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 .
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
Response to Arguments
This is in response to applicant’s amendment/response filed on June 23rd2026 which have been entered and made of record.
Applicant’s arguments regarding claim rejections under 112b for claims 5-14 have been fully considered and are persuasive. The previous rejection has been withdrawn.
Applicant’s arguments with respect to claim(s) 1-11 and 15-19 have been considered but are moot because the new ground of rejection does not rely on any reference applied in the prior rejection of record for any teaching or matter specifically challenged in the argument.
Applicant’s arguments regarding claim rejections under 35 U.S.C 103 for claims 12-14 have been fully considered and are not persuasive.
Applicant argues The Office Action acknowledged that Kitsunezuka did not teach reducing dimensionality of the plurality of attributes to the number of axes of the scatter plot(s) and relied on Matplotlib. Matplotlib's scatter documentation, however, is directed to plotting x and y data positions. It states that scatter (x, y) is a scatter plot of y versus x and treats x and y as the data positions. The cited portion of Matplotlib is a plotting API disclosure, not a teaching of dimensionality reduction of a plurality of attributes. The Office Action's present use of Matplotlib therefore conflates choosing plot coordinates with performing dimensionality reduction.
Claim 12 now recites selecting, through the user interface, a dimensionality- reduction method other than t-SNE and, in response, reducing dimensionality of the plurality of attributes to the number of axes of a scatter plot generated by the user interface. The specification expressly supports alternative dimensionality-reduction techniques. The Office Action does not show that Matplotlib teaches or suggests that missing reduction step, and it does not address the amended parent claim 1 limitations either.
Examiner respectably disagrees. Kitsunezuka teaches several other dimensional reduction models can be used to generate the scatter plot (Principal Component Analysis, Independent Component Analysis, Factor Analysis, Hierarchical Clustering, Latent Semantic Analysis, Page 7, Para. 2). The scatter plots are generated based on the monitoring data retrieved for each wafer (Page 8, Para. 6-7 to Page 9 Para. 1-2). Kitsunezuka also teaches a user can change setting information (Page 3 Para. 6). However, Kitsunezuka fails to explicitly show a user selecting a different dimensionality reduction method affecting the generation of the scatter plots axes, even though Kitsunezuka can use other dimensional reduction models. Hence, the Matplotlib reference is used to teach a user generating a scatter plot based on a different dimensionality reduction model and attribute data. As Matplotlib is a well-known data analysis tool for generating interactive visualizations and performing statistical analysis. Matplotlib can perform dimensionality reduction models and generate scatter plots (Matplotlib Page 1) based on user input through a user interface.
Applicant argues The Office Action's Davis-based theory is also insufficient against the amended claims. Davis describes a GUI for interactive inspection of modeled objects, including acquisition of image sets containing a 3D model, display of 2D and 3D views, visualization changes, operator markup, and optional manual or automatic labeling of zones on the surface of a virtual object. Davis further explains that the operator can switch among 2D and 3D views in a workflow that suits the inspection task, and that acquisition module 156 may retrieve and load image sets. That is not the same thing as Applicant's semiconductor-defect t-SNE scatter-plot interface, as recited in the claims, with a manually classified-defect 2D scatter plot, an automatically classified-defect 2D scatter plot, and pointwise reclassification between them.
The rationale for combining the references also remains insufficiently articulated. The Office Action appears to reason that, because Davis discloses an acquisition module and an interactive graphical user interface, a person of ordinary skill in the art could have used semiconductor-wafer data from Kitsunezuka and Nguyen as input to the Davis system. However, that reasoning identifies only a general possibility of combination, not a reasoned explanation why a skilled artisan would have modified the cited references to arrive at the particular reclassification operations now claimed in the specific context of scatter-plot-based semiconductor defect analysis. An obviousness rejection requires more than a showing that references could be combined; it requires articulated reasoning with a rational underpinning supporting the proposed modification. See MPEP § 2143; KSR Int'l Co. v. Teleflex Inc.; In re Kahn.
Examiner respectably disagrees. Claims 13 and 14 are directed to a user interface that can display a scatter plot of defects that were either automatically or manually classified and the user interface can be used to reclassify a select defect either automatically or manually. Kitsunezuka teaches generating 2D scatter plots of defect data (Kitsunezuka et al. Fig.5 and Nguyen et al. Fig. 3). Nguyen teaches generating 2D scatter plots of defect data, wherein the data is either classified manually (Nguyen et al. Page 2 Col. 1 Para. 1) or automatically (Nguyen et al. Fig. 2), as well as a user can select to view the defect image associated with the scatter plot(Nguyen et al. Fig. 3). Gkorou teaches classifying semiconductor wafers by a human operator/engineer (Para. 0084-0085) to train/reconfigure a model or automatically by the model (Para. 0039). Zheng teaches classifying semiconductor defects using machine learning algorithms (Zheng et al. Section: 1. Introduction Pages 1-2). All of these references fail to explicitly teach a user being able to reclassify a defect selected from the scatter plot. Thus, Davis a robust interactive graphical user interface used to inspect and analyze objects, which can be a semiconductor is referenced. Davis teaches visualizing 2D and 3D data and inspecting/marking-up the properties associated with the visualized data (2D Scatter Plot). The user can modify the 2D Visualizations inspection process (Defect Inspection) in a way that fits their workflow (Davis et al. Para. 0067). Which means a user can modify the data in a way they seek. The 2D visualizations are labeled during inspection manually or automatically (Davis et al Para. 0078). Davis’s Figure 5 shows a user can modify the type of defect present. Which could have been labeled manually or automatically. Thus, combining Kitsunezuka’s scatter plot generation, Nguyen, Gkorou, or Zheng’s classification with Davis’s interactive user interface. Results in an interactive user interface for inspecting objects visualized that can have the properties modified automatically or manually.
Regarding the remaining arguments applicant argues with respect to the amended claim language, which is fully addressed in the prior art rejections set forth below.
Claim Interpretation
Under MPEP 2143.03, "All words in a claim must be considered in judging the patentability of that claim against the prior art." In re Wilson, 424 F.2d 1382, 1385, 165 USPQ 494, 496 (CCPA 1970). As a general matter, the grammar and ordinary meaning of terms as understood by one having ordinary skill in the art used in a claim will dictate whether, and to what extent, the language limits the claim scope. Language that suggests or makes a feature or step optional but does not require that feature or step does not limit the scope of a claim under the broadest reasonable claim interpretation.
Under SuperGuide Corp. v. DirecTV Enters., Inc., 358 F.3d 870 (Fed. Cir. 2004), “the phrase ‘at least one of’ precedes a series of categories of criteria, and the patentee used the term ‘and’ to separate the categories of criteria, which connotes a conjunctive list. The district court correctly interpreted this phrase as requiring that the user select at least one value for each category; that is, at least one of a desired program start time, a desired program end time, a desired program service, and a desired program type.”, SuperGuide, 358 F.3d at 886.
In this case, applicant has presented the following categorical list, all of which must be present in order to reject the claims 1, 16, and 17:
a current image, a reference image, and a difference image.
Claim Rejections - 35 USC § 112
The following is a quotation of 35 U.S.C. 112(b):
(b) CONCLUSION.—The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention.
The following is a quotation of 35 U.S.C. 112 (pre-AIA ), second paragraph:
The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention.
Claims 1-19 are rejected under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), second paragraph, as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor (or for applications subject to pre-AIA 35 U.S.C. 112, the applicant), regards as the invention.
Claims 1, 16, and 17 state the following limitation “(e) displaying the selected defect image by repeatedly displaying first and second images of a same location one after the other”. Examiner is unsure how “first and second image of a same location” relate to “the selected defect image”. As well as it is unclear what “first and second image of a same location” is referring to. Examiner determines by broadest reasonable interpretation that “first and second image of a same location” to be any two images at the same location. Thus, the claims will be examined as best understood by the examiner.
Claims 2-15 inherit their indefiniteness from claim 1 from which they depend.
Claims 18 and 19 inherit their indefiniteness from claim 17 from which they depend.
Claims 2-15, 18, and 19 will be examined as best understood by the Examiner.
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole 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. Patentability shall not be negated by the manner in which the invention was made.
Claim(s) 1-11 and 15-19 are rejected under 35 U.S.C. 103 as being unpatentable over Kitsunezuka et al. WIPO WO 2024024633 A1 (hereinafter Kitsunezuka) in view of NPL “Combining unsupervised and supervised learning in microscopy enables defect analysis of a full 4H-SiC wafer” by Binh Duong Nguyen, Johannes Steiner, Peter Wellmann, and Stefan Sandfeld (hereinafter Nguyen) in further view of Gkorou et al. WIPO WO 2022012873 A1 (hereinafter Gkorou) in further view of NPL “Wafer Surface Defect Detection Based on Background Subtraction and Faster R-CNN” by “Jeibing Zheng and Tao Zhang (hereinafter Zheng) in further view of NPL YouTube Video “Toggle Images Using Javascript” by Chuewei Lu (hereinafter Lu).
Regarding claim 1, Kitsunezuka teaches a method of presenting (Displaying Defects for Analysing, Page 5 Para. 6 , Page 6 Para. 1) defects data (Monitoring Data that Includes Defect Data) produced by inspection of semiconductor wafers or masks, the method comprising(Page 2 Section: System Configuration Para. 3):
(a) receiving defects data(Monitoring Data that Includes Defect Data) comprising a plurality of attributes (Label Information or Monitoring Data associated with each Defect, Page 12 Para. 4) per defect (Page 2 Section: System Configuration Para. 3) ,the plurality of attributes including image attributes(Display Unit 14 Displays Various Images and Information Associated to Monitoring Data, Page 5 Para. 7 to Page 6 Para. 1) associated with at least one of a current (Displaying Defects), (Analysing Individual Differences of Defects) corresponding to the defect;
(b) using t-SNE (t-distributed Stochastic Neighbor Embedding) (Dimension Reduction Model) to embed the defects attributes(Label Information or Monitoring Data associated with each Defect) from a multi-dimensional attribute space into a two-dimensional (2D) plane(2D or 3D, Page 5 Para. 5);
and (c) displaying (Page 10 Para. 5) by a user interface (Interface Displayed by Display Unit 14, Page 5 Para. 7 to Page 6 Para. 1), the defect data embedded into the 2D plane on a 2D display (Display Unit 14, Page 5 Para. 7 to Page 6 Para. 1)as a 2D scatter plot(Scatter Diagram, Page 9 Para. 2 and Page 14, Para. 4). (Fig. 5, Fig. 13, Fig. 15) The Monitoring data can be multidimensional and can be graphed/displayed in 2D or 3D(Page 5 Para. 5);
However, Kitsunezuka fails to explicitly teach:
image attributes associated with at least one of a current image , a reference image, and a difference image corresponding to the defect;
(d) in response to selection of one defect in the 2D scatter plot, displaying, by the user interface, a selected defect image corresponding to the selected defect;
(e) displaying the selected defect image by repeatedly displaying first and second images of a same location one after the other, geometrically registered to each other, so that a defect appears to blink on and off;
and (f) receiving, through the user interface, manual classification input for the selected defect.
Kitsunezuka and Nguyen are analogous to the claimed invention because both of them are in the same field of analysing semiconductor wafer defects and displaying them through graphs.
Nguyen teaches:
image attributes(Label Information or Monitoring Data associated with each Defect, Page 12 Para. 4) associated with at least one of a current image(Images of Semiconductor Wafers include Defects, Fig.1 or Fig.3) ,
(d) in response to selection of one defect(Defect A-G) in the 2D scatter plot(Fig. 3), displaying, by the user interface, a selected defect image corresponding to the selected defect(Etch pit images associated with each Defect A-G);
and (f) receiving, through the user interface, manual classification(Nguyen et al. Page 2 Col. 1 Para. 1) input for the selected defect. Nguyen teaches the defects data can be either classified manually (Nguyen et al. Page 2 Col. 1 Para. 1) or automatically (Nguyen et al. Fig. 2).
Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Kitsunezuka’s 2D Scatter Plots to incorporate Nguyen’s Displaying Defect Images When Selecting a Point in the Scatter Plot and Manually Classifying Defect Images. Since doing so would provide the benefit of viewing data associated with each scatter point.
However, Kitsunezuka and Nguyen fails to teach:
a reference image, and a difference image corresponding to the defect
(e) displaying the selected defect image by repeatedly displaying first and second images of a same location one after the other, geometrically registered to each other, so that a defect appears to blink on and off;
Kitsunezuka, Nguyen, and Gkorou are analogous to the claimed invention because all of them are in the same field of analysing semiconductor defects.
Gkorou teaches:
a reference image(Reference Data Image 604, 606, or 608), and
(e) displaying the selected defect image(Displaying Selected Wafer Image 602 Side-by-side with Reference Data Image, Fig. 6 and Para. 0083-0086) by repeatedly displaying first(Selected Wafer Image 602) and second images(Reference Data Image 604, 606, or 608) of a same location (Location of Defect) one after the other, geometrically registered to each other (Based on Parameter Data, Para. 007-0011), The selected wafer image is a wafer with a defect and the reference data image is a wafer without the defect. The parameter data can be a common characteristic between wafers, areas, or specific features (Para. 008-0011).
Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Kitsunezuka’s 2D Scatter Plots altered by Nguyen’s Displaying Defect Images to incorporate Gkorou Selection and Displaying of Defect Images and Reference Images. Since doing so would provide the benefit of comparing defect wafers with reference images to enhance analysis of the defect and modify the model that groups the images (Para. 009).
However, Gkorou fails to teach:
a difference image
repeatedly displaying first and second images of a same location one after the other, so that a defect appears to blink on and off;
Kitsunezuka, Nguyen, Gkorou, and Zheng are analogous to the claimed invention because all of them are in the same field of analysing semiconductor defects.
Zheng teaches:
image attributes associated with at least one of a current image (Input Image (a) with Damage), a reference image (Background Image With No Damage (b)), and a difference image (Difference Image (c)) corresponding to the defect (Wafer Defect, Fig. 6); (Section: 2 Wafer Surface Defect Detection Method Pages 3-9)
However, Zheng fails to teach:
repeatedly displaying first and second images of a same location one after the other, so that a defect appears to blink on and off;
Kitsunezuka, Nguyen, Gkorou, Zheng, and Lu are analogous to the claimed invention because all of them are in the same field of analysing/displaying data.
Lu teaches:
repeatedly displaying first (Earth Image, Timestamp: 0:03) and second images (Venus Image, Timestamp: 004) of a same location (Same Location on Screen) one after the other, so that a (Difference) appears to blink on and off (Timestamps: 0:02 to 0:04); Toggling between images to spot the difference or highlight the differences in the images is a common method when analysing the difference between images. This technique is used across fields, such as analysing a variety of data such as medical, weather, satellite, or any imagery, spot the difference games, and editing tools. Lu teaches a simple method of creating this technique for a graphic user interface. Lu’s method can function with any type of image.
Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Kitsunezuka’s 2D Scatter Plots altered by Nguyen’s Displaying Defect Images, Gkorou Selection and Displaying of Defect/Reference Images Side-By-Side, and Zheng’s Generating Defect/Reference/Difference Images to incorporate Lu’s Toggling of Images. Since doing so would provide the benefit of an alternative method of toggling between images to notice the differences in the images, instead of displaying the images side-by-side. Choosing to display the images by toggling or by side-by-side is merely a design/preference choice.
Regarding claim 2, Kitsunezuka teaches the method according to claim 1 and further comprising:
(d) adding (Label Information or Monitoring Data associated with each Defect, Page 12 Para. 4) to at least one defect (Monitoring Data that Includes Defect Data): Kitsunezuka tracks if changes have happened when collecting monitoring data (Page 24, Para. 5 and Page 25, Para. 7) and updating the dimensional compression model to include the new monitoring data (Page 24, Para. 3).
While Kitsunezuka fails to explicitly teach perform steps (b) and (c) again. Updating the dimensional compression model with new data (New Monitoring Data, Page 24, Para. 5 and Page 25, Para. 7), would update the analysis using t-SNE and the scatter diagram(Page 24, Para. 3). As the scatter diagram is created based on the dimensional compression model and monitoring data. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Kitsunezuka’s Scatter Diagram and Dimension Reduction Model to incorporate Kitsunezuka’s own teaching of updating the Dimension Reduction Model to Repeat steps (b) and (c). Since doing so would provide the benefit of analysing changes over time in the monitoring data and updating the scatter diagrams with new data. (Page 26, Para. 6)
Regarding claim 3, Kitsunezuka fails to explicitly teach the method according to claim 1 wherein the attributes comprise attributes produced by input of a defect image to an image processing module trained to produce the attributes based on the defect image.
However, Nguyen teaches the method according to claim 1 wherein the attributes comprise attributes (Type of Dislocation, Components, or Features, Fig. 3 or Fig. B2) produced by input of a defect image to an image processing module (Image Processing Techniques, Section: 2.2.1 Identification of Image Regions that contain an etch pit) trained to produce the attributes based on the defect image. (Section: 1 Introduction Classical Image Analysis Methods, Fig. 3).Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Kitsunezuka Attributes to incorporate Nguyen’s Attributes based on Defect Images. Since doing so would provide the benefit of clustering image defects in various diagrams using the attributes. (Nguyen, Section: 2.2.3 Automated Clustering of Different Dislocation Types).
Regarding claim 4, Kitsunezuka fails to explicitly teach the method according to claim 1 wherein the attributes comprise attributes produced by input of a defect image to a machine learning module trained to produce the attributes based on the defect image.
However, Nguyen teaches the method according to claim 1 wherein the attributes comprise attributes(Type of Dislocation, Components, or Features, Fig. 3 or Fig. B2) produced by input of a defect image to a machine learning module (Automated Machine Learning Pipeline, Fig.2) trained to produce the attributes based on the defect image. (Section: 2 Materials and Methods) Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Kitsunezuka’s Attributes to incorporate Nguyen’s Attributes based on Defect Images Produced by a Machine Learning Module. Since doing so would provide the benefit of clustering image defects in various diagrams using the attributes. (Nguyen, Section: 2.2.3 Automated Clustering of Different Dislocation Types).
Regarding claim 5, Kitsunezuka teaches the method according to claim 1 wherein the 2D scatter plot (Scatter Diagram, Page 9 Para. 2 and Page 14, Para. 4). has an X-axis showing values along a first t-SNE axis and a Y- axis showing values along a second t-SNE axis (Fig. 5, Fig. 13, Fig. 15).
Regarding claim 6, Kitsunezuka teaches the method according to claim 1 and further comprising producing a three-dimensional (3D) (2D or 3D, Page 5 Para. 5)scatter plot by selecting three attributes, manually or automatically (Generates a Scatter Diagram Based on Displayed Data which contains Label Information or Monitoring Data associated with each Defect, Page 12 Para. 4) , and
While Kitsunezuka fails to explicitly teach a displaying the 3D scatter plot as a perspective view relative to a 3D volume The Monitoring data can be multidimensional and can be graphed/displayed in 2D or 3D(Page 5 Para. 5). Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Kitsunezuka’s User Interface that displays 2D Scatter plots to incorporate Kitsunezuka’s own teaching of Graphing/Displaying 3D Information into a 3D scatter plot as a perspective view relative to a 3D volume. Since doing so would provide the benefit of graphing and displaying 3D information as scatter diagrams, as scatter diagrams are useful for identifying correlations and detecting outliers.
Regarding claim 7, Kitsunezuka teaches the method according to claim 1 wherein the user interface displays two 2D scatter plots (Fig. 5), wherein:
the defects(Monitoring Data that Includes Defect Data) have been classified into categories; (Page 7 Section: Display Processing of Monitoring Data Para. 3 to Page 8 Para. 1 and Page 4 Para. 1) The monitoring data is classified into multiple categories, which includes defect data.
a manually classified-defect 2D scatter plot (Fig. 5, Either Scatter Plot) displays defects which have been classified manually;(Page 9 Para. 1) Both scatter plots have been classified based on the sensors present.
However, Kitsunezuka fails to teach:
and a
Nguyen teaches:
a manually classified-defect (Page 2 Col. Para. 1) Analysing defects manually or utilizing a classical image analysis method are common.
and an automatically classified-defect 2D scatter plot (Fig. 3 and Fig. B2) displays defects which have been classified by automatic classification. (Fig. 2) The defects are automatically classified utilizing machine learning based on type of dislocation, components, or features of the defect.
Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Kitsunezuka’s Categorization to incorporate Nguyen’s Automatic Machine Learning Categorization that Automatically Classifies Defects. Since doing so would provide the benefit of automating classification of defects which would accelerate the process of analysing defects. (Nguyen et al. Page 2 Col. 1 Para. 2) As well as allow someone to compare defect categorization that has been classified manually or automatic.
Regarding claim 8, Kitsunezuka teaches a user interface for creating 2D Scatter plots (Fig. 5, Fig. 13, Fig. 15) based on semiconductor monitoring data that includes defects.
Kitsunezuka fails to teach the method according to claim 1 wherein the user interface configured to display both the selected defect image and a selected difference image corresponding to the selected defect.
However, Nguyen teaches the method according to claim 1 wherein the user interface (Etch pit images associated with each Defect A-G). is configured to display both the selected defect image(Etch pit images associated with each Defect A-G, Fig. 3) and Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Kitsunezuka’s 2D Scatter Plots to incorporate Nguyen’s Displaying Defect Images When Selecting a Point in the Scatter Plot and Manually Classifying Defect Images. Since doing so would provide the benefit of viewing data associated with each scatter point.
However, Nguyen and Gkorou fails to teach:
display both the selected defect image and a selected difference image
Zhen teaches:
display both the selected defect image(Input Image (a) with Damage)and a selected difference image(Difference Image (c)) (Wafer Defect, Fig. 6); (Section: 2 Wafer Surface Defect Detection Method Pages 3-9).
Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Kitsunezuka’s 2D Scatter Plots altered by Nguyen’s Displaying Defect Images, Gkorou Selection and Displaying of Defect/Reference Images Side-By-Side to incorporate Zheng’s Generation/Displaying of Defect/Reference/Difference Images. Since doing so would provide the benefit of generating a difference image which is usually used in defect detection for product surfaces like semiconductors (Zheng et al. Section: 2.2.3 Image difference, Page 7).
Regarding claim 9, Kitsunezuka fails to teach the method according to claim 8 wherein the defect image comprises a digital image obtained from an e-beam inspection machine.
However, Nguyen teaches the method according to claim 8 wherein the defect image comprises a digital image obtained from an e-beam inspection machine (Images Produced by Scanning Electron Microscopy). (Section: 1 Introduction, Page 2 Col.1 Para. 1) Scanning Electron Microscopy is an e-beam inspection machine. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Kitsunezuka’s Sensors to include Nguyen’s Scanning Electron Microscopy. Since doing so would provide the benefit of generating images of the semiconductors wafer surface for inspection (Nguyen, Section: 1 Introduction, Page 2 Col.1 Para. 1).
Regarding claim 10, Kitsunezuka fails to teach the method according to claim 8 wherein the defect image comprises a digital image obtained from an optical inspection machine.
However, Nguyen teaches the method according to claim 8 wherein the defect image comprises a digital image obtained from an optical inspection machine(Images Produced by Optical Scanning). (Section: 1 Introduction, Page 2 Col.1 Para. 1) Optical Scanning would utilize an optical inspection machine. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Kitsunezuka’s Sensors to include Nguyen’s Optical Scanning Since doing so would provide the benefit of generating images of the semiconductors wafer surface for inspection (Nguyen, Section: 1 Introduction, Page 2 Col.1 Para. 1).
Regarding claim 11, Kitsunezuka and Nguyen, fail to teach the method according to claim 8 wherein the selected defect image is displayed by repeatedly displaying one image of the defect and one image of a same area without the defect, one after the other, in spatial alignment, to cause appearance of the defect to switch on and off.
Gkorou teaches: the method according to claim 8 wherein the selected defect image(Displaying Selected Wafer Image 602 Side-by-side with Reference Data Image, Fig. 6 and Para. 0083-0086) is displayed by repeatedly displaying one image of the defect(Selected Wafer Image 602) and one image of a same area without the defect(Reference Data Image 604, 606, or 608), (Side-by-Side), Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Kitsunezuka’s 2D Scatter Plots altered by Nguyen’s Displaying Defect Images to incorporate Gkorou Selection and Displaying of Defect Images and Reference Images. Since doing so would provide the benefit of comparing defect wafers with reference images to enhance analysis of the defect and modify the model that groups the images (Para. 009).
However, Gkorou fails to teach:
one after another in spatial alignment, to cause appearance of the defect to switch on and off.
Zheng teaches the method according to claim 8 wherein the selected defect image((Wafer Defect, Fig. 6) is displayed by repeatedly displaying one image of the defect(Input Image (a) with Damage)and one image of a same area without the defect(Background Image With No Damage (b)), (Side-By-Side), (Section: 2 Wafer Surface Defect Detection Method Pages 3-9) Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Kitsunezuka’s 2D Scatter Plots altered by Nguyen’s Displaying Defect Images, Gkorou Selection and Displaying of Defect/Reference Images Side-By-Side to incorporate Zheng’s Generation/Displaying of Defect/Reference/Difference Images. Since doing so would provide the benefit of generating a difference image which is usually used in defect detection for product surfaces like semiconductors (Zheng et al. Section: 2.2.3 Image difference, Page 7).
However, Zheng fails to teach:
one after another in spatial alignment, to cause appearance of the defect to switch on and off.
Lu teaches the method according to claim 8 wherein the selected (Toggling Between Images, Timestamps: 0:02-0:04) is displayed by repeatedly displaying one image(Earth Image, Timestamp: 0:03) f the (Venus Image, Timestamp: 004) of a same area(Same Location on Screen) (Timestamps: 0:02 to 0:04); Toggling between images to spot the difference or highlight the differences in the images is a common method when analysing the difference between images. This technique is used across fields, such as analysing a variety of data such as medical, weather, satellite, or any imagery, spot the difference games, and editing tools. Lu teaches a simple method of creating this technique for a graphic user interface. Lu’s method can function with any type of image. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Kitsunezuka’s 2D Scatter Plots altered by Nguyen’s Displaying Defect Images, Gkorou Selection and Displaying of Defect/Reference Images Side-By-Side, and Zheng’s Generating Defect/Reference/Difference Images to incorporate Lu’s Toggling of Images. Since doing so would provide the benefit of an alternative method of toggling between images to notice the differences in the images, instead of displaying the images side-by-side. Choosing to display the images by toggling or by side-by-side is merely a design/preference choice.
Regarding claim 15, Kitsunezuka teaches the method according to claim 1 wherein plurality of attributes(Label Information or Monitoring Data associated with each Defect, Page 12 Para. 4) further comprises include processing data including one or more of:
identity of a machine which produced the defect; (Chamber ID, Page 3 Para. 7 and Page 12 Para. 4-6)
date upon which the defect was produced;
time upon which the defect was produced; (Time Stamp, Page 3 Para. 6-7 and Page 12 Para. 4-6)
location of the defect on a die;
location of the defect on a wafer;
location of the defect on a mask;
identity of an inspection machine;
and identity of operator of the inspection machine.
Regarding claim 16, Kitsunezuka teaches a non-transitory computer-readable medium storing (Storage Unit 12, Page 3, Para. 4) instructions that, when executed by a processor (CPU, GPU, Quantum Processor, Page 3 Para. 3), cause the processor to perform operations of claim 1, therefore it is rejected under the same rationale as claim 1.
Regarding claim 17, Kitsunezuka teaches a system for inspecting wafers or masks, the system comprising a user interface for presenting(Displaying Defects for Analysing, Page 5 Para. 6-7 to Page 6 Para. 1) defect data(Monitoring Data that Includes Defect Data) produced by inspection of wafers or masks (Page 2 Section: System Configuration Para. 3), the user interface implementing a method of claim 1, therefore it is rejected under the same rationale as claim 1.
Regarding claim 18, Kitsunezuka teaches the system according to claim 17 and further comprising a database (Monitoring Data Storage Unit 12c, Page 3 Para. 7) for storing defect(Label Information or Monitoring Data associated with each Defect, Page 12 Para. 4) associated with the defectThe Monitoring Data Storage Unit 12c is a database, which stores monitoring data that includes defect data. (Page 2 Section: System Configuration Para. 3)
However, Kitsunezuka fails to explicitly teach defect images and defect image attributes.
Nguyen teaches defect images (Images of Semiconductor Wafers include Defects, Fig.1 or Fig.3) and defect image attributes(Type of Dislocation, Components, or Features, Fig. 3 or Fig. B2). Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Kitsunezuka’s Attributes to incorporate Nguyen’s Attributes based on Defect Images. Since doing so would provide the benefit of clustering image defects in various diagrams using the attributes. (Nguyen, Section: 2.2.3 Automated Clustering of Different Dislocation Types).
Regarding claim 19, Kitsunezuka teaches the system according to claim 17 and further comprising a database(Monitoring Data Storage Unit 12c, Page 3 Para. 7) for storing non-image attributes(Label Information or Monitoring Data associated with each Defect, Page 12 Para. 4) associated with the defect
However, Kitsunezuka fails to explicitly teach defect images.
Nguyen teaches defect images (Images of Semiconductor Wafers include Defects, Fig.1 or Fig.3). Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Kitsunezuka’s Attributes to incorporate Nguyen’s Attributes based on Defect Images. Since doing so would provide the benefit of clustering image defects in various diagrams using the attributes. (Nguyen, Section: 2.2.3 Automated Clustering of Different Dislocation Types).
Claim(s) 12 is rejected under 35 U.S.C. 103 as being unpatentable over Kitsunezuka et al. WIPO WO 2024024633 A1 (hereinafter Kitsunezuka) in view of NPL “Combining unsupervised and supervised learning in microscopy enables defect analysis of a full 4H-SiC wafer” by Binh Duong Nguyen, Johannes Steiner, Peter Wellmann, and Stefan Sandfeld (hereinafter Nguyen), in view of Gkorou et al. WIPO WO 2022012873 A1 (hereinafter Gkorou), in view of NPL “Wafer Surface Defect Detection Based on Background Subtraction and Faster R-CNN” by “Jeibing Zheng and Tao Zhang (hereinafter Zheng), and YouTube Video “Toggle Images Using Javascript” by Chuewei Lu (hereinafter Lu) in further view of NPL Matplotlib 3.9.0 Documentation for Creating Scatter Plots by Matplotlib (hereinafter Matplotlib).
Regarding claim 12, Kitsunezuka teaches the method according to claim 1 and further comprising selecting(User Changing Setting Information, Page 3 Para. 6), through the user interface, a dimensionality reduction method other than t-SNE(Other Dimensional Reductions Models – Principal Component Analysis, Independent Component Analysis, Factor Analysis, Hierarchical Clustering, Latent Semantic Analysis, Page 7, Para. 2) and, in response to the selection, (Scatter Plot is Generated by the Interface to be Viewed). The user can utilize several different analysis methods to reduce the dimensions.
However, Kitsunezuka fails to teach reducing dimensionality of the plurality of attributes to the number of axes of a scatter plot generated by the user interface.
Kitsunezuka, Nguyen, Gkorou, Zheng, Lu, and Matplotlib are analogous to the claimed invention because all of them are in the same field of analysing/displaying data.
Matplotlib teaches to reduce dimensionality of the plurality of attributes to the number of axes of the scatter plot(s). (Page 1, X and Y Values) Matplotlib is a well-known data analysis tool for generating interactive visualizations and performing statistical analysis using the popular programming language Python. When creating scatter plots in Matplotlib a user can choose the axes (X and Y values) and the data associated with the scatter plot. Thus, if a user wanted to create a scatter plot utilizing defect data that has been categorized. They could easily choose the axes to be certain defect attributes.
As stated above, Kitsunezuka teaches several other dimensional reduction models can be used to generate the scatter plot (Principal Component Analysis, Independent Component Analysis, Factor Analysis, Hierarchical Clustering, Latent Semantic Analysis, Page 7, Para. 2). The scatter plots are generated based on the monitoring data retrieved for each wafer (Page 8, Para. 6-7 to Page 9 Para. 1-2). Kitsunezuka also teaches a user can change setting information (Page 3 Para. 6). However, Kitsunezuka fails to explicitly show a user selecting a different dimensionality reduction method affecting the generation of the scatter plots axes, even though Kitsunezuka can use other dimensional reduction models. Hence, the Matplotlib reference is used to teach a user generating a scatter plot based on a different dimensionality reduction model and attribute data. As Matplotlib is a well-known data analysis tool for generating interactive visualizations and performing statistical analysis. Matplotlib can perform dimensionality reduction models and generate scatter plots (Matplotlib Page 1) based on user input through a user interface.
Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Kitsunezuka’s 2D Scatter Plots altered by Nguyen’s Displaying Defect Images, Gkorou Selection and Displaying of Defect/Reference Images Side-By-Side, Zheng’s Generating Defect/Reference/Difference Images, and Lu’s Toggling of Images to incorporate MatplotLib’s Scatter Plots Creation Tool that can Select Any Data as the Axes. Since doing so would provide the benefit of increasing the flexibility when generating scatter plots and enhance the user’s ability when analysing data.
Claim(s) 13-14 are rejected under 35 U.S.C. 103 as being unpatentable over Kitsunezuka et al. WIPO WO 2024024633 A1 (hereinafter Kitsunezuka) in view of NPL “Combining unsupervised and supervised learning in microscopy enables defect analysis of a full 4H-SiC wafer” by Binh Duong Nguyen, Johannes Steiner, Peter Wellmann, and Stefan Sandfeld (hereinafter Nguyen), in view of Gkorou et al. WIPO WO 2022012873 A1 (hereinafter Gkorou), in view of NPL “Wafer Surface Defect Detection Based on Background Subtraction and Faster R-CNN” by “Jeibing Zheng and Tao Zhang (hereinafter Zheng), and YouTube Video “Toggle Images Using Javascript” by Chuewei Lu (hereinafter Lu) in further view of Davis et al. U.S. Patent Application Publication 20080247636 A1 (hereinafter Davis).
Regarding claim 13, Both Kitsunezuka and Nguyen teach Generating 2D Scatter plot(s) based on semiconductor defect analysis (Kitsunezuka et al. Fig.5 and Nguyen et al. Fig. 3) and the utilization of a user interface to create said 2D Scatter plots(s). Nguyen teaches the defects data can be either classified manually (Nguyen et al. Page 2 Col. 1 Para. 1) or automatically (Nguyen et al. Fig. 2).
However, Kitsunezuka, Nguyen, Gkorou, Zheng, and Lu fail to explicitly teach the method according to claim 7 wherein the user interface enables a user to select one defect in the automatically classified-defect 2D scatter plot and classify the defect manually.
Kitsunezuka, Nguyen, Gkorou, Zheng, Lu, and Davis are analogous to the claimed invention because all of them are in the same field of analysing/displaying data.
Davis teaches the method according to claim 7 wherein the user interface (Para. 0006-0007 and 0040-0041) enables a user to select one defect in the automatically classified-defect 2D scatter plot(Fig. 5 - 2D Visualization) and classify the defect manually. (Scatter plot(s) are 2D Visualizations of data. The user can modify the 2D Visualizations inspection process in a way that fits their workflow, Para. 0067. As well as the 2D visualizations can be labeled during inspection manually or automatically, Para. 0078)
Davis’s robust interactive graphical user interface used to inspect and analysis objects, could be used for semiconductors that have been scanned/modeled. One of ordinary skill in the art would be able to utilize the semiconductor wafer data collected in Kitsunezuka and Nguyen as input into Davis’s Interactive Inspection/Analysis Tool. As Davis’s has an acquisition module 156 which can acquire stored data or data from other devices (Davis et al. Para. 0041), and Kitsunezuka stores its monitoring data (Defect data) into a database (Kitsunezuka et al. Monitoring Data Storage Unit 12c, Page 3 Para. 7) and has a communication unit 13 that can transmit/receive data (Kitsunezuka et al. Page 5 Para. 6).
As stated above, Kitsunezuka teaches generating 2D scatter plots of defect data (Kitsunezuka et al. Fig.5 and Nguyen et al. Fig. 3). Nguyen teaches generating 2D scatter plots of defect data, wherein the data is either classified manually (Nguyen et al. Page 2 Col. 1 Para. 1) or automatically (Nguyen et al. Fig. 2), as well as a user can select to view the defect image associated with the scatter plot(Nguyen et al. Fig. 3). Gkorou teaches classifying semiconductor wafers by a human operator/engineer (Para. 0084-0085) to train/reconfigure a model or automatically by the model (Para. 0039). Zheng teaches classifying semiconductor defects using machine learning algorithms (Zheng et al. Section: 1. Introduction Pages 1-2). All of these references fail to explicitly teach a user being able to reclassify a defect selected from the scatter plot. Thus, Davis a robust interactive graphical user interface used to inspect and analyze objects, which can be a semiconductor is referenced. Davis teaches visualizing 2D and 3D data and inspecting/marking-up the properties associated with the visualized data (2D Scatter Plot). The user can modify the 2D Visualizations inspection process (Defect Inspection) in a way that fits their workflow (Davis et al. Para. 0067). Which means a user can modify the data in a way they seek. The 2D visualizations are labeled during inspection manually or automatically (Davis et al Para. 0078). Davis’s Figure 5 shows a user can modify the type of defect present. Which could have been labeled manually or automatically. Thus, combining Kitsunezuka’s scatter plot generation, Nguyen, Gkorou, or Zheng’s classification with Davis’s interactive user interface. Results in an interactive user interface for inspecting objects visualized that can have the properties modified automatically or manually.
Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Kitsunezuka’s 2D Scatter Plots altered by Nguyen’s Displaying Defect Images, Gkorou Selection and Displaying of Defect/Reference Images Side-By-Side, Zheng’s Generating Defect/Reference/Difference Images, and Lu’s Toggling of Images to incorporate Davis’s Graphical User Interface Inspection/Analysis Tool. Since doing so would provide the benefit of analysing/inspecting defects in semiconductors utilizing an interactive graphical user interface, which would increase the user’s ability to perform data analysis and allow manipulation of the defect data.
Regarding claim 14, Both Kitsunezuka and Nguyen teach Generating 2D Scatter plot(s) based on semiconductor defect analysis (Kitsunezuka et al. Fig.5 and Nguyen et al. Fig. 3) and the utilization of a user interface to create said 2D Scatter plots(s). Nguyen teaches the defects data can be either classified manually (Nguyen et al. Page 2 Col. 1 Para. 1) or automatically (Nguyen et al. Fig. 2).
However, Kitsunezuka, Nguyen, Gkorou, Zheng, and Lu fail to explicitly teach the method according to claim 7 wherein the user interface enables a user to select one defect in the manually classified-defect 2D scatter plot and submit the defect to automatic classification.
Davis teaches the method according to claim 7 wherein the user interface (Para. 0006-0007 and 0040-0041) enables (Fig. 5 - 2D Visualization) and submit the defect to automatic classification. (Scatter plot(s) are 2D Visualizations of data. The user can modify the 2D Visualizations inspection process in a way that fits their workflow, Para. 0067. As well as the 2D visualizations can be labeled during inspection manually or automatically, Para. 0078)
Davis’s robust interactive graphical user interface used to inspect and analysis objects, could be used for semiconductors that have been scanned/modeled. One of ordinary skill in the art would be able to utilize the semiconductor wafer data collected in Kitsunezuka and Nguyen as input into Davis’s Interactive Inspection/Analysis Tool. As Davis’s has an acquisition module 156 which can acquire stored data or data from other devices (Davis et al. Para. 0041), and Kitsunezuka stores its monitoring data (Defect data) into a database (Kitsunezuka et al. Monitoring Data Storage Unit 12c, Page 3 Para. 7) and has a communication unit 13 that can transmit/receive data (Kitsunezuka et al. Page 5 Para. 6).
As stated above, Kitsunezuka teaches generating 2D scatter plots of defect data (Kitsunezuka et al. Fig.5 and Nguyen et al. Fig. 3). Nguyen teaches generating 2D scatter plots of defect data, wherein the data is either classified manually (Nguyen et al. Page 2 Col. 1 Para. 1) or automatically (Nguyen et al. Fig. 2), as well as a user can select to view the defect image associated with the scatter plot(Nguyen et al. Fig. 3). Gkorou teaches classifying semiconductor wafers by a human operator/engineer (Para. 0084-0085) to train/reconfigure a model or automatically by the model (Para. 0039). Zheng teaches classifying semiconductor defects using machine learning algorithms (Zheng et al. Section: 1. Introduction Pages 1-2). All of these references fail to explicitly teach a user being able to reclassify a defect selected from the scatter plot. Thus, Davis a robust interactive graphical user interface used to inspect and analyze objects, which can be a semiconductor is referenced. Davis teaches visualizing 2D and 3D data and inspecting/marking-up the properties associated with the visualized data (2D Scatter Plot). The user can modify the 2D Visualizations inspection process (Defect Inspection) in a way that fits their workflow (Davis et al. Para. 0067). Which means a user can modify the data in a way they seek. The 2D visualizations are labeled during inspection manually or automatically (Davis et al Para. 0078). Davis’s Figure 5 shows a user can modify the type of defect present. Which could have been labeled manually or automatically. Thus, combining Kitsunezuka’s scatter plot generation, Nguyen, Gkorou, or Zheng’s classification with Davis’s interactive user interface. Results in an interactive user interface for inspecting objects visualized that can have the properties modified automatically or manually.
Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Kitsunezuka’s 2D Scatter Plots altered by Nguyen’s Displaying Defect Images, Gkorou Selection and Displaying of Defect/Reference Images Side-By-Side, Zheng’s Generating Defect/Reference/Difference Images, and Lu’s Toggling of Images to incorporate Davis’s Graphical User Interface Inspection/Analysis Tool. Since doing so would provide the benefit of analysing/inspecting defects in semiconductors utilizing an interactive graphical user interface, which would increase the user’s ability to perform data analysis and allow manipulation of the defect data.
Conclusion
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). 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 BRIANNA R COCHRAN whose telephone number is (571)272-4671. The examiner can normally be reached Mon-Fri. 7:30am - 5:00pm.
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, Alicia Harrington can be reached at (571) 272-2330. 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.
/BRIANNA RENAE COCHRAN/Examiner, Art Unit 2615
/ALICIA M HARRINGTON/Supervisory Patent Examiner, Art Unit 2615
Prosecution Insights