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 .
Claim Objections
Claim 1 is objected to because of the following informalities:
In line 14, “deleting ones of the samples” should read “deleting one of the samples”.
Appropriate correction is required.
Claim Interpretation
Note that according to the Federal Circuit’s 2004 Superguide v. DirecTV decision, “at least one of … and … “ requires at least one instance of each and every item listed. Claim 5 contain such limitations, however, the specification supports both a conjunctive and disjunctive interpretation (see “and/or” in paragraph 50). For examination purposes, the limitations be interpreted under the broader disjunctive interpretation, requiring at least one instance of any of the items listed.
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.
Claim 1-20 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.
Claim 1 recites “Obtaining, for each of the plurality of training patterns, effective distances which are smallest values of distances from each of the plurality of training patterns to a pattern that is misaligned.”, which is indefinite. It is unclear how the “smallest values” are determined. For example, the claim requires that effective distances be obtained for each of the plurality of training patterns, that is, multiple distances are obtained per training pattern. The claim further requires that these effective distances are “smallest values” of distances between each training pattern and a misaligned pattern. However, it is unclear whether the claim is meant to require that each of the training patterns be associated with multiple distances (e.g., distances to multiple misaligned patterns) from which a smallest value is selected, or whether only a single distance is computed for each training pattern, in which case the recitation of “smallest values” is redundant or unclear. Thus, one of ordinary skill in the art could not ascertain the scope of the limitation, rendering the claim indefinite. For examination purposes, the limitation will be interpreted as obtaining effective distances which are smallest values of distances from the plurality of training patterns to at least one pattern that is misaligned.
Claim 1 further recites “determining whether all the samples are aligned”, which lacks antecedent basis. It is unclear if “all the samples” is meant to refer to the initial “plurality of patterns of SEM images” set as samples or to the subset of samples which remain after the deleting limitation. For examination purposes, the limitation will be interpreted as determining whether all samples are aligned either prior or post sample deletion.
Independent claims 9 and 15 contain elements found analogous to that of claim 1. Therefore, claims 9 and 15 are rejected for the same reason of indefiniteness.
Claim 2-8, 10-14, and 16-20 are rejected as being dependent on a rejected base claim.
Examiner’s Note
Examiner notes that no prior art was applied against claims 1-20. However, claims 1-20 stand rejected under 35 U.S.C. 112(b) due to indefiniteness, and cannot be considered to be allowable subject matter because the scope of the claims is not clearly defined.
The closest prior art of record is as follows:
Nam (“Precise Pattern Alignment for Die-to-Database Inspection Based on the Generative Adversarial Network” IEEE TRANSACTIONS ON SEMICONDUCTOR MANUFACTURING, VOL. 35, NO. 3, August 2022), (hereinafter Nam) teaches a training scanning electron microscope (SEM) image selection method (Nam, “Alignment between the reference layout (or target pattern) and the corresponding scanning electron microscope (SEM) image is a crucial task for the die-to-database (D2DB) inspection in the semiconductor industry… Here, we propose a new method enabling the precise pattern alignment.” See abstract, lines 1-11, see Fig. 1) comprising:
setting a plurality of patterns of SEM images as samples (Nam, “Thus, to improve the performance of the defect review task, it is necessary to properly align the misaligned dataset. To achieve this, an alignment framework consisting of the clustering for training data selection, the translation from a SEM image to a target-like CAD image, and the alignment between a SEM image and a reference layout was developed.” Pg. 533, 2nd column, lines 2-7);
performing training on each of a plurality of training patterns to generate a conversion model between a design image pattern and the plurality of training patterns (Nam, “The translation procedure is divided into two steps. First, with the selected training data, the generator was trained adversarially to translate SEM images to target-like CAD images.”, pg. 533, 2nd column, lines 9-12, see section C. Generative Model for Image Translation);
converting the plurality of training patterns into a plurality of converted design patterns using the conversion model; comparing the plurality of converted design patterns with the design image pattern to determine misalignment therebetween; determining whether all the samples are aligned (Nam, “The second step is the inference step in which the entire SEM images in the misaligned dataset are translated to the target-like CAD images with the trained generator. The alignment procedure performs alignment between the target-like CAD image and the corresponding reference layout to obtain the alignment coordinates. Using this coordinate, the SEM images were properly located in the reference layouts. Finally, the aligned datasets were fed back to the equipment to measure the CD. Fig. 1 shows a graphical explanation of the proposed alignment framework and alignment results.”, pg. 533, lines 12-22, see Fig. 2 and sections F. Pattern Alignment and D. Alignment Results).
Nam teaches determining distances between target CAD patterns for clustering training data selection (Nam, see pg. 533, section B. Clustering for Training Data Selection), but does not teach obtaining or determining effective distances between the training patterns and a misaligned pattern. Further, Nam does not teach extracting a maximum effective distance from the effective distances or deleting samples which fall within the maximum effective distance.
Zhang (US 20220404712 A1), (hereinafter Zhang) teaches a training scanning electron microscope (SEM) image selection method (Zhang, “A method for training a machine learning model to generate a predicted measured image, the method including obtaining (a) an input target image associated with a reference design pattern, and (b) a reference measured image associated with a specified design pattern printed on a substrate, wherein the input target image and the reference measured image are non-aligned images; and training, by a hardware computer system and using the input target image, the machine learning model to generate a predicted measured image.”, see abstract) comprising:
setting a plurality of patterns of SEM images as samples; performing training on each of a plurality of training patterns to generate a conversion model between a design image pattern and the plurality of training patterns (Zhang, “FIG. 12 is a block diagram of training an image generator model to predict a measured image, according to an embodiment. In some embodiments, the image generator model 920 is a machine learning model that is trained to generate a predicted measured image for any given input target image. Training the image generator model 920 includes adjusting model parameters, such as weights and biases of the image generator model 920, such that a cost function in generating a predicted measured image is minimized. The cost function can be a measure of a difference between a predicted measured image ( e.g., an output of the image generator model 920) and a real measured image obtained (e.g., using a SEM tool) from a printed substrate.”, pg. 12, paragraph 0166, see Fig. 12);
converting the plurality of training patterns into a plurality of converted design patterns using the conversion model; comparing the plurality of converted design patterns with the design image pattern to determine misalignment therebetween; determining whether all the samples are aligned (Zhang, “The predicted measured image 925 and the measured image 320 are input to a first aligner 1110, which aligns the measured image 320 with the predicted measured image 925 using frequency domain, as described at least with reference to FIGS. 8 and 10… After the measured image 320 is aligned using frequency domain by the first aligner 1110, it is determined whether the alignment specification is satisfied at 1115.”, pgs. 11 and 12, paragraphs 0162 and 0163, see Figs. 8-10).
Zhang teaches determining distances between predicted images and reference measured images as part of a model’s cost function (Zhang, pg. 12, paragraph 0163 and pg. 14, paragraphs 0192-0193, see Fig. 17), but does not teach obtaining or determining effective distances between training patterns and a misaligned pattern. Further, Zhang does not teach extracting a maximum effective distance from the effective distances or deleting samples which fall within the maximum effective distance.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to CONNOR LEVI HANSEN whose telephone number is (703)756-5533. The examiner can normally be reached Monday-Friday 9:00-5:00 (ET).
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, Sumati Lefkowitz can be reached at (571) 272-3638. 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.
/CONNOR L HANSEN/Examiner, Art Unit 2672
/SUMATI LEFKOWITZ/Supervisory Patent Examiner, Art Unit 2672