Image segmentation for inspecting semiconductor samples

Abstract

A system for inspecting a semiconductor sample that includes a plurality of layers at respective different depths, and a plurality of holes. Each well has a top portion at the surface of the sample and a bottom portion received in one of the layers. The system includes processing and memory circuitry (PMC) configured to provide an inspection image indicative of the aperture without using a shape characterization model, and to process an image of the aperture in the inspection image. The processing includes segmenting the inspection image and determining data indicative of a contour of a top portion of the aperture, and further segmenting the inspection image and determining data indicative of a contour of a shape enclosed within the contour of the top of the aperture.

Description

Technical Field

[0001] The presently disclosed subject matter relates generally to the field of image segmentation for inspecting semiconductor samples. Background Art

[0002] The current demand for high density and performance associated with ultra-large scale integration of manufactured devices requires sub-micron features, increased transistor and circuit speeds, and improved reliability. As semiconductor processes advance, pattern dimensions such as line widths and other types of critical dimensions continue to shrink. These demands require that device features be formed with high precision and uniformity, which in turn requires careful monitoring of the manufacturing process, including automated inspection of the device while it is still in semiconductor wafer form.

[0003] Nondestructive inspection tools can be used to provide inspection during or after the manufacture of the sample to be inspected. Inspection generally involves generating some output (e.g., image, signal, etc.) of the sample by directing light or electrons to a wafer and detecting the light or electrons from the wafer. Various nondestructive inspection tools include, by way of non-limiting example, scanning electron microscopes, atomic force microscopes, optical inspection tools, and the like.

[0004] The inspection process may include multiple inspection steps. The manufacturing process of the semiconductor device may include various processes, such as etching, deposition, planarization, growth (such as epitaxial growth), implantation, etc. The inspection step may be performed multiple times, for example, after certain process steps and / or after manufacturing certain layers, etc. Additionally or alternatively, each inspection step may be repeated multiple times, for example for different wafer positions or for the same wafer position with different inspection settings.

[0005] Inspection processes are used at various steps during semiconductor manufacturing to perform metrology-related operations and / or defect-related operations. The effectiveness of inspection can be improved by automating certain process(es), such as, for example, defect detection, automatic defect classification (ADC), automatic defect review (ADR), image segmentation, automated metrology-related operations, etc. Automated inspection systems ensure that manufactured parts meet expected quality standards and provide useful information about adjustments that may be needed in manufacturing tools, equipment, and / or components, depending on the type of error identified, in order to promote higher yields. Summary of the invention

[0006] According to certain aspects of the presently disclosed subject matter, there is provided a system for inspecting a semiconductor sample comprising a plurality of layers at respective different depths, and a plurality of holes, each hole having a top portion at a surface of the sample and a bottom portion received in one of the layers, the system comprising a processor and memory circuitry (PMC) configured to: a. Providing an inspection image indicating at least one hole in the well; b. Processing at least one hole image in the inspection image, comprising: i. segmenting the inspection image and determining data indicating the contour of the top portion of the hole; ii. segmenting the inspection image or a derivative thereof and determining data indicative of an outline of a shape of the bottom of the hole enclosed within an outline of the top of the hole; wherein the segmenting comprises classifying the image into at least two clusters of pixels, each cluster of pixels having pixel values ​​discernible relative to the other clusters of pixels, and determining data indicative of the outline of the shape based on the pixel values ​​of one of the clusters; Inspection of the sample is thereby facilitated, including utilizing at least the profile data indicative of the shape of the bottom of the hole for measuring at least a property of the bottom of the hole.

[0007] In addition to the features described above, the system according to this aspect of the presently disclosed subject matter may include one or more of the features (i) to (ix) listed below, in any desired combination or arrangement that is technically possible: (i) wherein the semiconductor sample is a stepped 3D NAND, and wherein the hole is a high aspect ratio (HAR) via, and wherein the bottom portion is accommodated in a tungsten layer of the sample. (ii) wherein the segmentation specified in (b)(ii) does not use a shape representation model. (iii) wherein the PMC is further configured to segment the image, comprising: applying at least two iterations of a KMeans function, wherein K>2 in a first iteration, K decreases in each subsequent iteration, and K=2 in a final iteration, thereby producing an image having at least two pixel clusters characterized by pixel values ​​that are distinguishable relative to other pixel clusters, wherein data indicative of an outline of a shape is determined based on the pixel values ​​of one of the clusters. (iv) wherein the computerized system comprises two iterations, and wherein in the first iteration K=3 and in the second iteration K=2. (v) wherein the PMC is further configured to apply a smoothing operation in at least two of the iterations. (vi) wherein the PMC is further configured to apply at least one morphological operation after the last iteration for obtaining data indicative of a different contour representing a corrected shape of the bottom of the hole. (vii) wherein the PMC is further configured to determine data indicating the outline of the top portion of the hole such that peripheral pixels are excluded and then apply the at least two iterations; wherein the peripheral pixels are pixels at the periphery of the hole having pixel values ​​indicating brighter pixels compared to pixels near the peripheral pixels. (viii) wherein the shape characterization model characterizes a geometry that is obtained at least from a priori design or drawing and is used to determine the sought shape or shapes. (ix) measuring at least one of the following properties at the top or bottom of the hole: CD - fit the contour to an ellipse and calculate 2*(major axis*minor axis) 1 / 2 . Area - fits the contour to an ellipse and calculates the area of ​​the ellipse. GL - Grayscale statistics, such as mean and standard deviation. The tilt is between the top and bottom COG (center of gravity): Direction – The angle between the vector connecting the COG and the X axis. Length – The Euclidean distance between COGs.

[0008] According to other aspects of the presently disclosed subject matter, there is provided a computerized method for inspecting a semiconductor sample comprising a plurality of layers at respective different depths, and a plurality of holes, each hole having a top portion at a surface of the sample and a bottom portion received in one of the layers, the method comprising, by a processor and memory circuitry (PMC): a. providing an inspection image indicating at least one of the holes in the hole; b. Processing at least one hole image in the image, comprising: i. segmenting the inspection image and determining data indicating the contour of the top portion of the hole; ii. segmenting the inspection image or a derivative thereof and determining data indicating the contour of a shape enclosed within the contour of the top of the hole without using a shape characterization model; Inspection of a sample is thereby facilitated, including utilizing at least data indicative of the contour of the shape of the bottom of the hole for measuring at least a property of the bottom of the hole.

[0009] This aspect of the presently disclosed subject matter may include one or more of features (i), (iii) to (ix), and (x) listed above with respect to the system, wherein the segmentation includes classifying the image into at least two pixel clusters, each pixel cluster having pixel values ​​that are distinguishable relative to the other pixel clusters, and determining data indicative of the outline of the shape based on the pixel values ​​of one of the clusters; after necessary modifications, wherein the features (i), (iii) to (ix) and (x) may be utilized in any desired combination or arrangement that is technically possible.

[0010] According to other aspects of the presently disclosed subject matter, there is provided a non-transitory computer-readable storage medium tangibly embodying a program of instructions that, when executed by a computer, causes the computer to perform a method for inspecting a semiconductor sample comprising a plurality of layers at respective different depths, and a plurality of holes, each hole having a top portion at a surface of the sample and a bottom portion received in one of the layers, the method comprising: a. Providing an inspection image indicating at least one hole in the well; b. Processing at least one hole image in the image, comprising: i. segmenting the inspection image and determining data indicating the contour of the top portion of the hole; ii. segmenting the inspection image or a derivative thereof and determining data indicative of a contour of a shape enclosed within the contour of the top of the hole without using a shape characterization model, Inspection of the sample is thereby facilitated, including utilizing at least the profile data indicative of the shape of the bottom of the hole for measuring at least a property of the bottom of the hole.

[0011] This aspect of the presently disclosed subject matter may include one or more of features (i), (iii) to (ix), and (x) listed above with respect to the system, wherein the segmentation includes classifying the image into at least two pixel clusters, each pixel cluster having pixel values ​​that are distinguishable relative to the other pixel clusters, and determining data indicative of the outline of the shape based on the pixel values ​​of one of the clusters; after necessary modifications, wherein the features (i), (iii) to (ix) and (x) may be utilized in any desired combination or arrangement that is technically possible. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] In order to understand the present disclosure and to see how it may be implemented in practice, various embodiments will now be described, by way of non-limiting examples only, with reference to the accompanying drawings, in which:

[0013] Figure 1 shows a generalized block diagram of an inspection system according to certain embodiments of the presently disclosed subject matter;

[0014] Figure 2 A semiconductor step 3D NAND sample with multiple high aspect ratio (HAR) vias is schematically shown;

[0015] Figure 3 schematically illustrates a plurality of contours in an image indicating corresponding different hole bottom shapes in a semiconductor sample according to certain embodiments of the presently disclosed subject matter;

[0016] Figure 4 shows a general flow chart of a model-free sequence of operations for segmenting an image and outlining the shape of a bottom of a hole in a semiconductor sample, according to certain embodiments of the presently disclosed subject matter;

[0017] Figure 5 shows a more detailed flow chart of a model-free sequence of operations for segmenting an image and outlining the shape of a bottom of a hole in a semiconductor sample, according to certain embodiments of the presently disclosed subject matter;

[0018] Figure 6 According to some embodiments of the presently disclosed subject matter Figure 4 A graphical illustration of the sequence of operations depicted in ;

[0019] Fig. 7A schematically illustrates a general flow chart for removing bright rings in an image according to certain embodiments of the presently disclosed subject matter;

[0020] Figure 7B schematically illustrates an auxiliary data structure used in the process of removing "bright" pixels according to some embodiments of the presently disclosed subject matter;

[0021] FIG. 8A to FIG. 8B schematically illustrates a plurality of images obtained as a result of applying a first Kmeans iteration and a second Kmeans iteration according to some embodiments of the presently disclosed subject matter;

[0022] Fig. 9 schematically illustrates an image obtained as a result of applying Gaussian smoothing according to some embodiments of the presently disclosed subject matter; and

[0023] Fig.10 An image indicating the shape of the bottom of a hole in a semiconductor sample obtained as a result of applying a profile correction operation according to some embodiments of the presently disclosed subject matter is schematically illustrated. DETAILED DESCRIPTION

[0024] A semiconductor sample (e.g., a stepped 3D NAND sample) may include multiple layers at correspondingly different depths, and multiple holes. The holes may have different hole bottom shapes, among other reasons, stemming from the HAR (high aspect ratio) characteristics of the holes. The specified HAR characteristics may affect the etching process, resulting in specified shape differences between different bottoms. For various purposes, it may be necessary to accurately determine the shape of the hole, such as defect inspection / detection, defect classification, segmentation, etc. However, given the wide differences between different possible bottom shapes, it is difficult to accurately determine the bottom shape.

[0025] Thus, according to certain embodiments of the present invention, there is provided a system (and related method) for inspecting a semiconductor sample comprising a plurality of layers at respective different depths, and a plurality of holes. Each hole has a top portion at the sample surface and a bottom portion contained in one of the layers. The system includes a processing and memory circuit system (PMC) configured to provide an inspection image indicating the holes without using a shape characterization model, and to process the hole image in the inspection image. The processing may include segmenting the inspection image and determining data indicating the contour of the top portion of the hole, and further segmenting the inspection image and determining data indicating the contour of the shape of the bottom of the hole enclosed within the contour of the top of the hole, thereby facilitating the inspection of the sample, including using the data indicating the contour of the shape of the bottom of the hole, for example, for measuring a characteristic of at least the bottom of the hole.

[0026] With this in mind, attention turns to Figure 1 , Figure 1 A functional block diagram of an inspection system according to some embodiments of the presently disclosed subject matter is shown.

[0027] Figure 1 The inspection system 100 shown may be used for inspection of semiconductor samples (e.g., wafers, dies, or portions thereof) as part of a sample manufacturing process. As noted above, inspection as referred to herein may be interpreted as covering any kind of operations related to defect inspection / detection, defect classification, segmentation, and / or metrology operations (such as, for example, critical dimension (CD) measurements, overlay, etc.) with respect to the sample. The system 100 includes one or more inspection tools 120 configured to scan the sample and capture images of the sample for further processing for various inspection applications.

[0028] The term "inspection tool(s)" as used herein should be interpreted broadly to cover any tool that can be used for inspection-related processes, including, by way of non-limiting example, scanning (in a single or multiple scans), imaging, sampling, reviewing, measuring, classifying and / or other processes provided with respect to a sample or portion thereof.

[0029] Without limiting the scope of the present disclosure, it should also be noted that the inspection tool 120 can be implemented as various types of inspection machines, such as an optical inspection machine, an electron beam inspection machine (e.g., a scanning electron microscope (SEM), an atomic force microscope (AFM), or a transmission electron microscope (TEM), etc.), etc. In some cases, the same inspection tool can provide low-resolution image data and high-resolution image data. The resulting image data (low-resolution image data and / or high-resolution image data) can be transmitted to the system 101 - directly or via one or more intermediate systems. The present disclosure is not limited to any specific type of inspection tool and / or the resolution of the image data generated by the inspection tool.

[0030] In some embodiments, at least one of the inspection tools 120 has metrology capabilities and can be configured to capture images and perform metrology operations on the captured images. Such inspection tools are also referred to herein as metrology tools.

[0031] According to some embodiments, the metrology tool may be an electron beam tool, such as, for example, a scanning electron microscope (SEM). A SEM is an electron microscope that generates an image of a sample by scanning the sample with a focused electron beam. The electrons interact with the atoms in the sample, generating various signals containing information about the surface topography and / or composition of the sample. The SEM is capable of accurately measuring features during semiconductor wafer manufacturing. As an example, the metrology tool may be a critical dimension scanning electron microscope (CD-SEM), configured to perform metrology operations relative to structural features of the sample based on a captured image.

[0032] It should be noted that the term "image(s)" as used herein may refer to an original image of a sample captured by an inspection tool during a manufacturing process, a derivative of a captured image obtained through various pre-processing stages, and / or a computer-generated design image based on data. It should be noted that in some cases, the images referred to herein may include image data (e.g., captured images, processed images, etc.) and associated digital data (e.g., metadata, handcrafted attributes, etc.). It should be further noted that the image data may include data associated with one or more layers of interest of the sample.

[0033] Semiconductor manufacturing processes typically require the manufacture of a series of layers, at least some of which include various structural features (also referred to as structures or features) manufactured by one or more processing steps (also referred to as process steps). Structural features may refer to elements or modules with a specifically designed structure and function to be manufactured on a layer. Metrology operations may be performed at various processing steps during the manufacturing process to provide measurements for the purpose of monitoring and controlling the process. Such measurements may include, for example, critical dimension (CD) measurements, overlay measurements, CD uniformity (CDU) measurements, etc.

[0034] In some cases, measurements can be obtained based on image segmentation, which delivers key information about the shape and size of structural features in the image, such as those indicating holes. For example, an image (indicating holes in different layers) can be segmented for the purpose of identifying the contours of the top portion of the hole and the contours representing the shape of the bottom portion of the hole (as will be explained in detail, these contours may vary from one hole to another). Once the contours are identified, various measurements can be applied. In this case, the quality of the measurements can also depend on the performance of the image segmentation.

[0035] Accordingly, certain embodiments of the presently disclosed subject matter provide a sample comprising a plurality of layers at respective different depths, and a plurality of wells. Each well has a top portion at the sample surface and a bottom portion received in one of the layers, all of which will be described below, for example, with reference to Figure 2 Description Certain embodiments provide methods and systems capable of accurately determining data indicative of a hole bottom shape profile, thereby facilitating inspection of a sample, for example, measuring various characteristics of the hole and / or hole bottom.

[0036] According to certain embodiments of the presently disclosed subject matter, inspection system 100 includes a computer-based system 101 operably connected to inspection tool 120 and capable of segmenting acquired images and determining a certain shape and contour of a hole.

[0037] Specifically, system 101 includes a processor and memory circuitry (PMC) 102 operatively connected to a hardware-based I / O interface 126. PMC 102 is configured to provide processing required by an operating system, such as reference Figures 2 to 10 In further detail, and including one or more processors (not shown separately) operably connected to a memory (not shown separately). The (multiple) processors of PMC 102 can be configured to execute several functional modules according to computer-readable instructions implemented on a non-transitory computer-readable memory included in the PMC. Such functional modules are hereinafter referred to as being included in the PMC.

[0038] Functional modules included in the PMC 102 of the system 101 may include, for example, a segmentation and contour determination module 104 , and an auxiliary module 106 .

[0039] The PMC 102 may be configured to obtain, at runtime, an inspection image of the semiconductor sample (an image indicating one of the holes and possibly other data) via the I / O interface 126 and from the inspection tool 120. The image may be acquired by one (or more) tools (e.g., a VeritySEM tool known per se). The present invention is not limited to this example. The segmentation and contouring module 104 may be configured to segment the acquired image (possibly in an iterative process) to determine data indicating the contour of the top of the hole and indicating the shape of the bottom of the hole relative to the one or more holes. Auxiliary modules may assist in auxiliary operations, all of which will be discussed in more detail below.

[0040] The operation of the system 100, the system 101, the system 110 and the PMC(s) 110 therein, and the functional modules therein will be referred to as follows. Figures 2 to 10 Described in further detail.

[0041] In some cases, the inspection system 100 may include one or more inspection modules, such as, for example, a defect detection module, and / or an automatic defect review module (ADR), and / or an automatic defect classification module (ADC), and / or additional metrology-related modules, and / or other inspection modules that may be used to inspect semiconductor samples, in addition to the system 101. The one or more inspection modules may be implemented as stand-alone computers, or the functionality of the one or more inspection modules (or at least portions thereof) may be integrated with the inspection tool 120. In some cases, the output of the system 101, such as, for example, an image associated with data indicating a bottom profile of a hole, may be provided to the one or more inspection modules for further processing.

[0042] According to some embodiments, the system 101 may include a storage unit 122. The storage unit 122 may be configured to store any data required to operate the system 101, for example, data related to the input and output of the system 101, and intermediate processing results generated by the system 101. As an example, the storage unit 122 may be configured to store images of samples produced by the inspection tool 120 and / or derivatives thereof. Thus, these images may be retrieved from the storage unit 122 and provided to the PMC 102 for further processing. The output of the system 101 may be sent to the storage unit 122 for storage.

[0043] In some embodiments, the system 100 may optionally include a computer-based graphical user interface (GUI) 124 that is configured to implement user-specified inputs related to the system 101. For example, a visual representation of a sample, including image data of the sample, may be presented to the user (e.g., via a display forming part of the GUI 124). The user may be provided with the option of defining certain operating parameters via the GUI. The user may also annotate reference images via the GUI. The user may also view the results of the operation on the GUI.

[0044] In some cases, the system 101 may be further configured to send the output data to the inspection tool 120 and / or one or more of the one or more inspection modules as described above for further processing via the I / O interface 126. In some cases, the system 101 may be further configured to send certain output data to the storage unit 122 and / or an external system (e.g., a yield management system (YMS) of a manufacturing plant (FAB)).

[0045] Those skilled in the art will readily appreciate that the teachings of the presently disclosed subject matter are not limited to Figure 1 104 and 106 and / or the operations performed thereby, as described below with reference to Figures 2 to 10 Equivalent and / or modified functionality may be combined or divided in another manner and may be implemented in any suitable combination of software and firmware and / or hardware.

[0046] It should be noted that Figure 1 The system shown can be implemented in a distributed computing environment where Figure 1 The aforementioned components and functional modules shown may be distributed across several local and / or remote devices and may be linked via a communication network. For example, the inspection tool 120 and the system 101 may be located at the same entity (in some cases hosted by the same device), or distributed across different entities.

[0047] It should also be noted that in some embodiments, at least some of the inspection tools 120, storage unit 122, and / or GUI 124 may be external to the inspection system 100 and in data communication with the system 100 and the system 101 via the I / O interface 126. The system 101 may be implemented as a (multiple) stand-alone computer used in conjunction with the inspection tools and / or additional inspection modules as described above. Alternatively, the various functions of the system 101 may be at least partially integrated with one or more inspection tools 120, thereby facilitating and enhancing the functions of the inspection tools 120 in the inspection-related processes.

[0048] Although Figure 1As shown, in some cases the functionality of system 110 may be at least partially integrated with system 100. As an example, the functional modules of system 110 may be incorporated into PMC 102 of system 101.

[0049] Although not necessarily so, the system 101 and the operating process of the system 100 may correspond to the Figures 2 to 10 Some or all of the stages of the method described. Figures 2 to 10 The described methods and possible implementations thereof can be implemented by the system 101 and the system 100. Therefore, it should be noted that Figures 2 to 10 The discussed embodiments may be implemented mutatis mutandis as various embodiments of system 101 and system 100 , and vice versa.

[0050] Attention shifts Figure 2 , Figure 2 A semiconductor step 3D NAND sample with multiple high aspect ratio (HAR) through holes as described above is schematically shown. It should be noted that for clarity, the sample is shown in a translucent view for a better understanding of the sample structure. It should be noted that the latter 3D NAND step sample is a non-limiting example of a semiconductor sample to be tested according to certain embodiments of the presently disclosed subject matter. For clarity of explanation, the description will focus on a specific example of a 3D NAND step sample, but those skilled in the art will readily understand that the present invention is not limited to this specific example, and the present invention can be applied to, for example, a HAR process requiring CD measurement of the bottom.

[0051] As shown, the 3D NAND sample 200 includes multiple layers (e.g., only two layers 201 and 202 of the multiple layers are labeled). These layers may reside at different depths relative to the top surface 203 of the sample. The 3D NAND step sample further includes a plurality of holes (also referred to as "vias"), of which 204 and 205 are labeled. Vias are typically characterized by a high aspect ratio (HAR) between the via depth and the via area (e.g., 1:40). Of course, the present invention is not limited by this numerical example. The vias penetrate the sample and ultimately enter the layers, for example, the bottom portion of via 204 resides in the (e.g., tungsten) layer 201, and the bottom portion of via 205 resides in the (e.g., tungsten) layer 202. Obviously, different holes may reside in different layers and the specified examples of 204 and 205 are by no means limiting. While the top portion of each hole may have substantially the same shape (e.g., circular), this is not the case for the bottom portion of the hole, which may vary significantly from one hole to another. See, for example Figure 3, shows one or more inspection images including a plurality of holes 300, each image showing a different internal (lighter) shape indicating the bottom of the corresponding hole. Thus, for example, hole image 301 encloses a shape of the bottom of hole 302 (indicated by a lighter color than the shape surrounding it), which is very different from shape 303 of the bottom of hole 304 (indicated by a lighter color than the shape surrounding 303). The same is true for bottom 305 of hole 306 and bottom 307 of hole 308. The designated shapes (which are provided for illustrative purposes only) indicate the bottoms of the corresponding holes, and the reasons why they are so different relative to each other, among other reasons, stem from the HAR characteristics of the holes, etc. The designated HAR characteristics can affect the etching process, resulting in differences in the designated shapes between the bottoms of different through holes.

[0052] It is important to characterize the shape of the corresponding hole to facilitate measuring hole characteristics (e.g., bottom or top), all of which will be explained in more detail below. These measurement data can be used during various manufacturing stages and other stages, such as discussed above in the background section.

[0053] Known techniques exist in the art that aim to identify the shape of a hole bottom by defining an outline of the shape of the hole bottom, while relying on a shape characterization model. However, given the wide variety of possible shapes, it is difficult to define a general model that will envision all possible shapes, and therefore, existing techniques that utilize shape characterization models may not only be prone to error, but may also cause computational overhead when trying to match the correct model to the shape being sought.

[0054] The hole property measurements using data indicative of at least the shape of the hole bottom and possibly the hole top wall may include at least one of: Calculate CD, area, and GL for the top and bottom contours separately: CD - fit the contour to an ellipse and calculate 2*(major axis*minor axis) 1 / 2 . Area - fits the contour to an ellipse and calculates the area of ​​the ellipse. GL - Grayscale statistics, such as mean and standard deviation. The tilt is between the top and bottom COG (center of gravity): Direction – The angle between the vector connecting the COG and the X axis. Length – The Euclidean distance between COGs.

[0055] These measurements can be performed by any inspection tool known per se. These data can be used in various processes as discussed above. Of course, the present invention is not restricted by these specific examples, and other examples can be applied instead of or in addition to at least one of the specified measures.

[0056] With this in mind, attention turns to Figure 4 , Figure 4 A general flow chart of a model-free sequence of operations for segmenting an image and determining the contour of the shape of the bottom of a hole in a semiconductor sample according to certain embodiments of the presently disclosed subject matter is shown. Attention is also turned to Figure 6 , Figure 6 Schematically depicts a schematic diagram of a system according to some embodiments of the presently disclosed subject matter. Figure 4 The sequence of operations depicted in .

[0057] Thus, in step 410, an inspection image is provided that indicates at least one of the holes. The image may be obtained by the PMC 102 via the I / O interface 126, and the image may be acquired by the inspection tool 120 during runtime inspection of the sample. The inspection image may be from a variety of inspection (multiple) modalities, such as, for example, by an optical inspection tool, an electron beam tool, a specific VeritySEM tool, etc., and the present disclosure is not limited by the specific inspection modality used to acquire the image. The inspection image may indicate one or more holes (through holes). The inspection image may include one or more perspective images captured from one or more viewing angles of the inspection tool.

[0058] After the inspection image has been obtained, the inspection image can be subjected to processing (e.g., by the segmentation and contour determination module 104 and the auxiliary module 106 of the PMC 102). The image is processed with respect to at least one of the holes (see, for example, the image 601 including the hole 602). The processing step includes (420) segmenting the hole and determining data indicating the contour of the top part of the hole (see, for example, the contour 603 of the hole image 602). There are known segmentation techniques in the art, such as GMM-Gaussian mixture model. Obviously, the present invention is not limited to this example. It should be noted that the data indicating the contour data can be represented in many possible ways known per se, such as a digital representation of the contour, an image (e.g., 603) including a graphical indication of the contour (the latter can be used, for example, to output to the user via, for example, the GUI module 124) and / or other ways.

[0059] Having identified the contour of the hole top, the inspection image or a derivative thereof (e.g., an inspection image with an indication of the hole top contour) may be subjected to further processing as described with respect to step 430, namely, segmenting the inspection image or a derivative thereof without using a shape characterization model, and determining data indicative of the hole bottom shape contour enclosed within the hole top contour (603). As previously described, the specified shape may vary significantly from one through-hole to another, among other reasons, because the HAR of a typical through-hole may affect the etching process and result in a different hole bottom shape. Non-limiting examples of a specific sequence of operations for achieving the specified determination of data indicative of the hole bottom contour will be described in further detail below. It should be noted that an area 605 of the image that is outside the shape contour (such as 604), but is still enclosed within the hole top contour (such as 603), may indicate a hole wall.

[0060] Data indicative of the hole bottom shape profile may facilitate inspection of the sample (eg, using various inspection tools 120 ), including measuring various characteristics of the hole and / or hole bottom, all as discussed above.

[0061] It should be noted that prior to processing a designated inspection image (eg, step 420, step 430), it may be necessary to localize the image (by methods known per se) to better process the hole image (eg, for processing each through-hole individually), as discussed above.

[0062] It should also be noted that where the inspection image includes more than one well image, the designated procedure described above may be repeated with respect to the other wells.

[0063] With this in mind, attention now turns to Figure 5 , Figure 5 A more detailed flowchart of a model-free sequence of operations for segmenting an image and determining a contour of the shape of the bottom of a hole in a semiconductor sample (e.g., by the segmentation and contour determination module 104 and auxiliary modules 106 of the PMC 102) according to certain embodiments of the presently disclosed subject matter is shown.

[0064] The program according to some embodiments includes segmenting the inspection image, including (possibly in an iterative manner), classifying the image into at least two pixel clusters, each pixel cluster having pixel values ​​that are distinguishable relative to the other pixel clusters, and determining data indicative of the outline of the shape based on the pixel values ​​of one of the clusters. It should be noted that the difference in pixel values ​​(grayscale) arises from the fact that the pixel values ​​represent different materials, for example, the bottom of the through hole may be made of tungsten material (which may be represented by a brighter (such as higher) pixel value), while the through hole wall may be made of silicon oxide material (which is represented by a darker (such as lower) pixel value).

[0065] Figure 5The illustrated process will now be described with reference to a specific segmentation technique (e.g., the K-means technique). It should be noted that the invention is not bound by the specified technique and that other techniques may be used, such as the OTSU technique known per se. Thus, having determined the data indicative of the contour of the top of the hole, the PMC may be configured to segment the image (this term also encompasses its derivatives), for example, by applying at least two iterations of the KMeans function, wherein in the first generation K>2, in each subsequent iteration K is decreased, and in the last iteration K=2.

[0066] Thus, as a non-limiting example, K=4 in the first iteration, K=3 in the second iteration and K=2 in the third iteration, or as another non-limiting example, two iterations, with K=3 in the first iteration and K=2 in the second iteration.

[0067] This will result in an image having at least two clusters of pixels, each cluster of pixels being characterized by pixel values ​​that are distinguishable relative to the other cluster of pixels, wherein the data indicative of the outline of the shape is determined based on the pixel values ​​of one of the clusters.

[0068] refer to Figure 5 The description of illustrates a specific non-limiting example of only two iterations, where K=3 in the first iteration (ie, applying the 3means technique), followed by K=2 iterations, ie, applying the 2means technique).

[0069] Initially, an initial optional step 501 is performed in which data indicative of a determined hole top contour is processed to produce a modified contour in which peripheral pixels are excluded. Peripheral pixels are pixels on the periphery of the hole that have "brighter" pixel values ​​than pixels near the peripheral pixels. Referring now to Fig. 7A , the "brighter" pixels (marked as 707) on the circumference of the hole are excluded so that these pixels are not enclosed within the contour, thereby obtaining a modified contour.

[0070] Back to Figure 5 This optional step may be advantageous for obtaining a hole bottom profile obtained in a subsequent processing step. The sequence of operations detailing the step of excluding peripheral "brighter" pixels will be discussed in more detail below with reference to FIG.

[0071] Continue to Figure 5 , a first Kmeans operation is applied (step 502), which, by this example, will produce an image with clusters of K pixel groups (eg, K=3).

[0072] Then, a smoothing operation (eg, as a non-limiting example, Gaussian smoothing) may optionally be applied to the image (step 503).

[0073] Subsequently (504), a second iteration of the Kmeans operation (eg, K=2) is applied to obtain a smaller number of clusters than the previous iteration.

[0074] Finally (505), optional morphological operations may be applied, such as contour correction operations, which will be described in more detail below.

[0075] Therefore, it should be noted that according to some embodiments, the Kth value is reduced in each iteration, thereby generating a smaller number of clusters compared to the previous iteration, until two clusters are obtained, one of which represents the shape of the bottom of the hole.

[0076] Before proceeding to illustrate an image obtained as a result of applying a Kmeans operation, a person skilled in the art will readily appreciate that a shape characterization model is not utilized. Intuitively, a shape characterization model is intended to model a given known shape or shape clusters that can be used by a detection tool (e.g., for measurement, classification, etc.) to classify the shape being sought. As discussed in detail above, because of the huge differences between the possible shapes representing the corresponding different through-hole bottoms, defining a shape characterization model that will encompass all possible bottom shapes is difficult and error-prone.

[0077] It should be noted that although Figure 5 Three auxiliary operations are shown (step 501, step 503, and step 505) - all three of these auxiliary operations may be implemented in the auxiliary tool 106 of the PMC 102, but the present invention is not constrained to utilize any of these three auxiliary operations, and accordingly, one or more of these three auxiliary operations may be eliminated or replaced by different auxiliary operations not disclosed herein. Alternatively, auxiliary operations not disclosed herein may be added to the reference Figure 5 It should also be noted that this consideration can be applied mutatis mutandis to other segmentation techniques, not necessarily the specified K means operation.

[0078] With this in mind, attention turns to Fig. 7A , Fig. 7A Schematically illustrates a flow chart for removing bright rings in an image according to some embodiments of the presently disclosed subject matter ( Figure 5 As specified above, this auxiliary operation is optional and is intended to improve the result of the subsequent segmentation step and can be implemented in an auxiliary tool 106 such as PMC 102.

[0079] For clarity, Fig. 7AThe flowchart of is shown in an image sequence, whereby image 701 depicts a hole 702 and an outline 703 indicating the top portion of the hole. As shown, image 701 includes a partial pixel ring 704 at the periphery of image 701, which is brighter than nearby pixels (in terms of grayscale value). This phenomenon can be encountered, for example, when the top contour is not close enough to the through hole, and a bright ring appears. According to certain embodiments, the presence of this "ring" (as an example of brighter pixels outside the periphery) can adversely affect the accuracy of clustering of the internal shape of the bottom of the indicating hole. Therefore, it is desirable to obtain data indicating a modified outline, which represents the top portion of the hole without including these "brighter" pixels. Modified outline 705 is described in image 706, wherein the specified "bright pixel" 707 is excluded.

[0080] Before implementing the next step, a top mask and an erosion mask may be placed. It should be noted that typically an erosion mask may be used to erode the shape of the top contour to determine at which pixels the desired percentiles will be calculated. The use of masks and their function are generally known per se.

[0081] Reference image 708 and Figure 7B The corresponding arrays in illustrate one possible technique for obtaining a modified contour. Thus, by this non-limiting example, the image 701 is sampled at different radii (possibly after using a mask). For simplicity, a slice 709 is depicted in the image 708, and the slice 709 is bounded by dividing lines labeled "2" and "3" (interchangeably referred to as radii). Obviously, the specified graphical representation of the dividing lines is only an example, and as another non-limiting example, a radius that can be represented as data indicating a radius (e.g., the angle of the dividing line relative to a reference angle such as "0 degrees" (not shown)) can be used. In addition, the number of radii may vary depending on the application.

[0082] continue Figure 7B , assuming that the entire hole depicted in image 708 includes multiple radii, such as 709. The circular hole 708 can then be mapped to a rectangle 710 in which the dividing lines "1", "2", and "3" are arranged at positions corresponding to the positions on the hole in image 708 (starting at "1", moving counterclockwise to dividing line "2", and then to "3").

[0083] As will be explained below, each separator line is processed to identify the outermost bright pixels (which should be removed), and then a modified outline is drawn based on the inner "non-bright" pixels of each separator line. Thus, starting with, for example, separator line "1", in order to "remove" the outermost bright pixels residing in the outer portion of separator line "1", a configurable threshold is set to, for example, an 8-bit representation (representing grayscale values ​​from 0 to 255), and the threshold is set to, for example, 210. This threshold indicates that any pixel with a value exceeding 210 (meaning that the pixel is "bright") will be removed (all numbers are provided for illustrative purposes and are by no means binding). Before continuing, it should be noted that each separator line can be associated with a corresponding auxiliary array having n cells (see, for example, Figure 7B 751, 752, and 753 corresponding to separation lines "1", "2", and "3", respectively). Thus, by this example, cell #0 in array 751 indicates the outermost pixel of separation line "1", and cell #n-1 indicates the innermost pixel of separation line "1". The same is true for arrays 752 and 753 corresponding to separation lines "2" and "3", respectively. Obviously, there may be additional separation lines (not depicted in hole image 708).

[0084] Continuing with the example of separator line "1", the pixel value in cell #0 (the outermost pixel) is compared to the threshold. Assuming that the value of the pixel value exceeds the threshold (i.e., in the latter example, the grayscale value of the pixel is higher than 210, i.e., the pixel is a "bright" pixel), the cell value is set to "0", indicating that the pixel should be removed (i.e., this is a "brighter" surrounding pixel). Then, the next pixel in cell #1 is processed, also assuming that the value of the pixel exceeds the threshold, so the pixel is also set to "0" (i.e., the pixel should be removed). This process continues until the first position is encountered, which indicates a pixel whose pixel value drops below the threshold of 210. Assuming that the pixel is found in cell #5, cell #5 is then set to "1". At this stage, processing of separator line "1" has been completed to indicate that the five outermost "brighter" pixels (positions 0-4) should be removed. This process is repeated for separator line "2", resulting in three pixels being removed (see positions 0-2 in array 752), and no pixels being removed for separator line "3" (see value "1" at position "1" in array 753). The same procedure is repeated for all separator lines (not shown in the figure). A modified contour can now be constructed for each separator line based on the pixels that were not removed (e.g., a modified contour consisting of the sixth pixel in line "1", the fourth pixel in separator line "2", and the first pixel in separator line "3", and so on for all other separator lines not shown in the figure), thereby producing a modified contour 705 that excludes bright pixel 707.

[0085] As discussed above, the use of separator lines and auxiliary arrays is of course a non-limiting example of the proposed procedure for removing external bright pixels. It should be noted that although for clarity, the description focuses on "bright" or "brighter" pixels (such as corresponding to grayscale values ​​exceeding a given threshold), those skilled in the art will readily note that the terms "bright" or "brighter" also include equivalent situations, such as, as a non-limiting example, representing a "bright pixel" as a "dark pixel" and the value sought for removal is a value that falls below a given threshold. For example, in the example of Figure 7, the external excluded pixels 707 each have a pixel value that represents a grayscale value of a pixel that is "darker" than the grayscale value of the pixels near the external excluded pixel 707 (enclosed in the modified outline 750).

[0086] Having described a non-limiting example of removing external bright pixels, attention turns to FIG. 8A to FIG. 8B , FIG. 8A to FIG. 8B Schematically illustrated are multiple images obtained as a result of applying a first Kmeans iteration and a second Kmeans iteration according to some embodiments of the presently disclosed subject matter. It should be noted that the present invention is not constrained by a specific Kmeans technique known per se for segmenting an image. As specified, these operations may be implemented in a segmentation tool 106 such as PMC 102.

[0087] Therefore, if Fig. 8A As shown, the input image (after being subjected to determining the outline indicating the top portion of the hole and possibly removing the external "bright pixels" as discussed in detail above) is fed to a Kmeans function (K=3 for this example), resulting in three clusters of pixels, each with pixel values ​​that are distinguishable relative to the other pixel clusters. The resulting image is shown in 802, with three groups corresponding to the three clusters 803, 804 and 805, respectively. For clarity, the different groups are shown in images 806, 807 and 808, respectively. The three cluster images can then be fed to a 2means (k=2) operation (see Figure 8B ), thereby reducing the three cluster images to two (or more) clusters with pixel values ​​that are distinguishable relative to the other clusters. As shown, the output image produced by the 2Means operation includes the surface 821, the wall 822, and the bottom 820 (3 clusters for this example), thereby producing (after contour segmentation) a contour 830. The latter indicates the bottom 820 of the hole.

[0088] It should be noted that Kmeans (k=2) is accurate because the previous K>2 iterations reduce the number of clusters, thus facilitating better results for the subsequent Kmeans (k=2) operations. Figure 8BAs shown, once two clusters have been determined, it is possible to define data indicative of a shape 830 and a contour representing that shape of the bottom of the hole without incurring the HAR of the hole and at the same time without using a shape characterization model.

[0089] As suggested, the present invention is neither constrained by the specified Kmeans function nor by the number of iterations.

[0090] According to some embodiments, known smoothing procedures may be applied between one or more iterations using, for example, Gaussian smoothing, e.g., see Fig. 9 As shown. The latter shows the application of a smoothing function between the 3kmeans and 2kmeans operations. Image 901 is the input (fed as the output of the 3kMeans operation), which may include integer representations of three clusters (e.g., values ​​1, 2, and 3 representing the corresponding three clusters) and in the resulting smoothed image (see 902). Each pixel can be represented as a floating point number (rather than an integer) that varies, for example, in the range of 1 to 3, thereby allowing better classification into two binary clusters in the subsequent 2Kmeans operation. The smoothed image will then be fed to the 2means module for obtaining two clusters in a specified manner, which can produce better results in view of the smoothing process.

[0091] Now turn to Fig.10 , Fig.10 Schematically illustrates an image obtained as a result of applying a morphological operation (e.g., a contour correction operation including removing redundant portions of the contour) according to some embodiments of the presently disclosed subject matter, the image indicating the shape of the bottom of a hole in a semiconductor sample. It should be noted that the morphological operation(s) may be, for example, part of an expansion or contraction of the contour shape.

[0092] Therefore, after determining the data indicating the shape and contour of the bottom of the hole, one or more morphological operations, such as contour correction operations, may be applied. For this example, the contour may modify the contour shape and / or size, or better represent the bottom of the hole, by, for example, expanding or contracting the shape. The latter may be performed by the auxiliary module 106 of the PMC 102.

[0093] As shown by way of non-limiting example, the contour of shape 1003 (which may be found by a function known per se after a 2means operation) may include a redundant portion 1003 which is unlikely to be part of the shape representing the bottom of the hole, and which is therefore removed, resulting in a corrected shape and contour (e.g., 1004).

[0094] As yet another example, a shape is shrunk (possibly using morphological operations), e.g., as a result of a user decision based on the data (e.g., removing more pixels that should have been classified as "bright," or vice versa in the case of expansion).

[0095] It should be noted that the examples shown in the present disclosure (such as, for example, input and generated images, various processing techniques, numerical values, etc.) are shown for exemplary purposes only and should not be considered to limit the present disclosure in any way. Other suitable examples / implementations may also be used in addition to or in place of the above examples.

[0096] One of the advantages of certain embodiments of the subject matter described herein is providing data indicative of the accurate shape of the bottom of a hole in a semiconductor sample, thereby facilitating more accurate inspection and improved processes, such as semiconductor manufacturing processes.

[0097] In the detailed description, numerous specific details are set forth in order to provide a thorough understanding of the present disclosure. However, those skilled in the art will appreciate that the subject matter currently disclosed can be practiced without these specific details. In other instances, well-known methods, procedures, components, and circuits are not described in detail to avoid obscuring the subject matter currently disclosed.

[0098] Unless otherwise expressly stated, as will be apparent from the discussion, it should be understood that throughout the specification discussion, terms such as "examine," "obtain," "provide," "acquire," "identify," "determine," "locate," "process," "segment," "utilize," "apply," "generate," "execute," "measure," "correlate," "classify," "segment," etc., refer to computer actions and / or processes that manipulate and / or transform data into other data, the data being represented as physical (such as electronic) quantities and / or the data representing physical objects. The term "computer" should be broadly interpreted to include, for example, reference to Figure 1 Any kind of hardware-based electronic device described having data processing capabilities.

[0099] The processor referred to in the present disclosure may represent one or more general processing devices, such as a microprocessor, a central processing unit, etc. More specifically, the processor may be a complex instruction set computing (CISC) microprocessor, a reduced instruction set computing (RISC) microprocessor, a very long instruction word (VLIW) microprocessor, a processor implementing other instruction sets, or a processor implementing a combination of instruction sets. The processor may also be one or more special-purpose processing devices, such as an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA), a digital signal processor (DSP), a network processor, etc. The processor is configured to execute instructions for performing the operations and steps discussed herein.

[0100] The memory mentioned in this article may include main memory (e.g., read-only memory (ROM), flash memory, dynamic random access memory (DRAM) (such as synchronous DRAM (SDRAM) or bus DRAM (RDRAM)), etc.) and static memory (e.g., flash memory, static random access memory (SRAM), etc.).

[0101] The terms "non-transient memory" and "non-transient storage medium" used herein should be broadly interpreted as covering any volatile or non-volatile computer memory suitable for the subject matter currently disclosed. These terms should be understood 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 groups of instructions. These terms should also be understood to include any medium that can store or encode a group of instructions for execution by a computer and cause the computer to execute any one or more methods in the method disclosed herein. Therefore, these terms should be understood to include, but are not limited to, read-only memory ("ROM"), random access memory ("RAM"), disk storage media, optical storage media, flash memory devices, etc.

[0102] The term "sample" used in this specification should be broadly interpreted to cover any kind of physical object or substrate, including wafers, masks, reticles, and other structures, combinations and / or parts thereof used to manufacture semiconductor integrated circuits, magnetic heads, flat panel displays, and other semiconductor manufactured products. Samples are also referred to herein as semiconductor samples and can be produced by manufacturing equipment that performs the corresponding manufacturing process.

[0103] The term "inspection" used in this specification should be broadly interpreted to cover any kind of operation related to various types of defect detection, defect review and / or defect classification, segmentation and / or metrology operations during and / or after the sample manufacturing process. Inspection is provided using a non-destructive inspection tool during or after the manufacture of the sample to be inspected. As a non-limiting example, the inspection process may include runtime scanning (in a single or multiple scans), imaging, sampling, detection, review, measurement (for example, including measurement of characteristics of the sample hole and the bottom of the hole), classification and / or other operations provided about the sample or part of the sample using the same or different inspection tools. Similarly, the inspection can be provided before the sample to be inspected is manufactured, and may include, for example, generating (multiple) inspection schemes and / or other setup operations. It should be noted that unless otherwise explicitly stated, the term "inspection" or its derivatives used in this specification are not limited in terms of the resolution or size of the inspection area. As non-limiting examples, various non-destructive inspection tools include scanning electron microscopes (SEMs), atomic force microscopes (AFMs), optical inspection tools, etc.

[0104] The term "shape characterization model" used in this specification should be interpreted broadly to cover any kind of model that characterizes a geometric shape, for example, the model is obtained (e.g., by a user or a computer) from a priori design, (multiple) drawings and is used to determine the sought shape or shapes (e.g., the shape of the bottom of a hole). A priori in this context means that a model is prepared first and then used to determine the sought shape.

[0105] In the structure of vertical NAND, memory cells are stacked on top of each other, and each memory cell is activated by its corresponding word line (WL). The connections to the WL are arranged in a stepped structure, thereby independently enabling access to each WL layer. Contact holes are etched through silicon oxide and fall on the tungsten wire steps, thereby establishing connections between word lines in different layers. The term HAR should be interpreted as the ratio between the width and length of the hole or its equivalent. The term through-hole should be interpreted as a vertical hole included between two or more adjacent layers for performing electrical connections.

[0106] It should be understood that, unless otherwise specifically stated, certain features of the presently disclosed subject matter described in the context of separate embodiments may also be provided in combination in a single embodiment. Conversely, various features of the presently disclosed subject matter described in the context of a single embodiment may also be provided individually or in any suitable subcombination. In the following detailed description, numerous specific details are set forth in order to provide a thorough understanding of the method and apparatus.

[0107] It should be noted that, according to certain embodiments, the order of the computational stages described herein with reference to the accompanying drawings is not necessarily binding. For example, the order of the steps may be changed, the steps may be modified or deleted, and / or other steps may be added in place of or in addition to the steps disclosed herein.

[0108] It is to be understood that the disclosure is not limited in its application to the details set forth in the description contained herein or illustrated in the drawings.

[0109] It should also be understood that the system according to the present disclosure can be implemented at least in part on a suitably programmed computer. Likewise, the present disclosure contemplates a computer program readable by a computer for performing the method of the present disclosure. The present disclosure further contemplates a non-transitory computer-readable memory tangibly embodying an instruction program executable by a computer for performing the method of the present disclosure.

[0110] The present disclosure is capable of other embodiments and can be practiced or performed in various ways. Therefore, it should be understood that the wording and terminology employed herein are for descriptive purposes and should not be considered restrictive. Therefore, those skilled in the art will appreciate that the concepts upon which the present disclosure is based can be easily used as the basis for designing other structures, methods and systems for achieving the several purposes of the presently disclosed subject matter.

[0111] It will be easily appreciated by those skilled in the art that various modifications and changes may be applied to the embodiments of the present disclosure described above without departing from the scope of the present disclosure as defined by the accompanying claims.

Claims

1. A system for inspecting a semiconductor sample, the semiconductor sample comprising a plurality of layers at respective different depths, and a plurality of holes, each hole having a top portion at a surface of the sample and a bottom portion received in one of the layers, the system comprising a processor and memory circuitry (PMC) configured to: a. providing an inspection image indicating at least one of the holes in the hole; b. Processing at least one hole image in the inspection image, comprising: i. segmenting the inspection image and determining data indicating the contour of the top portion of the hole; ii. segmenting said inspection image or a derivative thereof and determining data indicative of an outline of a shape of said bottom of said hole enclosed within said outline of said top of said hole; wherein said segmentation comprises classifying said image into at least two clusters of pixels, each cluster of pixels having pixel values ​​discernible relative to the other clusters of pixels, and determining said data indicative of said outline of said shape based on said pixel values ​​of one of said clusters; Inspection of the sample is thereby facilitated, including utilizing at least the data indicative of the contour of the shape of the bottom of the hole for measuring at least a property of the bottom of the hole.

2. The computerized system of claim 1, wherein the semiconductor sample is a step 3D NAND, and wherein the hole is a high aspect ratio (HAR) via, and wherein the bottom portion is contained in a tungsten layer of the sample.

3. The computerized system of claim 1, wherein the segmentation specified in (b)(ii) does not use a shape characterization model.

4. The computerized system of claim 1, wherein the PMC is configured in (ii) to segment the image, comprising: Applying at least two iterations of the KMeans function, wherein K>2 in the first iteration, K decreases in each subsequent iteration, and K=2 in the last iteration, thereby producing an image having at least two pixel clusters, wherein the at least two pixel clusters are characterized by pixel values ​​that are distinguishable relative to other pixel clusters, wherein the data indicating the outline of the shape is determined based on the pixel values ​​of one of the clusters.

5. The computerized system of claim 4, comprising two iterations, wherein K=3 in a first iteration and K=2 in a second iteration.

6. The computerized system of claim 4, wherein the PMC is further configured to apply a smoothing operation in at least two of the iterations.

7. The computerized system of claim 4, wherein the PMC is further configured to apply at least one morphological operation after a last iteration for obtaining data indicative of a different contour representing a corrected shape of the bottom of the hole.

8. A computerized system as described in claim 4, wherein the PMC is further configured to determine data indicative of the outline of the top portion of the hole such that peripheral pixels are excluded and then apply the at least two iterations; wherein the peripheral pixels are pixels at the periphery of the hole having pixel values ​​indicating brighter pixels compared to pixels near the peripheral pixels.

9. The computerized system of claim 1, wherein the shape characterization model characterizes a geometric shape obtained at least from a priori design or drawing and used to determine the sought one or more shapes.

10. The computerized system of claim 1, further comprising measuring at least one of the following properties of the top or bottom of the hole: CD - fit the contour to an ellipse and calculate 2*(major axis*minor axis) 1 / 2 ; Area - fits the contour to an ellipse and calculates the area of ​​the ellipse; GL——Grayscale statistics, including mean and standard deviation; The tilt is between the top and bottom COG (center of gravity); Direction – the angle between the vector connecting the COG and the X-axis. Length – The Euclidean distance between COGs.

11. A computerized method for inspecting a semiconductor sample comprising a plurality of layers at respective different depths, and a plurality of holes, each hole having a top portion at a surface of the sample and a bottom portion received in one of the layers, the method comprising: a. providing an inspection image indicating at least one of the holes in the hole; b. Processing at least one hole image in the image, comprising: i. segmenting the inspection image and determining data indicating the contour of the top portion of the hole; ii. segmenting the inspection image or a derivative thereof and determining data indicative of a contour of a shape enclosed within the contour of the top of the hole without using a shape characterization model; Inspection of the sample is thereby facilitated, including utilizing at least the data indicative of the contour of the shape of the bottom of the hole for measuring at least a property of the bottom of the hole.

12. The computerized method of claim 11, wherein the semiconductor sample is a step 3D NAND, and wherein the hole is a high aspect ratio (HAR) via, and wherein the bottom portion is received in a tungsten layer of the sample.

13. A computerized method as described in claim 11, wherein said (ii) includes segmenting the image, including classifying the image into at least two pixel clusters, each pixel cluster having pixel values ​​that are distinguishable relative to the other pixel clusters, and determining the data indicating the outline of the shape based on the pixel values ​​of one of the clusters.

14. A non-transitory computer-readable storage medium tangibly embodying a program of instructions that, when executed by a computer, cause the computer to perform a method for inspecting a semiconductor sample, the semiconductor sample comprising a plurality of layers at respective different depths, and a plurality of holes, each hole having a top portion at a surface of the sample and a bottom portion received in one of the layers, the method comprising: a. providing an inspection image indicating at least one of the holes in the hole; b. Processing at least one hole image in the image, comprising: i. segmenting the inspection image and determining data indicating the contour of the top portion of the hole; ii. segmenting the inspection image or a derivative thereof and determining data indicative of a contour of a shape enclosed within the contour of the top of the hole without using a shape characterization model, Inspection of the sample is thereby facilitated, including utilizing at least the data indicative of the contour of the shape of the bottom of the hole for measuring at least a property of the bottom of the hole.

15. The non-transitory computer-readable storage medium of claim 14, wherein the semiconductor sample is a step 3D NAND, and wherein the hole is a high aspect ratio (HAR) via, and wherein the bottom portion is received in a tungsten layer of the sample.

16. A non-transitory computer-readable storage medium as described in claim 14, wherein the (ii) includes segmenting the image, including classifying the image into at least two pixel clusters, each pixel cluster having pixel values ​​that are distinguishable relative to other pixel clusters, and determining the data indicating the outline of the shape based on the pixel values ​​of one of the clusters.

17. The non-transitory computer-readable storage medium of claim 14, wherein said (ii) comprises: Segmenting the image comprises: applying at least two iterations of the KMeans function, wherein K>2 in the first iteration, K decreases in each subsequent iteration, and K=2 in the last iteration, thereby producing an image having at least two pixel clusters, wherein the at least two pixel clusters are characterized by pixel values ​​that are distinguishable relative to other pixel clusters, wherein the data indicating the outline of the shape is determined based on the pixel values ​​of one of the clusters.

18. The non-transitory computer readable storage medium of claim 17, comprising two iterations, wherein K=3 in a first iteration and K=2 in a second iteration.

19. The non-transitory computer-readable storage medium of claim 14, wherein the shape characterization model characterizes a geometric shape obtained at least from a priori design or drawing and used to determine the sought one or more shapes.

20. The non-transitory computer readable storage medium of claim 14, wherein the method further comprises measuring at least one of the following properties in the top or bottom of the hole: CD - fit the contour to an ellipse and calculate 2*(major axis*minor axis) 1 / 2 ; Area - fits the contour to an ellipse and calculates the area of ​​the ellipse; GL——Grayscale statistics, including mean and standard deviation; The tilt is between the top and bottom COG (center of gravity): Direction – the angle between the vector connecting the COG and the X-axis. Length – The Euclidean distance between COGs.

Images