may include a reference contour (RC) of the pattern, which may be selected from any or all of the following: a reference self-contour RO; a reference sub-contour RCh; a reference neighbor contour RN; and a captured neighbor contour CN (which may be used as a reference contour if it also has its own deformed portion).

[0095] In some implementations, the contour (CMP) of the deformed portion may be a portion of the same drawing or a portion of a neighboring drawing. The contour of interest (COI) may be the same as itself (imaginary or real) or a COI of a neighboring drawing. If CD parameters are measured for a drawing containing a COI (e.g., its own CD parameter measurement), the processing device in block 24 may apply a DT function to a reference contour RC near the COI / CMP to measure its own CD parameter value. The processing device may apply the DT function based on blocks 24a, 24b, and 24c.

[0096] In block 24a, the processing device applies a DT function using its own COI and its own CMP, where RC is a reference self-profile (RO). In some embodiments, instead of the self-CMP, a capture self-profile CO near the COI can be used. In block 24b, the processing device applies a DT function using its own COI and its own CMP, where RC is a reference sub-profile (RCh). In some embodiments, instead of the self-CMP, a capture self-profile CO near the COI can be used. In block 24c, the processing device may apply a DT function where RC is a reference self-profile (RO) and uses its own COI (e.g., no CMP). One or more extrema of the DT function can be obtained by applying the DT function along other points where the COI references the RO.

[0097] Reference Figure 10 In block 25a, the processing device determines the extreme value of the DT function for each CD parameter candidate in blocks 24a, 24b, and 24c. In block 25b, the processing device may determine the CD parameter candidate and / or CDV for each CD parameter candidate measured in block 25a. In block 25c, the processing device may determine the CD parameter candidate that produces the maximum CDV. In some embodiments, the CD parameter candidate that produces the minimum CDV may be determined.

[0098] like Figure 10 The illustration shows how proximity CD parameter measurements can be determined for adjacent, contiguous, or immediately adjacent patterns. For example, in block 26, the processing device can apply a DT function to a reference profile (RC) near the COI / CMP to measure the proximity CD parameter value. The DT function can be applied based on blocks 26a and / or 26b. For example, in block 26a, the processing device applies a DT function where the RC is a reference adjacent pattern profile (RN), and the DT function uses its own COI and its own CMP. In some embodiments, instead of its own CMP, a capture profile CO near the COI can be used. In block 26b, the processing device applies a DT function to determine the proximity CD parameter between two deformed portions of two different adjacent patterns. For example, two pattern profiles RC can be considered (e.g., the first RC could be a reference adjacent profile (RN), and the second RC would be a reference self-profile (RO)). Specifically, deformed portion MP1 can be the MP of its own pattern, and deformed portion MP2 can be the MP of an adjacent pattern. In the same or alternative implementations, RC can be selected as the capture profile of a neighboring pattern (CN illustrated in 26b), and DT can use its own COI and CMP. Therefore, several CD parameter values ​​that measure the distance between two deformed portions can be detected.

[0099] In block 27a, the processing device determines the CD parameter candidates as the extreme values ​​of the DT function received in blocks 26a and / or 26b. In block 27b, the processing device determines the proportion and / or CDV for each of the CD parameter candidates found in block 27a (e.g., based on blocks 26a and / or 26b). In block 28, the processing device can collect the CD parameter and / or CDV values ​​determined from all branches and sub-branches of the method into two combined sets: (1) self-CDV and (2) proximity CD / CDV parameters. The combined set may contain the self-CD / CDV of the pattern with real deformation, and / or the self-CD parameter determined for any pattern given a hypothetical COI. If the pattern contains more than one deformed portion MP, the self-CD / CDV can be selected as the maximum CD / CDV from those defined for each MP. The combination group may further include proximity CD parameters for several neighboring patterns, each of which may be subject to the largest change (e.g., having the largest scale / CDV) due to the presence of a deformed portion MP on the base pattern (pattern of interest) at a specific location. The second group may include combined proximity CD parameters selected from the mentioned proximity parameters to determine the neighboring pattern most affected by the MP on the base pattern. In some embodiments, the second group may include dual-combined proximity CD parameters selected for the patterns, having several MPs and several proximity parameters. Dual-combined CDs may indicate the maximum dimensional change (maximum CDV) resulting from the combination of a specific MP with a specific neighboring pattern.

[0100] The determination of CD parameter candidates can be further based on the calculation of ratios (e.g., Dd / Dr) and the derivation of CDV as a selected ratio function according to formula [1]: CDV = f(Ratio). The determination can be further based on the calculation of ratios and the derivation of CDV in percentage form, for example according to formula [2]:

[0101] CDV = max|(1-Ratio)|*100 (for any case, self-CD or proximity CD)

[0102] In the same or alternative implementations, CDV is determined based on formula [3]:

[0103]

[0104] In some implementations, the ratio is CD capture / CD reference, C is a weight factor, and α, β, γ are measurement directions for CD selected for the reference profile RC, the profile of interest COI, and the deformed profile of interest MCOI, respectively (see, for example, [reference]). Figure 3A The angle formed by ).

[0105] In block 28, the processing device may receive a combination set containing CD parameters and CDV values, which describe the impact of the deformed portion on its own deformation pattern and its adjacent area. In some embodiments, smoothing operations may be used to improve robustness and reduce the effects of noise (see, for example, [link to documentation]). Figure 5C ).

[0106] Figure 11 This is a flowchart of an exemplary method for determining the variation in critical dimensions for a pattern of arbitrary shape. Figure 11 The method can be executed by processing logic, which may include hardware (e.g., processing device, circuit system, special-purpose logic, programmable logic, microcode, device hardware, integrated circuit, etc.), software (e.g., instructions that run or execute on the processing device), or a combination of the above. In some implementations, it may be executed by... Figure 1A The critical dimension component (SP) execution method.

[0107] Generally, it can receive input, which may contain two binary images of the pattern (a reference image and a captured or inspected image), and the positional information of the deformed parts can be obtained as the difference between the two images. A contour group can be similar to... Figure 10 The outline group illustrated in box 22. In some embodiments, boxes 34-38 may correspond to... Figure 10 Boxes 24-28. For example, a contour group may include RO, RCh, and MPC. The COI can be obtained as the boundary between MP (MPC) and RO. In some implementations, the COI may be the boundary between MP and RCh. For each deformed portion MP detected on the drawing, some actions as follows can be performed.

[0108] In block 34, the processing device can apply the DT function to the COI for the RC (RO or RCh) to determine the function extrema. The RC can be selected as a reference profile excluding the COI and can be adjacent to the (closest) deformed portion (e.g., blocks 34.1, 34.2, and 34.3). In block 35, the processing device can use the extrema found on the RC to determine the measurement direction by creating a set of rays between the RC and the COI point, and perform a pair of measurements for the reference distance Dr and the capture / deformation / defect distance Dd in each of those directions (e.g., blocks 35.1 and 35.2). These measurements will then generate CD parameter candidates. While Dr can be measured between the RC and the COI, Dd can be measured in the same direction as the rays between the RC and the deformed portion profile MPC. Based on the measurements of Dr and Dd, the scale and / or CDV (e.g., for each direction / ray) can be estimated for each pair of such measurements.

[0109] In block 38, the processing device can determine or measure its own CD parameters and the pattern. For example, in block 38.1, the processing device selects a CD candidate with the largest scale and / or CDV. In block 38.2, the processing device can search for CD candidates for each deformed pattern if there is more than one deformed portion in the pattern. All CD candidates can then be compared to select the most critical self-CD parameter of the pattern as the self-CD measurement result (having the largest scale and CDV among all deformed portions).

[0110] In box 38.2, the processing device can measure the selected self-CD parameter. The selected self-CD parameter can also be (preferably in proportion to the selected self-CD parameter and CDV) transmitted to the combined CD group for this pattern (see boxes 18 and 28). It should be noted that for each MP (e.g., if there are more than one MP on the pattern of interest), a CD and CDV group corresponding to this MP will be generated, and the largest self-CD (and / or CDV) will then be selected from all MPs. See below. Figure 12 Explain the proximity CD binding groups (generated by boxes 38.3 and 38.4).

[0111] Figure 12 The illustration demonstrates an exemplary method for determining the proximity critical dimension of a pattern of arbitrary shape and adjacent patterns. Therefore, the proximity CD parameter can be measured. Figure 12 The method can be executed by processing logic, which may include hardware (e.g., processing device, circuit system, special-purpose logic, programmable logic, microcode, device hardware, integrated circuit, etc.), software (e.g., instructions that run or execute on the processing device), or a combination of the above. In some implementations, it may be executed by... Figure 1A The critical dimension component (SP) execution method.

[0112] In some implementations, input may be received, wherein the input may include a pattern and its surrounding binary image (referencing its own image, referencing neighboring images, capturing neighboring images), and the positional information of the deformed portion may include MP on its own image and MP' on neighboring images. The contour group may be similar to... Figure 10 The contour group illustrated in box 22. For example, the contour group may contain RO, RN, CN, and MPC. CN (the contour of the captured neighboring region) can be used as RC if the neighboring pattern also has deformed portions on the base pattern side. COI can be obtained as the boundary between MP (MPC) and RO. The method can be performed on each deformed portion (MP) detected on the base pattern itself.

[0113] In block 36, the processing device may apply the DT function to the COI over the RC (which is RN or CN) to find function extrema. For example, the processing device may use the function extrema (minimum, maximum) and endpoints found on the RC to determine the measurement direction by creating a set of rays between the RC (which is RN or CN) and the COI point, and perform a pair of measurements for Dr and Dd in each of those directions (e.g., blocks 36.1, 36.2, and 36.3). Each such pair of measurements will then generate proximity CD candidates.

[0114] In block 37 (e.g., 37.1 and 372), the processing device can measure Dr between RC and COI, and can measure Dd in the same direction as the ray between RC and the deformed portion profile MPC. Based on the measurements of Dr and Dd, the proximity ratio and / or CDV can be estimated for each such direction / ray.

[0115] In block 38 (e.g., block 38.3), the processing device can select proximity CD candidates based on the maximum proportion / CDV of a specific MP / defect on the base drawing. In block 38.4, the processing device can determine proximity CD candidates for each base drawing if there is more than one deformed portion in the base drawing. All CD candidates can then be compared to select the most critical (combined) proximity CD parameter of the drawing as the CD measurement result with the largest proportion and CDV in the deformed portion. For each MP on the base drawing, the proximity CD of the drawing can be selected as the CD that gives the largest CDV value among all neighboring drawings. If all options are performed together, the combined CD / CDV can be converted into a dual combined CD / CDV.

[0116] In box 38.4, the processing device selects the proximity CD parameter for the drawing. The selected proximity CD is transmitted to the combined CD group for this drawing (see boxes 18 and 28).

[0117] In some implementations, the proximity CD can be measured for each deformed portion (e.g., defect) on the base drawing (with or without applying a DT function). For example, the CD can be defined between the defect and neighboring drawings as the distance between the defect profile and the neighboring drawing to determine the maximum CDV. If a defect is also present on a neighboring drawing, the CD change or proportion caused by the defect or by both defects can be determined. A criterion (CDV) for the change can be calculated, and an optimal segment associated with the maximum CDV value (and the maximum proximity proportion) can be selected. This process can then be repeated for each MP on the drawing of interest for all neighboring drawings, and the neighboring drawing CD that gives the maximum CD change for this MP can be selected by combining the proximity CD. In some implementations, the neighboring drawing closest to the MP gives the maximum CD change. This process can be repeated for all defects (MPs) on the base drawing to select the maximum proximity proportion / CDV value among them, and an appropriate segment for measuring the dual-combined proximity CD of the base drawing.

[0118] Figure 13 The icon illustrates an example machine of computer system 1300, in which a set of instructions can be executed to cause the machine to perform one or more of the methods discussed herein. In alternative implementations, the machine may be connected (e.g., network-connected) to other machines in a LAN, internal network, external network, and / or the Internet. The machine may operate as a server or client machine in a client-to-server network environment, or as a peer machine in a peer-to-peer (or distributed) network environment, or as a server or client machine in a cloud computing infrastructure or environment.

[0119] This machine can be a personal computer (PC), tablet computer, set-top box (STB), personal digital assistant (PDA), cellular phone, web page device, server, network router, switch or bridge, or any machine capable of (sequentially or otherwise) executing a set of instructions specifying the actions to be taken by this machine. Furthermore, although a single machine is shown, the term "machine" should also be considered as encompassing any collection of machines that individually or jointly execute a set (or more) of instructions to perform any one or more of the methods discussed herein.

[0120] The example computer system 1300 includes a processing device 1302, a main memory 1304 (e.g., read-only memory (ROM), flash memory, dynamic random access memory (DRAM) such as synchronous DRAM (SDRAM) or Rambus DRAM (RDRAM), etc.), a static memory 1306 (e.g., flash memory, static random access memory (SRAM), etc.), and a data storage device 1318, which communicates with each other via a bus 1330.

[0121] Processing device 1302 represents one or more general-purpose processing devices, such as microprocessors, central processing units, etc. More specifically, the processing device may be a Complex Instruction Set Computing (CISC) microprocessor, a Reduced Instruction Set Computing (RISC) microprocessor, a Very Long Instruction Word (VLIW) microprocessor, or a processor implementing other instruction sets, or a processor implementing combinations of instruction sets. Processing device 1302 may also be one or more special-purpose processing devices, such as application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), digital signal processors (DSPs), network processors, etc. Processing device 1302 is configured to execute instructions 1326 to perform the operations and steps discussed herein.

[0122] The computer system 1300 may further include a network interface device 1308 for communication via a network 1320. The computer system 1300 may also include a video display unit 1310 (e.g., a liquid crystal display (LCD) or a cathode ray tube (CRT)), an alphanumeric input device 1312 (e.g., a keyboard), a cursor control device 1314 (e.g., a mouse), a graphics processing unit 1322, a signal generation device 1316 (e.g., a speaker), a graphics processing unit 1322, a video processing unit 1328, and an audio processing unit 1332.

[0123] Data storage device 1318 may include machine-readable storage medium 1324 (also referred to as computer-readable medium) having stored thereon one or more sets of instructions or software 1326 embodying any one or more methods or functions described herein. Instructions 1326 may also reside wholly or at least partially within main memory 1304 and / or processing device 1302 during execution by computer system 1300, which also constitute machine-readable storage media.

[0124] In one implementation, instruction 1326 includes instructions to implement a component corresponding to a critical size (e.g., Figure 1A The functionality of the critical size component (SP). Although the machine-readable storage medium 1324 is shown as a single medium in the exemplary embodiment, the term "machine-readable storage medium" should be considered to include a single medium or multiple media (e.g., a centralized or distributed database, and / or associated caches and servers) that store one or more sets of instructions. The term "machine-readable storage medium" should also be considered to include any medium capable of storing or encoding a set of instructions that is to be executed by a machine and to enable the machine to perform any one or more methods of the present disclosure. The term "machine-readable storage medium" should accordingly be considered to include (but is not limited to) solid-state memory, optical media, and magnetic media.

[0125] Some portions of the foregoing embodiments have been presented based on algorithms and symbolic representations of operations on data bits within computer memory. These algorithmic descriptions and representations are the means by which those skilled in the art of data processing most effectively communicate the essence of their work to others skilled in the art. Algorithms herein (and generally) are considered to be self-consistent sequences of operations that produce desired results. Operations require the manipulation of physical quantities. Typically (but not necessarily), these quantities are in the form of electrical or magnetic signals that can be stored, combined, compared, and otherwise manipulated. Sometimes, primarily for reasons of general use, it has proven convenient to refer to these signals as bits, values, elements, symbols, characters, items, numbers, etc.

[0126] However, it should be remembered that all these and similar terms should be associated with appropriate physical quantities and are merely convenient labels applicable to those quantities. Unless otherwise explicitly stated, it will be apparent from the above discussion that throughout the specification, the use of terms such as “identify,” “determine,” “execute,” “run,” “collect,” “create,” or “transmit,” etc., refers to the operation and procedures of a computer system or similar electronic computing device, which manipulate and transform data represented as physical (electronic) quantities in the registers and memory of the computer system into other data similarly represented as physical quantities in the computer system's memory or registers or other such information storage devices.

[0127] This disclosure also relates to apparatus for performing the operations described herein. Such apparatus may be specifically constructed for the intended purpose, or it may comprise a general-purpose computer selectively activated or reconfigured by a computer program stored in a computer. This computer program may be stored in a computer-readable storage medium, such as (but not limited to) any type of disk, including floppy disks, optical disks, CD-ROMs, and magneto-optical discs, read-only memory (ROM), random access memory (RAM), EPROM, EEPROM, magnetic cards or optical cards, or any type of medium suitable for storing electronic instructions, each coupled to a computer system bus.

[0128] The algorithms and displays presented herein are not inherently related to any particular computer or other device. Various general-purpose systems can be used with the programs taught herein, or it may be proven convenient to construct more specialized devices to execute the methods. The structures of various such systems will be presented below. Furthermore, this disclosure is not illustrated with reference to any particular programming language. It should be understood that the teachings of the disclosure described herein can be implemented using various programming languages.

[0129] This disclosure may be provided as a computer program product or software, which may include a machine-readable medium having instructions stored thereon, the instructions being used to program a computer system (or other electronic device) to perform a process according to this disclosure. A machine-readable medium includes any mechanism for storing information in a form readable by a machine (e.g., a computer). For example, a machine-readable (e.g., computer-readable) medium includes machine-readable storage media such as read-only memory (“ROM”), random access memory (“RAM”), disk storage media, optical storage media, flash memory devices, and the like.

[0130] In the foregoing description, embodiments of the disclosure have been described with reference to specific exemplary embodiments. It will be apparent that various modifications can be made without departing from the broad spirit and scope of the disclosure as set forth in the appended claims. The specification and drawings should therefore be regarded as illustrative rather than restrictive.

Claims

1. A method comprising: receiving input data comprising an image of a pattern and location data identifying a changed portion of the pattern; determining, by a processing device, a first parameter of a first dimension within the pattern by measuring a critical dimension parameter value at a location where a profile of the changed portion changes most relative to at least one of an outer profile of the changed portion or an inner profile of the changed portion; determining, by the processing device, a second parameter of a second dimension outside the pattern; generating a combination set based on the first parameter and the second parameter; and classifying a defect associated with the changed portion based on a critical dimension parameter determined based on a first value from the combination set and a second value associated with a reference image.

2. The method of claim 1, further comprising: determining a set of boundaries based on the input data, wherein the set of boundaries comprises at least one of a reference profile, a capture profile, the profile of the changed portion, or a profile of interest.

3. The method of claim 1, wherein determining the second parameter comprises measuring a critical dimension parameter value with respect to an adjacent pattern.

4. The method of claim 1, wherein determining the second parameter comprises measuring a critical dimension parameter value with respect to another changed portion of an adjacent pattern.

5. The method of claim 1, wherein the critical dimension parameter has a maximum difference between the first value from the combination set and the second value associated with the reference image.

6. A system comprising: a memory; and a processing device operably coupled with the memory for: receiving input data comprising an image of a pattern and location data identifying a changed portion of the pattern; determining a first parameter of a first dimension within the pattern by measuring a critical dimension parameter value at a location where a profile of the changed portion changes most relative to at least one of an outer profile of the changed portion or an inner profile of the changed portion; determining a second parameter of a second dimension outside the pattern; generating a combination set based on the first parameter and the second parameter; and classifying a defect associated with the changed portion based on a critical dimension parameter determined based on a first value from the combination set and a second value associated with a reference image.

7. The system of claim 6, wherein the processing device is further for: determining a set of boundaries based on the input data, wherein the set of boundaries comprises at least one of a reference profile, a capture profile, the profile of the changed portion, or a profile of interest.

8. The system of claim 6, wherein the processing device is further for measuring a critical dimension parameter value with respect to an adjacent pattern to determine the second parameter.

9. The system of claim 6, wherein the processing device is further for measuring a critical dimension parameter value with respect to another changed portion of an adjacent pattern to determine the second parameter.

10. The system of claim 6, wherein the critical dimension parameter has a maximum difference between the first value from the combination set and the second value associated with the reference image. ​ ​ ​ 11. A non-transitory computer-readable medium comprising instructions that, when executed by a processing device, cause the processing device to perform operations comprising: receiving input data comprising an image of a pattern and location data identifying a changed portion of the pattern; determining a first parameter of a first dimension within the pattern by measuring a critical dimension parameter value at a location where a profile of the changed portion changes most relative to at least one of an outer profile of the changed portion or an inner profile of the changed portion; determining a second parameter of a second dimension outside the pattern; generating a combination set based on the first parameter and the second parameter; and classifying a defect associated with the changed portion based on a critical dimension parameter determined based on a first value from the combination set and a second value associated with a reference image.

12. The non-transitory computer-readable medium of claim 11, wherein the operations further comprise: determining a set of boundaries based on the input data, wherein the set of boundaries comprises at least one of a reference profile, a capture profile, the profile of the changed portion, or a profile of interest.

13. The non-transitory computer-readable medium of claim 11, wherein the processing device further performs operations comprising measuring a critical dimension parameter value with respect to a neighboring pattern to determine the second parameter.

14. The non-transitory computer-readable medium of claim 11, wherein the processing device further performs operations comprising measuring a critical dimension parameter value with respect to another changed portion of a neighboring pattern to determine the second parameter. ​

Images