DETAILED ACTION
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Response to Amendment
Applicant filed a Reply on 23 March 2026 including replacement Fig. 1 which overcomes the drawing objection.
Response to Arguments
Applicant's arguments filed 23 March 2026 have been fully considered but they are not persuasive.
Applicant argues that Humphris’ height image data, which includes height measurements of various areas, does not correspond to “the 2D inspection image of an inspection area of a semiconductor pattern” as recited in claim 1.
In response, a “2D inspection image” is a broad term. The plain meaning of this term is a two-dimensional array of numbers in which each value in the array represents a value such a greyscale value, a color value, or a depth/height value. Indeed, Humphris uses the term “height image data”.
Moreover, the BRI of the term “the 2D inspection image of an inspection area” includes 2D gray scale images formed by optical inspection systems and scanning electron microscopes and height image data formed by Atomic Force Microscopy (AFM) which drags a physical probe (cantilever tip) across a surface to measure the height/depth of the surface. See [0067] of the instant published application and claim 14. Given that Applicant has established the equivalence of using height image data or gray level image for the 2D inspection image of an inspection area, the further arguments against Humphri’s “height-value-based approach” as being different from 2D image based segmentation are not convincing and contrary to Applicant’s own definition of the term. Moreover, the machine learning model cares not and does not distinguish between height and gray level images during the feeding/training or segmentation steps.
Applicant also argues that Humprhis does not disclose a machine learning model that identifies different features of the same structural element because Humprhis classifies image regions solely on their height values. In response, the claims merely recite the basic technique of segmenting based on height (the first and second features are defined solely using their relative heights such that any “given structural element” that may be present in the image and which includes two different heights is segmented by Humphris as claimed. Moreover, there is no claim element directed to, for example, labeling the “given structural element” in the training data such that the machine learning model can segment the “same given structural element” into the first and second height features; as such, the claims do not positively recite distinguishing features to ensure that a particular “same given structural element” is segmented. Instead, claim 1 broadly encompasses segmentation of any structural element that happens to have two different heights into the first and second segments which is what Humphris discloses.
Wang is applied to demonstrate the obviousness of extending Humphris to inspecting semiconductor specimens but Applicant focuses instead on Wang’s alleged lack of the identification by segmentation step. But Humphris is applied to disclose this identification by segmentation step, not Wang. In response to applicant's arguments against the references individually, one cannot show nonobviousness by attacking references individually where the rejections are based on combinations of references. See In re Keller, 642 F.2d 413, 208 USPQ 871 (CCPA 1981); In re Merck & Co., 800 F.2d 1091, 231 USPQ 375 (Fed. Cir. 1986).
Applicant also argues against Wang’s machine learning model as being trained using AFM height data which does not correspond to the claimed features. This argument is incorrect and contrary to the BRI established by Applicant. See above.
Claim Rejections - 35 USC § 103
The text of those sections of Title 35, U.S. Code not included in this action can be found in a prior Office action.
Claims 1-20 are rejected under 35 U.S.C. 103 as being unpatentable over
Humphris (US 2024/0096081 A1) and Wang US 2024/0203019 A1).
Claim 1
In regards to claim 1, Humphris discloses a system comprising one or more processing circuitries {processor in [0001], [0014]} configured to:
obtain a 2D inspection image of an inspection area
The plain meaning of this term is a two-dimensional array of numbers in which each value in the array represents a value such a greyscale value, a color value, or a depth/height value. Indeed, Humphris uses the term “height image data”.
Moreover, the BRI of the term “the 2D inspection image of an inspection area” includes 2D gray scale images formed by optical inspection systems and scanning electron microscopes and height image data formed by Atomic Force Microscopy (AFM) which drags a physical probe (cantilever tip) across a surface to measure the height/depth of the surface. See [0067] of the instant published application and claim 14. Given that Applicant has established the equivalence of using height image data or gray level image for the 2D inspection image of an inspection area, the further arguments against Humphri’s “height-value-based approach” as being different from 2D image based segmentation are not convincing and contrary to Applicant’s own definition of the term. Moreover, the machine learning model cares not and does not distinguish between height and gray level images during the feeding/training or segmentation steps.}, and
feed the 2D inspection image to a trained machine learning model to identify, by segmentation of the 2D inspection image, a first feature of a given structural element present in the inspection area and a second different feature of the same given structural element
{Fig. 7 illustrating feeding a new inspection image 605 into neural network 601 implementing a trained machine learning model to segment the image as per [0056]-[0061], [0064], [0067], Figs. 6A. 6B, 8C wherein the trained neural net segments the image into portions (first segment) corresponding to the feature data 403 and portions (second segments 402 that do not correspond to the feature). See also the masking process which also segments the image. See also Fig. 9 for the machine learning training process}, wherein:
the first feature is identified as a first segment S'1 of the inspection area, the first segment corresponding to a first region of the inspection area having a height profile pattern corresponding to a first height profile pattern and
the second feature is identified as a second segment S'2 of the inspection area, the second segment corresponding to a second region of the inspection area having a height profile pattern corresponding to a second height profile pattern, wherein the first height profile pattern is different from the second height profile pattern.
{See above in which the neural network segments the image into portions (first segment) corresponding to the feature data 403 and portions (second segments 402 that do not correspond to the feature) wherein the feature may be a height dimension (height profile pattern) that deviates from a mean height dimension such that the first segment S'1 has a different height profile pattern that deviates from (different from) the second segment’s S’2 height which is the mean height dimension}
Although Humphris obtains an inspection image representative of 2D information of an inspection area, the sample 7 being inspected is not specified as a semiconductor specimen. Nevertheless, Humphris’ inspecting sample 7 is functionally consistent with inspecting a semiconductor specimen particularly as broadly recited therein and because Humphris seeks to identify microscopic height features on the sample which notably deviates from a desired height dimension as per [0043] which is highly analogous to the instant invention’s identification of microscopic height features on a semiconductor sample.
Wang is a highly analogous reference from the same field of semiconductor metrology and machine learning and solves the same or reasonably pertinent problem of determining height (3D) characterizations of a semiconductor based on a 2D image. See abstract, technical field, [0001]-[0002] and cites below. It is noted that molybdenum disulfide is the specimen being examined and that this material is a well-known semiconductor and is used, for example, to construct nano-transistors as per [0002].
Wang also teaches a system comprising one or more processing circuitries configured to:
-obtain a 2D inspection image of an inspection area of a semiconductor specimen
- {Fig. 1 optical image acquisition, [0006] using a microscope, [0030]}, and
- feed the 2D inspection image to a trained machine learning model to identify, by segmentation of the 2D inspection image, a first feature of a given structural element present in the inspection area and a second different feature of the same given structural element
{see Fig. 1, machine learning model training by a dataset including color features and height data} including outputting a 3D image having segments (e.g. first and second pixel regions) which having first and second (different) height profile patterns, [0008]-[0012], [0031]-[0036]} and
the first feature is identified as a first segment S'1 of the inspection area, the first segment corresponding to a first region of the inspection area having a height profile pattern corresponding to a first height profile pattern and
the second feature is identified as a second segment S'2 of the inspection area, the second segment corresponding to a second region of the inspection area having a height profile pattern corresponding to a second height profile pattern, wherein the first height profile pattern is different from the second height profile pattern
{e.g. different height features in the 3D model characterizing the sample as per above and [0020]-[0023]}.
It would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains to have modified Humphris which already obtains a 2D inspection image of an inspection area such that the sample being inspected is a semiconductor specimen as taught by Wang because inspecting sample 7 is functionally consistent with inspecting a semiconductor specimen particularly as broadly recited therein and because Humphris seeks to identify microscopic height features on the sample which notably deviates from a desired height dimension as per [0043] which is highly analogous to the instant invention’s and Wang’s identification of microscopic height features on a semiconductor sample; because there is a reasonable expectation of success; and/or because doing so merely combines prior art elements according to known methods to yield predictable results.
Claim 2
In regards to claim 2, Humphris discloses the system configured to, during run-time examination of the specimen, use the trained machine learning model to determine, in the inspection image, a plurality of segments corresponding to different features of interest of the same given structural element present in the inspection area.
{see above including, e.g., different height features of interest of the same given structural element present in the inspection area as per above and [0020]-[0023]}.
Claim 3
In regards to claim 3, Humphris discloses wherein the features of interest, or data informative thereof, have been used in training of the machine learning model {see above cites for claim 1 wherein the features of interest (height data in the height image) have been used in training the ML model}.
Claim 4
In regards to claim 4, Humphris discloses the system configured to:
for each given structural element of a plurality of structural elements present in the inspection area, use the trained machine learning model to determine segments of the inspection image corresponding to different features of interest of said given structural element, thereby obtaining a set of segments, and use the set of segments to determine metrology data informative of the plurality of the structural elements {the neural network determines features and portions of features in the structural elements present in the inspection area (e.g. columns or wells as per [0043], [0051]-[0053]) to determine segments of the inspection image corresponding to different features of interest (e.g. defective, non-defective) thereby obtaining a set of segments (see claim 1 cites) and use the set of segments to determine metrology data (e.g. height data), [0064]}.
Claim 5
In regards to claim 5, Humphris discloses wherein the trained machine learning model is operative to identify, in the 2D inspection image of the inspection area, the first feature and the second feature of the same given structural element
Wang teaches wherein the trained machine learning model is operative to determine height data for identifying, in the 2D inspection image of the inspection area, the first feature and the second feature of the same given structural element, without receiving 3D information on the inspection area
{see above cites and [0023] clarifying that once the machine learning model has been trained it can determine height data for segmenting without receiving 3D information on the inspection area (without AFM providing the height data)}.
It would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains to have modified Humphris which already discloses wherein the trained machine learning model is operative to identify, in the 2D inspection image of the inspection area, the first feature and the second feature of the same given structural element such that this segmenting is performed without receiving 3D information on the inspection area as taught by Wang because there is a reasonable expectation of success; and/or because doing so merely combines prior art elements according to known methods to yield predictable results.
Claim 6
In regards to claim 6, Humphris discloses wherein the machine learning model has been trained using training images, wherein each training image has been segmented into a first segment S’1 corresponding to the first segment S'1 of the inspection image and a second segment S2 corresponding to the second segment S'2 of the inspection image
{Fig. 7 illustrating feeding a new inspection image 605 into neural network 601 implementing a trained machine learning model to segment the image as per [0056]-[0061], [0064], [0067], Figs. 6A. 6B, 8C wherein the trained neural net segments the image into portions (first segment) corresponding to the feature data 403 and portions (second segments 402 that do not correspond to the feature). See also the masking process which also segments the image. See Fig. 9, [0068] for the machine learning training process in which the height image patches correspond to the segmented training images}.
Claim 7
In regards to claim 7, Humphris discloses wherein the machine learning model has been trained using data informative of the first height profile pattern and of the second height profile pattern {Fig. 9, [0068] for the machine learning training process in which the model has been trained using data informative of the first height profile pattern and of the second height profile pattern (height image patches)}.
Claim 8
In regards to claim 8, Humphris discloses wherein data informative of the first height profile pattern and of the second height profile pattern includes label data comprising, for each given training image of a plurality of training images used to train the machine learning model, at least one segment of the given training image with a height profile pattern corresponding to the first height profile pattern, and at least another segment of the given training image with a height profile pattern corresponding to the second height profile pattern {Fig. 9, [0068] wherein the height measurements 801 sever as label data}.
Claim 9
In regards to claim 9, Humphris discloses wherein the data informative of the first height profile pattern and of the second height profile pattern have been obtained using three- dimensional data informative of one or more areas of one or more semiconductor specimens, wherein the three-dimensional data have been acquired by an examination tool {see Fig. 1, [0027]-[0039] scanning probe microscopy system}.
Claim 10
In regards to claim 10, Humphris discloses wherein the first height profile pattern and the second height profile pattern each correspond to a height profile pattern of a characteristic feature of said same given structural element present in the inspection area {see above wherein the first segment S1’ corresponds to a first (defective, varies from mean height) feature of a given structural element and the second segment S’2 corresponds to a second (non-defective, at mean height) different feature of the same given structural element}.
Claim 11
In regards to claim 11, Humphris discloses wherein at least one of the first feature or the second feature is informative of at least one of: a foot of an element present in the inspection area, a slope of an element present in the inspection area, an edge of an element present in the inspection area, a round edge of an element present in the inspection area, a top edge of an element present in the inspection area {see above cites for claim 1 wherein the segment(s)/features are at least broadly “informative” of a slope, edge and top edge of an element present in the inspection area}.
Claim 12
In regards to claim 12, Humphris discloses wherein the machine learning model has been trained using, for each given area of a plurality of areas of at least one semiconductor specimen {[0060]-[0064]}:
- a given image representative of 2D information of the given area acquired by an examination tool
{Fig. 9 providing height image data step 801, [0043]-[0045], [0068] wherein the height image data 201 includes height measurements of corresponding areas of the surface of the sample, with each height measurement being represented by a pixel and wherein the height image data 201 is “representative” of the underlying 2D information/structure of the sample 7},
- given label data informative of a segmentation of the given image into at least a first segment Si and a second segment S2 {Fig. 9, [0068] wherein the height measurements 801 sever as label data for each of the height image patches}, wherein:
o the first segment Si corresponds to a first region of the given area which has a height profile corresponding to the first height profile pattern, and o the second segment S2 corresponds to a second region of the given area which has a height profile corresponding to the second height profile pattern {see [0068] wherein the height image patches include at least a first and second region with corresponding height profiles as claimed}.
Claim 13
In regards to claim 13, Humphris discloses wherein, for each given area, the given label data has been obtained using an image representative of 3D information of the given area acquired by an examination tool
{Fig. 9, [0068] wherein the height measurements 801 sever as label data. For examination tool see Fig. 1, [0027]-[0039] scanning probe microscopy system}.
Claim 14
In regards to claim 14, Humphris discloses wherein the image representative of 3D information of the given area has been acquired by an Atomic Force Microscope or a Scanning Transmission Electron Microscope
{Fig. 1, [0027]-[0039] scanning probe microscopy system is an atomic force microscope using a cantilever}.
Claim 15
In regards to claim 15, Humphris discloses configured to use at least one of the first segment S'1 or the second segment S'2 to determine metrology data informative of the inspection area {see above wherein the determination of height features and height defects are metrology data informative of the inspection area as claimed}.
Independent Claim 16
In regards to claim 16, Humphris discloses a method comprising, by one or more processing circuitries {processor in [0001], [0014]}:
- obtaining, for each given area of a plurality of areas of a
{Fig. 9 including providing height image data step 801, [0043]-[0045], [0068] wherein the height image data 201 includes height measurements of corresponding areas of the surface of the sample, with each height measurement being represented by a pixel and wherein the height image data 201 is “representative” of the underlying 2D information/structure of the sample 7
The plain meaning of this term is a two-dimensional array of numbers in which each value in the array represents a value such a greyscale value, a color value, or a depth/height value. Indeed, Humphris uses the term “height image data”.
Moreover, the BRI of the term “the 2D inspection image of an inspection area” includes 2D gray scale images formed by optical inspection systems and scanning electron microscopes and height image data formed by Atomic Force Microscopy (AFM) which drags a physical probe (cantilever tip) across a surface to measure the height/depth of the surface. See [0067] of the instant published application and claim 14. Given that Applicant has established the equivalence of using height image data or gray level image for the 2D inspection image of an inspection area, the further arguments against Humphri’s “height-value-based approach” as being different from 2D image based segmentation are not convincing and contrary to Applicant’s own definition of the term. Moreover, the machine learning model cares not and does not distinguish between height and gray level images during the feeding/training or segmentation steps.},
o given label data informative of a segmentation of the given 2D image into at least a first segment S1 and a second segment S2
{Fig. 9, [0068] wherein the height measurements 801 serve as label data}, wherein:
- the first segment Si corresponds to a first region of the given area which has a first height profile pattern, and - the second segment S2 corresponds to a second region of the given area which has a second height profile pattern, wherein the second height profile pattern is different from the first height profile pattern
{see [0068] wherein the height image patches include at least a first and second region with corresponding height profiles as claimed},
wherein the first segment Si corresponds to a first feature of a given structural element present in the given area, and the second segment S2 corresponds to a second different feature of the same given structural element
{see above wherein the first segment S1’ corresponds to a first (defective, varies from mean height) feature of a given structural element and the second segment S’2 corresponds to a second (non-defective, at mean height) different feature of the same given structural element},
- for each given area, feeding the given 2D image and the given label data to a machine learning model for its training {[0062], Fig. 9, [0068],
wherein the machine learning model is operative, after its training, to identify, by segmentation of a 2D inspection image of an inspection area, the first feature and the second feature of a same structural element present in the inspection area
{Fig. 7 illustrating feeding a new inspection image 605 into neural network 601 implementing a trained machine learning model to segment the image as per [0056]-[0061], [0064], [0067], Figs. 6A. 6B, 8C wherein the trained neural net segments the image into portions (first segment) corresponding to the feature data 403 and portions (second segments 402 that do not correspond to the feature). See also the masking process which also segments the image}.
Although Humphris obtains an inspection image representative of 2D information of an inspection area, the sample 7 being inspected is not specified as a semiconductor specimen. Nevertheless, Humphris’ inspecting sample 7 is functionally consistent with inspecting a semiconductor specimen particularly as broadly recited therein and because Humphris seeks to identify microscopic height features on the sample which notably deviates from a desired height dimension as per [0043] which is highly analogous to the instant invention’s identification of microscopic height features on a semiconductor sample.
Wang is a highly analogous reference from the same field of semiconductor metrology and machine learning and solves the same or reasonably pertinent problem of determining height (3D) characterizations of a semiconductor based on a 2D image. See abstract, technical field, [0001]-[0002] and cites below. It is noted that molybdenum disulfide is the specimen being examined and that this material is a well-known semiconductor and is used, for example, to construct nano-transistors as per [0002].
Wang also teaches a system comprising one or more processing circuitries configured to:
- obtaining, for each given area of a plurality of areas of a semiconductor specimen a given 2D image of the given area {Fig. 1 optical image acquisition, [0006] using a microscope, [0030]}, and
- feed the inspection image to a trained machine learning model operative to determine height data of the semiconductor based on the optical image {see Fig. 1, machine learning model training by a dataset including color features and height data} including outputting a 3D image having segments (e.g. first and second pixel regions) which having first and second (different) height profile patterns, [0008]-[0012], [0031]-[0036]} and
wherein the first segment S'1 corresponds to a first feature of a given structural element present in the inspection area, and the second segment S'2 corresponds to a second different feature of the same given structural element {e.g. different height features in the 3D model characterizing the sample as per above and [0020]-[0023]}.
It would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains to have modified Humphris which already obtains a 2D inspection image of an inspection area such that the sample being inspected is a semiconductor specimen as taught by Wang because inspecting sample 7 is functionally consistent with inspecting a semiconductor specimen particularly as broadly recited therein and because Humphris seeks to identify microscopic height features on the sample which notably deviates from a desired height dimension as per [0043] which is highly analogous to the instant invention’s and Wang’s identification of microscopic height features on a semiconductor sample; because there is a reasonable expectation of success; and/or because doing so merely combines prior art elements according to known methods to yield predictable results.
Claim 17 and 20
The rejection of device system claims 5 and 1 above applies mutatis mutandis to the corresponding limitations of method claim 17 and computer readable medium claim 20 while noting that the rejection above cites to both device and method disclosures. For the computer readable storage medium storing program limitations of claim 20 see Humpris [0016], claim 16.
Claim 18
In regards to claim 18, Humphris discloses performing (i) or (ii):
i) for each given area, obtaining a given second image representative of 3D information of the given area, wherein, for each given area, the given label data is determined using the given second image, or
(ii) for each given area, obtaining a given second image representative of 3D information of the given area acquired by an Atomic Force Microscope or a Scanning Transmission Electron Microscope, wherein, for each given area, the given label data is determined using the given second image
{Fig. 1, [0027]-[0039] scanning probe microscopy system is an atomic force microscope using a cantilever that obtains a given second image including height data and the given label data is determined from this given second image. As such, Humphris discloses both options i) and ii)}.
Claim 19
In regards to claim 19, Humphris discloses wherein:
the segmentation of the given 2D image is performed by the one or more processing circuitries using the given second image, data informative of the first height profile pattern and data informative of the second height profile pattern {see mapping of claims 1 and 18 in which segmentation of the given image is performed by processing circuitry using the identified data in the claim}; or
the segmentation of the given 2D image is performed using feedback of a user and the given second image.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Young (US 2022/0043357 A1) discloses predicting height of a 3D structure using a machine learning model trained using dark and brightfield images 500, 502. See Figs. 1, 4, and 5.
THIS ACTION IS MADE FINAL. Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to Michael R Cammarata whose telephone number is (571)272-0113. The examiner can normally be reached M-Th 7am-5pm EST.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Matthew Bella can be reached at 571-272-7778. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/MICHAEL ROBERT CAMMARATA/Primary Examiner, Art Unit 2667
Prosecution Insights