US 10345247

I'll search for authoritative information on this patent, including USPTO records and any CAFC 2026 docket activity.

Let me run additional targeted searches for litigation/docket activity and the official USPTO record.

Based on the authoritative full text provided plus the search results, here is my summary. Note on scope: I limited searching to the exact number 10345247 / 10,345,247, and I explicitly did not include lookalikes that surfaced in the searches — e.g., German publication DE10345247A1 (a semiconductor-component/pressing-composition patent) and an unrelated "10345247" identifier appearing on plainsite for a Verso Paper record. Those are different documents.

---

US 10,345,247 B2 — Bibliographic Summary

---

Abstract (as issued)

An apparatus for detecting a degree of particulate contamination on a flat panel, including: an illuminator for producing a radiation beam which results in scattered radiation from its scattering by contaminants on a surface of the flat panel under test and reflected radiation from its reflection by the surface; a detector for collecting the scattered radiation, the detector having a radiation collection surface perpendicular to a normal of the surface of the flat panel under test; and a beam trimmer for separating the reflected radiation from the scattered radiation. The beam trimmer includes a first optical member disposed in correspondence with the scattered radiation (directing it to the detector) and a second optical member disposed in correspondence with the reflected radiation (directing it not to be collected by the detector).

---

The Single Independent Claim — Plain-Language Overview

Claim 1 is the only independent claim. The remaining claims (2–11) all depend, directly or indirectly, on claim 1.

Claim 1 covers a contamination-detection apparatus with three required elements:

1. An illuminator that produces a radiation beam. The beam plays two roles at the flat panel under test: it is scattered by contaminants on the panel surface (giving "scattered radiation"), and it is reflected by the panel surface itself (giving "reflected radiation").
2. A detector that collects the scattered radiation. Critically, the detector's radiation-collection surface must be oriented perpendicular to the normal of the panel surface — i.e., the detector faces straight down onto the panel, rather than sitting at the oblique collection angle used in prior designs.
3. A beam trimmer located to separate the reflected radiation from the scattered radiation. It has two parts:
   • a first optical member positioned to correspond to the scattered radiation, which routes the scattered radiation so the detector collects it; and
   • a second optical member positioned to correspond to the reflected radiation, which routes the reflected radiation away from the detector (so the specular/zero-order reflection does not corrupt the measurement).

In short: the claim is directed to a dark-field-type contamination detector in which a beam-steering "trimmer" downstream of the panel sends the unwanted specular reflection off-axis while steering the wanted scatter onto a downward-facing sensor.

Where the dependent claims add specificity

• Claim 2 (dep. 1): the scattered radiation must strike the detector's collection surface at right angles.
• Claim 3 (dep. 1): the detector comprises an imaging optical path plus a time-delay integration (TDI) camera, with scatter converged onto the TDI camera after separation.
• Claim 4 (dep. 1): defines the beam trimmer concretely as three reflectors — a first reflector facing the scattered radiation, a second reflector that redirects the scatter to the detector, and a third reflector that diverts the reflected radiation away. This is the "discrete mirrors" embodiment of FIG. 9.
• Claim 5 (dep. 4): recites geometric design equations for the three reflectors (angles θ5, θ6, and minimum lengthwise dimensions based on d1, d3, h1, h2, alpha1, theta1, beta).
• Claim 6 (dep. 5): the illuminator creates a linear illumination field of view.
• Claim 7 (dep. 1): defines the beam trimmer as a reflecting prism whose outer surface has first and second radiation-transmitting areas plus an outer radiation-reflecting area; the transmitting areas and an internal reflecting area form the first optical member, and the outer reflecting area forms the second optical member (FIG. 6/7, FIG. 8).
• Claim 8 (dep. 7): specifies the prism cross-section as a trapezoid (top and bottom surfaces parallel to the panel; a first and second slanted side), and specifies the internal double-reflection path (enter via the first slanted side, reflect off the second slanted side, then the first slanted side, exit via the bottom surface).
• Claim 9 (dep. 8): recites the prism geometry equations — θ1 = 90° − (90° − theta1); θ2 = 180° − 1.5×theta1; θ3 = 1.5×theta1; θ4 = 180° − theta1; and l = 2×d1×tan(alpha1).
• Claim 10 (dep. 8): the prism cross-section is specifically a right trapezoid.
• Claim 11 (dep. 9): linear illumination field of view.

⚠️ Literal-reading note (not auto-corrected): In the granted claim text, claim 5 renders the angle formula as "θ5=180°1.5×theta1" and the height formula as "h2=d2/tan(theta1/alpha1)." The corresponding specification text states θ5 = 180° − 1.5×theta1 and h2 = d2/tan(theta1 − alpha1). I am reporting the claim wording exactly as it appears and flagging the discrepancy rather than correcting it.

---

CAFC 2026 Docket Search — No Results Found

I searched for Federal Circuit 2026 docket activity tied to patent 10,345,247 / 10345247 and found no CAFC appeal, briefing, or docket entry referencing this patent. My CAFC-related search hits returned only unrelated 2026 appeals (e.g., AeroVironment, Salesforce/Applications In Internet Time, Contour IP v. GoPro, Enanta v. Pfizer, WIPQTUS v. Samsung) with no connection to this patent or to Amies Technology / Shanghai Micro Electronics.

Uncertainty statement: I cannot affirmatively certify from these searches that no such litigation exists — a live PACER / CourtListener / USPTO Patent Center query would be needed to be definitive. But nothing in the available results indicates any CAFC or district-court proceeding involving US 10,345,247, and no PTAB proceeding surfaced either. The only post-grant activity I found is the recorded 2025 assignment to Amies Technology Co., Ltd.

---

Sources

• Google Patents full text (primary/authoritative for this task): https://patents.google.com/patent/US10345247/en
• Justia Patents record: https://patents.justia.com/patent/10345247
• Pre-grant publication record: https://patents.justia.com/patent/20180246040 and https://uspto.report/patent/app/20180246040
• Assignee/legal-events history (as reflected in the Google Patents legal-events table)

If you'd like, I can next pull the claim charts against the three cited prior-art references most relevant to the beam-trimmer concept (e.g., US 4,594,997 "Apparatus and method for detecting defects and dust on a patterned surface," and US 2011/0019197 A1 "Scattered Light Separation" assigned to KLA-Tencor).

I'll search for litigation involving US patent 10345247 across patent litigation databases.

Let me search more targeted litigation databases and the patent's own records.

Litigation Search Report: U.S. Patent No. 10,345,247

Bottom line

I found no known U.S. litigation involving U.S. Patent No. 10,345,247 (the "10345247" patent). No district court complaint, ITC Section 337 investigation, CAFC appeal, or PTAB proceeding asserting or challenging this specific patent surfaced in my searches. I am reporting this as "no known litigation" rather than fabricating case data.

What I searched and what I checked

I ran targeted queries against litigation-oriented sources (including searches keyed to the patent number on patent-litigation and Unified Patents-related materials, and general web searches for the exact number in a litigation context). I also examined the authoritative Google Patents record for US10345247B2 (furnished in the case file).

Key observations from that record:

• The Google Patents page for US10345247B2 contains no "Litigation" section. Google Patents typically populates a litigation/assertion timeline when a patent is asserted in court. None appears here.
• The "Families Citing this family," "Citations," and "Family Cites Families" entries are patent citations (prior art and later-citing patents), not lawsuits. For example, CN115248531A and CN112540082A are later publications, not case records.
• The only noted legal events are prosecution, maintenance-fee payment, and assignment — no litigation events.

Important caveats

1. Search-tool limits. My web-search results were noisy and did not return hits from PACER, Docket Navigator, or the Unified Patents litigation caselist for this exact number. I cannot rule out that a case exists but was not indexed in the results I received. You should verify independently on Unified Patents litigation portal, PACER, and Docket Navigator using the exact number 10,345,247.
2. Exact-number discipline. I was careful not to substitute similar numbers (e.g., 10,311,073, 10,045,696, 10,621,228, RE45543) that appeared in unrelated PTAB materials during the search. None of those relate to this patent.
3. Foreign/non-U.S. enforcement. The patent is held by Amies Technology Co., Ltd. (China), having been assigned from Shanghai Micro Electronics Equipment (Group) Co., Ltd. on 2025-06-25. Its family includes CN108507909B, JP6559822B2, KR20180099567A, and TWI667465B. Enforcement, if any, would more plausibly occur in China (e.g., CN108507909B) or via the other national counterparts — which are outside the scope of a "U.S. patent 10345247" litigation search. Note the Korean counterpart KR20180099567A is recorded as "Ceased."

Patent reference data (for context)

Recommendation

If you need a definitive record, run the exact number 10,345,247 through PACER's nationwide case search, Docket Navigator, and the Unified Patents litigation caselist, and check Lex Machina for the assignees Amies Technology and Shanghai Micro Electronics Equipment. Based on the evidence available to me, however, the correct answer is: no litigation is known for this patent.

I'll verify the structured data against live sources before finalizing.

Proceedings overview

Total AIA trial proceedings on US 10,345,247: ZERO. Per the canonical structured source ("PTAB proceedings on file"), the USPTO Open Data Portal returns no IPR, PGR, or CBM involving this patent; my independent web searches (logged below) surfaced no petition, institution decision, FWD, or Federal Circuit appeal naming 10,345,247 / 10345247 either. Bottom-line posture for a defendant: the patent is un-narrowed and untested — all 11 claims stand exactly as granted on 2019-07-09, there is no FWD you can lean on, and no § 315(e)(2) estoppel has been created against anyone. The defensive posture is the opposite of "hardened by two IPRs" and the opposite of "claims 1–5 are canceled." You have a clean slate — and so does every other accused infringer, which means nobody has burned your best art yet.

---

Proceedings

No proceedings to chart. Rather than manufacture entries, here is the record:

No AIA trial proceeding identified

• Type: N/A (no IPR / PGR / CBM on file)
• Filed: N/A
• Status: N/A — the structured PTAB-proceedings block is empty; no institution decision, FWD, or termination exists to report verbatim.
• Judge panel: N/A — no panel has been designated.
• Petition grounds: N/A — no petition.
• Institution decision: N/A.
• Final Written Decision: N/A — no claim of US 10,345,247 has been canceled, confirmed, or even construed by the Board.
• Settlement / termination: N/A.
• Appeal: N/A — and consistent with the previously generated CAFC section, no Federal Circuit docket entry references this patent.
• Defensive value: Zero precedential value either way. You get no free claim construction, no free invalidity finding, and no free estoppel against the patent owner — but you also face no adverse FWD and no "the Board already rejected that art" problem.

Identifier-diligence note (do not auto-correct, but do not conflate)

Searches for the bare string "the '247 patent" returned unrelated documents that must not be folded into this record:

Non-PTAB attrition worth knowing about: the KR family member KR20180099567A is recorded as not_active / Ceased — that is a Korean national-phase abandonment, not an invalidity event, and it has no effect on US 10,345,247. The CN (CN108507909B), TW (TWI667465B), and JP (JP6559822B2) counterparts all granted.

---

Strategic summary

Canceled vs. sustained vs. untested. Nothing is canceled and nothing is sustained, because no claim has ever been before the Board. All 11 claims (claim 1 independent; claims 2–11 dependent) are UNTESTED. For a defendant, that means the full claim set — including the narrow, geometry-heavy claims 5 and 9 — is live and available to the patent owner, and that the validity record is entirely undeveloped. Two structural observations matter more than anything else here:

1. The only AIA vehicle left is IPR, and it is limited to § 102/§ 103 over patents and printed publications (35 U.S.C. § 311(b)). The transcription anomalies flagged earlier in this analysis — granted claim 5's "θ5=180°1.5×theta1" and "h2=d2/tan(theta1/alpha1)", where the specification says "180° − 1.5×theta1" and "d2/tan(theta1 − alpha1)" — are § 112 written-description/indefiniteness problems you cannot raise in an IPR. They would have had to go through a PGR, and the PGR window under 35 U.S.C. § 321(c) closed 2020-04-09 (nine months after the 2019-07-09 grant). CBM is categorically unavailable — this patent claims a contamination-detection optical apparatus, not a financial product or service, and the CBM program's petition sunset (2020-09-16) has long passed. So claims 5 and 9's formula defects are a district-court / ITC argument, not a Board argument. Preserve them.
2. No § 315(b) clock is running. The one-year IPR bar is triggered by service of a complaint alleging infringement of the patent. I found no such complaint; if you have been served, calendar the bar immediately, because it is a hard statutory deadline with no good-cause extension.

Estoppel landscape. There is no § 315(e)(2) estoppel against anyone — that provision binds a petitioner (and its privies/real parties in interest) only after an FWD, and no FWD exists. Practical consequences: (a) every prior-art ground is still available to every defendant, including the 12 references cited on the face of the patent (US 4,394,524; US 4,464,050; US 4,598,997; US 4,794,264; US 4,794,265; JPH07297248A; US 2001/0048761; US 2002/0122174; US 2007/0229833; US 2008/0174771; US 2011/0019197 A1 "Scattered Light Separation" (KLA-Tencor); US 2016/0293052) and the separately listed family-cited art, e.g. CN105372256B and CN205538567U (both assigned within the Shanghai Micro Electronics orbit — potentially useful as the patent owner's own prior work); and (b) if you file and lose, you personally lose those grounds for the rest of the case, so filing should be a considered decision, not reflexive.

Pattern signals. No petitioner has filed once, let alone twice. No evidence of a defensive aggregator (Unified Patents or similar) in the chain. The patent owner has pursued no PTAB appeal — there is nothing to appeal. The signal that does flash is corporate, not procedural: the patent was assigned from Shanghai Micro Electronics Equipment (Group) Co., Ltd. to Amies Technology Co., Ltd., effective 2025-06-25 and recorded 2025-09-17 — a recent, entity-level change of owner, into a name that appears in no PTAB, ITC, or Federal Circuit record I could find. Recently reassigned patents in an active family (with a live, fee-paid US member and four granted foreign counterparts) are a common precursor to an assertion campaign. Treat the newly-named owner as the entity to diligence, and treat the absence of any IPR as a timing fact rather than a validity fact: well-asserted patents eventually attract IPRs, and this one has not been asserted yet in any forum I can identify.

---

Recommended next steps

1. Do not cite an FWD — there isn't one. If you are drafting an invalidity contention or an IPR, there is no Board disposition to quote and no claim to "avoid" as canceled. Build the challenge from scratch against claims 1–11 as issued.
2. Confirm the negative directly, then calendar it. Re-run the check against the primary sources before you rely on my result: PTAB case tracking at https://ptacts.uspto.gov/ptacts/public-informations/ and https://www.courtlistener.com/?q=%2210%2C345%2C247%22 (CourtListener also indexes the Federal Circuit). The canonical ODP extract in this prompt is the cleanest statement of the negative, but ODP ingest can lag by days on a brand-new filing — so if you are within weeks of a trigger event, verify live.
3. If you have been served with a complaint asserting this patent, file within one year of service (35 U.S.C. § 315(b)). With no competing petition, there is no § 325(d) "same art already before the Office" risk and no General Plastic / Fintiv-style multi-petition or parallel-litigation overhang beyond the ordinary Fintiv factors — which, with no co-pending action identified, currently cut in your favor. If a district court case exists, the stay calculus is also unusually favorable: there are no prior PTAB proceedings to defer to.
4. Split your theories by forum. § 102/§ 103 grounds (e.g., US 4,598,997 "Apparatus and method for detecting defects and dust on a patterned surface"; US 2011/0019197 A1 "Scattered Light Separation," KLA-Tencor — the closest art to the claimed beam-trimmer concept) go to the PTAB. The § 112 attacks on claims 5 and 9's recitations of "θ5=180°1.5×theta1" and "h2=d2/tan(theta1/alpha1)" — irreconcilable with the specification's "180° − 1.5×theta1" and "d2/tan(theta1 − alpha1)" — must go to the district court or the ITC, because IPR jurisdiction does not reach them.
5. Diligence the new owner. Obtain the 2025-06-25 assignment instrument (recorded at Reel/Frame 072912/0466 per the Google Patents legal-events table) and identify Amies Technology Co., Ltd.'s affiliates, real parties in interest, and any parallel CN/TW/JP/KR enforcement. If this patent is part of a portfolio being monetized, the first-filer advantage on the US member is real and perishable.

Uncertainty statement (unchanged and important): I cannot affirmatively certify the absence of PTAB activity from web search alone. My conclusion rests on (a) the structured ODP-derived block stating no AIA trial proceedings on file, and (b) four web searches that returned zero petitions, decisions, or appeals tied to US 10,345,247 or to Amies Technology. A live PTAB E2E / Patent Center query would be required to make the negative airtight.

I'll reconstruct the assignment record for US 10,345,247, grounding everything in the recorded reel/frame entries from the case file and supplementing with targeted searches for correspondents and entity status.

Let me try to pin down the recording correspondents and confirm the spin-off nature of the 2025 transfer.

Ownership-Chain Reconstruction — US 10,345,247 B2

Building on the prior sections: this adds the assignment-record analysis and entity-status detail. Where the earlier "Patent summary" and "Litigation summary" already established no U.S. litigation and the 2025 assignment to Amies Technology, I do not repeat that here — I resolve what kind of transfer it was and who the parties actually are.

---

Inventors

All three are listed as joint assignors in the recorded assignment of 2018-04-06 (reel 045854/0032, effective 2018-02-28), confirming they were SMEE employees at filing and assigned their rights to the company in the ordinary course.

Unusual-pattern check: No evidence of any of the three inventors departing SMEE within 12 months of filing, and no inventor-level assignments beyond the initial one to the employer. I have no data on later inventor movements — I am not inferring a departure. There is no "inventors retained rights and later sold" pattern here; this is a clean employer-assignment chain.

---

Original assignee

• Entity on the issued patent (as printed / Google Patents metadata): Shanghai Micro Electronics Equipment Co., Ltd.
• Entity recorded as assignee on reel 045854/0032: Shanghai Micro Electronics Equipment (Group) Co., Ltd. (上海微电子装备(集团)股份有限公司)

Reconciliation note (flagged, not auto-corrected): The patent front-matter name ("Co., Ltd.") and the recorded assignee name ("(Group) Co., Ltd.") differ. Both reels and the Google legal-events table consistently point to the same corporate filing, and the "(Group)" form is the name SMEE has used across its ~110 granted U.S. patents — this reads as a corporate name/group-structure change, not a third-party transfer. I could not independently confirm the exact date of the rename from the available record.

Primary line of business: SMEE is China's leading domestic lithography equipment maker (steppers/scanners plus the metrology and stages that orbit them) — the state-owned "national team" for front-end semiconductor lithography. The portfolio record on this assignment (110 U.S. grants, 2015–2024, primary CPC class G03F) is consistent with a large operating manufacturer, not a holding shell.

Does it ship a product embodying the claims? Yes — this patent is a flat-panel/mask particulate-contamination detection apparatus, i.e., an inspection tool, which sits squarely inside SMEE's tool portfolio.

Current status: Operating. State-owned, Shanghai-based, still in business; note it is subject to U.S. export-control/sanctions pressure as of the mid-2020s (a reputational/operational fact, not an assignment fact).

---

Assignment timeline

Both records below are drawn from the authoritative legal-events table in the case file. Correspondent of record is not exposed in that table — I could not retrieve it from the sources available and will not guess it (see caveat at the end).

• 2018-02-28 (executed/effective) / recorded 2018-04-06 — Reel 045854/0032

  • Conveyance: Assignment (Assignment of Assignors' Interest)
  • Assignors: SHEN, Yongqiang; YANG, Xiaoqing; HAN, Xueshan (all three inventors)
  • Assignee: SHANGHAI MICRO ELECTRONICS EQUIPMENT (GROUP) CO., LTD.
  • Correspondent: not determinable from the available record — if a single recurring filing attorney appears here, that is where the "repeat correspondent" tell would live, so this field must be pulled directly from the Assignment Center.
  • Context: Initial inventor→employer assignment (routine employment assignment), recorded on filing.

• 2025-06-25 (executed/effective) / recorded 2025-09-17 — Reel 072912/0466

  • Conveyance: Assignment of Assignor's Interest
  • Assignor: SHANGHAI MICRO ELECTRONICS EQUIPMENT (GROUP) CO., LTD.
  • Assignee: AMIES TECHNOLOGY CO., LTD. (上海芯上微装科技股份有限公司)
  • Correspondent: not determinable from the available record (same caveat).
  • Context: Corporate spin-off / carve-out — not an acquisition, fire-sale, securitization, or transfer-to-asserter. AMIES was incorporated 2025-02-08 as an SMEE spin-off; the patent transfer followed the incorporation by roughly four months (effective 2025-06-25).

That is the complete recorded chain: two links, seven-and-a-quarter years apart, both between affirmatively identified corporate entities in the same corporate family.

---

Timeline diagram

timeline
    title Ownership of US 10345247
    2017 : Priority filed by SMEE in China
    2018 : Inventors assign rights to SMEE
         : US application filed
    2019 : US patent issued
    2025 : AMIES spun off from SMEE
         : SMEE transfers patent to AMIES

---

NPE / troll-pattern signals

1. Shell-entity transfer — NOT PRESENT. Reel 072912/0466 moves the patent to AMIES Technology Co., Ltd., which is a Chinese joint-stock company (企业类型: 其他股份有限公司) formed 2025-02-08, USCC 91310000MAEBC0CJXF, with ~600 staff, a real plant address (1169 Huadong Road, Pudong, Shanghai), and government-backed shareholders. No "IP / Holdings / Licensing / Ventures" suffix; not a Delaware/Texas single-purpose LLC; not a registered-agent mailbox. It manufactures and ships equipment.

2. Known asserter in the chain — NOT PRESENT. Neither SMEE nor AMIES appears on any public NPE list (Acacia, Marathon, Intellectual Ventures, IPNav, Wi-LAN, Mosaid/Conversant, Vringo, Pendrell, Round Rock, Spangenberg entities, etc.). No matches surfaced via Unified Patents / RPX-style directories.

3. Repeat correspondent across the chain — UNCLEAR (not assessable). The authoritative record available to me does not include the correspondent of record for either reel 045854/0032 or reel 072912/0466, so I cannot test recurrence. I am explicitly declining to infer this from the entity names. Action item: pull both frames in the Assignment Center and compare the recording attorney/firm — recurrence across a 7-year gap between a Chinese state-owned manufacturer and its own spin-off would, by itself, be weak evidence anyway.

4. Cascading transfers — NOT PRESENT. Two recorded assignments total, ~7.25 years apart (2018 and 2025), both corporate, with no chained LLCs, no shared registered-agent address, and no short-interval hand-offs.

5. Pre-litigation transfer — NOT PRESENT. There is no known litigation involving this patent (per the prior section), so the 2025 transfer cannot be a venue/standing clean-up. The transfer date (eff. 2025-06-25) is tied to the spin-off incorporation date (2025-02-08), not to any filing.

6. Bankruptcy fire-sale — NOT PRESENT. SMEE is a going concern (state-owned); reel 072912/0466 is a spin-off transfer, not an insolvency sale.

7. Privateering — NOT PRESENT. This is the inverse: SMEE moved the asset down into its own newly created operating subsidiary, which makes and ships its own tools. There is no transfer to an unrelated asserting vehicle aimed at the parent's competitors.

8. Defensive aggregator — NOT PRESENT. The chain does not terminate at RPX, AST, LOT, Unified Patents, or OIN. It terminates at an operating manufacturer.

Cross-check — product/claim alignment: AMIES publicly lists an FPD (flat-panel display) line — FPD Scanner, TP/CD measurement, photo-alignment, FMM tension equipment — and AOI inspection tools. This patent is a flat-panel contamination-detection apparatus. The transferee therefore plausibly practices the claimed subject matter, which is the opposite of the "monetization-only, no-products" NPE profile.

---

Verdict

Operating-company assertion — read as an operating-company ownership chain with no assertion on record; the "assertion" prong is not documented.

Justification (2–3 sentences): The entire recorded chain is two corporate links — reel 045854/0032 (2018-04-06) assigning inventor rights to SMEE and reel 072912/0466 (recorded 2025-09-17, effective 2025-06-25) moving the patent from SMEE to its own February-2025 spin-off, AMIES — with zero shell entities, zero known asserters, zero cascading LLC transfers, and no bankruptcy or privateering. The current assignee, AMIES Technology Co., Ltd., is a real manufacturer (≈600 staff, shipped its 500th lithography stepper in August 2025) with a flat-panel inspection product line that plausibly reads on these claims, and no litigation, ITC action, or PTAB proceeding is known — so this is an operating-company ownership structure, not an NPE. Strictly, because no suit is on record, the accurate label is operating-company ownership (currently non-asserting); I selected the closest bucket in the taxonomy and am flagging the unmet prong rather than overstating it.

Verification link: https://assignmentcenter.uspto.gov/ (search 10345247); mirror: https://assignment.uspto.gov/patent/index.html

---

Caveats / open items

1. Correspondents missing. The case file's legal-events table gives reel/frame, conveyance, parties, and dates but not the recording correspondent. I did not fabricate names. To complete the "repeat correspondent" signal, retrieve both frames (045854/0032 and 072912/0466) in the Assignment Center and capture the attorney/firm of record.
2. No contradiction with prior sections — the earlier "no known litigation" and "Amies held it" findings are consistent with this chain analysis. One nuance newly surfaced: the 2025 assignee is not a third-party acquirer but an SMEE spin-off, which materially changes the NPE read (down from "possible transfer" to "internal carve-out").
3. Entity names for AMIES/SMEE (Chinese and English) are taken from company/brand sources and the assignment record; the Chinese registered name 上海芯上微装科技股份有限公司 maps to AMIES Technology Co., Ltd. per the company's own profile.

I'll research the cited prior-art references for US 10,345,247 to build accurate citations and anticipation analysis.

Let me research the remaining cited references.

Prior-Art Analysis for US 10,345,247 B2

Building on the earlier summary and litigation sections, this analysis focuses strictly on the 12 references appearing in the "Citations (12)" / "Patent Citations (12)" table of the authoritative Google Patents record for US 10,345,247. Those are the references of record for this patent. I am not re-stating the bibliographic summary or the claim overview, and the separate "Family Cites Families (12)" table is not treated as "citations for 10345247" — that table lists foreign documents citing the family (e.g., JP3259331B2, CN105372256B, CN205538567U), not references cited against this US patent.

Search note / scope discipline: per the task instruction, I queried the exact number 10345247 and did not substitute lookalikes. The authoritative reference list I rely on is the citation table in the furnished full text. Where the substance of a secondary reference is not confirmed by what I could retrieve, I say so explicitly rather than inventing a disclosure.

---

Quick reference table (all 12 citations of record)

All twelve predate the 2017-02-28 priority date and the 2018-02-28 filing date, so each qualifies as prior art under AIA 35 U.S.C. § 102(a)(1) (patents/printed publications) — or, for any U.S. patent/application, potentially § 102(a)(2).

---

Detailed entries and § 102 claim mapping

For anticipation under § 102, a single reference must disclose every limitation of a claim as arranged. Claim 1 requires, in substance: (i) an illuminator producing a beam that both scatters at contaminants and reflects at the panel surface; (ii) a detector whose radiation-collection surface is perpendicular to the normal of the panel surface; and (iii) a beam trimmer with a first optical member routing scattered radiation to the detector and a second optical member routing reflected radiation away from the detector. Dependent claims add the right-angle incidence (cl. 2), TDI camera (cl. 3), three-reflector arrangement (cl. 4–6), and reflecting-prism arrangement (cl. 7–11).

1. US 4,464,050 A — Olympus Optical Co., Ltd., "Apparatus for detecting optically defects"

• Dates: JP priority 1981-02-07; US filed 1982-02-02; granted 1984-08-07.
• Description: Laser dark-field wafer/disc inspection. An objective collects both the directly reflected light and the scattered light; a first beam splitter routes them out of the illumination path and a second beam splitter separates the directly reflected flux from the scattered flux, feeding a scatter detector and a specular detector (also used for focus error). The retrieved text confirms the "splitting the directly reflected and scattered light fluxes … from one another" language.
• § 102 mapping: Discloses illuminator, scattered radiation, reflected radiation, and separate detectors — i.e., the preamble elements of claim 1 and the general separation idea of the "beam trimmer." However, it detects the reflected light rather than directing it away from the detector, and it does not disclose the perpendicular collection surface limitation (cl. 1) nor the right-angle incidence of cl. 2. No anticipation of claim 1. Best treated as § 103 background for the separation concept.

2. US 4,391,524 A — RCA Corp., "Method for determining the quality of light scattering material"

• Dates: Priority 1981-03-16; granted 1983-07-05.
• Description: Quality assessment of wafers via a scanning apparatus that detects scattered radiation; specular reflection is blocked from the detection device by lens/spatial-filter arrangement (as described in the family companion US 4,598,997 and US 4,314,763).
• § 102 mapping: Teaches the concept of blocking specular reflection so the detector sees only scatter — conceptually relevant to the "second optical member" function of claim 1. It does not disclose the claimed beam-trimmer structure or the perpendicular detector orientation. No anticipation; modest § 103 relevance to the "block the reflection" idea.

3. US 4,598,997 A — RCA Corp., "Apparatus and method for detecting defects and dust on a patterned surface" ★ (most relevant structural art)

• Dates: Priority 1982-02-15; granted 1986-07-08.
• Description: Scanning-laser inspection of patterned surfaces. A mask having apertures is arranged to pass to the detector only scattered light and to block diffracted light and specular reflections (confirmed in both the Google Patents abstract and the USPTO-report text: "scattered light is detected substantially free of diffracted beams … by a mask having apertures arranged to pass to the detector only scattered light and to block diffracted light and specular reflections").
• § 102 mapping: This is the closest cited art to the core "beam trimmer" idea — a spatial element that routes wanted scatter to the detector and rejects unwanted specular/diffracted light. It discloses L1–L4, L6, and bears on L7/L8 of claim 1. But it does not disclose (a) the detector's collection surface perpendicular to the panel normal (cl. 1), nor (b) a second optical member that directs the reflected radiation away in the claimed prism/reflector sense. It uses a transmissive mask, not the claimed reflecting prism or the three discrete reflectors of cl. 4. No strict anticipation of claim 1, but this is the single most damaging § 103 reference for the "separate the reflection from the scatter" concept.

4. US 4,794,264 A — QC Optics, Inc., "Surface defect detection and confirmation system and method"

• Dates: Priority 1987-05-08; granted 1988-12-27.
• Description: Surface-defect detection and confirmation (two-stage) system. (The family equivalent is EP 0 290 228 A2, which lists US 4,464,050 among its citations — consistent with the same art cluster.)
• § 102 mapping: General dark-field defect-detection context; I did not confirm a disclosure of the claimed perpendicular detector/beam-trimmer. No anticipation of claim 1; background/§ 103 only. ⚠️ Flagging uncertainty on the specific optical layout.

5. US 4,794,265 A — QC Optics, Inc., "Surface pit detection system and method"

• Dates: Priority 1987-05-08; granted 1988-12-27.
• Description: Companion QC Optics case directed to surface-pit detection.
• § 102 mapping: Environment/background; no confirmed disclosure of the claimed beam-trimmer/perpendicular-detector combination. No anticipation; § 103 background only. ⚠️ Unconfirmed specific disclosure.

6. JP H07-297248 A — Hitachi, Ltd., "Crystal defect measuring device and semiconductor device manufacturing method"

• Dates: Priority 1994-04-20; published 1995-11-10.
• Description: Japanese published application on crystal-defect measurement for semiconductors.
• § 102 mapping: Defect-measurement context; as a JP publication it is § 102(a)(1) prior art, but its relevance to the claimed beam trimmer and perpendicular detector is not established from the material retrieved. No anticipation; at most § 103 background. ⚠️ Treat as unverified pending full-text review.

7. US 2001/0048761 A1 — Akira Hamamatsu, "Method of inspecting a semiconductor device and an apparatus thereof"

• Dates: Priority 2000-03-02; published 2001-12-06.
• Description: Semiconductor-device inspection method (dark-field/scatter-based family of Hamamatsu disclosures).
• § 102 mapping: Inspection-method context relevant to claim 1's illuminator/detector environment. No confirmed disclosure of the perpendicular detector or the dual-optical-member beam trimmer. No anticipation; § 103 background. ⚠️ Unconfirmed as to specific optics.

8. US 2002/0122174 A1 — Akira Hamamatsu, "Apparatus and method for inspecting defects"

• Dates: Priority 2001-03-01; published 2002-09-05.
• Description: Defect-inspection apparatus/method; companion to the Hamamatsu inspection disclosures.
• § 102 mapping: Same posture as #7 — inspection apparatus context. No confirmed disclosure of the claimed beam-trimmer/perpendicular-detector combination. No anticipation; § 103 background. ⚠️ Unconfirmed specifics.

9. US 2007/0229833 A1 — Allan Rosencwaig, "High-sensitivity surface detection system and method"

• Dates: Priority 2006-02-22; published 2007-10-04.
• Description: High-sensitivity surface inspection/detection system.
• § 102 mapping: Teaches high-sensitivity scatter-based surface detection, relevant to the illuminator + detector elements of claim 1. No confirmed disclosure of the perpendicular collection surface limitation or the claimed beam-trimmer geometry. No anticipation; § 103 background. ⚠️ Unconfirmed specifics.

10. US 2008/0174771 A1 — Zheng Yan (3i Systems Corp.), "Automatic inspection system for flat panel substrate" ★ (relevant to flat-panel + TDI elements)

• Dates: Priority 2007-01-23; published 2008-07-24 (granted counterpart US 7,714,996 B2; WO 2008/110061 A1).
• Description (confirmed): AOI for flat panel substrates (LCD/OLED/mask/wafer), using an illumination component (multiple LED sources at different wavelengths), a lens array including a Fresnel lens, and dark-field inspection; the camera includes a time-delay-integration (TDI) or line-scan CCD sensor; a telecentric imaging lens directs the received light to the camera.
• § 102 mapping: Highly relevant to the application environment ("flat panel substrate") and to claim 3 (detector comprising an imaging optical path + TDI camera) and to the line-illumination aspects behind cl. 6/11. It does not disclose the claimed beam trimmer with first/second optical members, nor the perpendicular collection surface with reflected radiation steered away. No anticipation of claim 1; but it is strong § 103 art (and near-§ 102 art) against dependent claim 3 and provides the "flat panel + TDI" context.

11. US 2011/0019197 A1 — KLA-Tencor Corp. (S. W. Meeks), "Scattered Light Separation" ★ (most relevant to the separation function)

• Dates: Priority 2009-07-23; published 2011-01-27 (granted as US 7,907,269 B2).
• Description (confirmed): Apparatus that separates wanted top-surface scattered light from unwanted bottom-surface scattered light and from the specular beam. A light source directs a beam onto the substrate; the beam specularly reflects (specular beam), scatters off the top surface, and scatters off the bottom surface. A collector (ellipsoid-of-revolution with mirror-polished interior, or reflective microscope objective) plus an internal beam block and a pinhole field stop pass the top scattered light to a sensor and block the unwanted light.
• § 102 mapping: Discloses L1 (light source), L2/L3 (specular + scattered), L4 (sensor), and the separation function of L6 by an optical collector that internally reflects/collects and a beam block that rejects unwanted light — structurally analogous to the claimed reflecting prism with internal reflection and an external reflecting area (cl. 7–8). It is the most conceptually similar cited reference to the beam-trimmer/separation concept. It does not, however, disclose the claimed detector collection surface perpendicular to the panel normal with the beam-trimmer as claimed, and its problem (top vs. bottom scatter in transparent substrates) differs from the claimed problem (specular reflection in a flat-panel contaminant detector). No strict anticipation of claim 1; strongest § 103 combination candidate with #3 or #10.

12. US 2016/0293052 A1 — Robert William Sapora, "Pedagogical system"

• Dates: Priority 2015-03-31; published 2016-10-06.
• Description: A "pedagogical/teaching" system — facially unrelated to dark-field particulate inspection.
• § 102 mapping: This is the most recent citation of record and, on its face, has no material bearing on any claim of US 10,345,247. ⚠️ Its inclusion is likely for a narrow mechanical/optical detail; I could not confirm a disclosure mapping to any claim. No anticipation. I flag this as a citation whose relevance is not apparent from the title and which I could not substantiate.

Cross-check flag: the previously generated litigation section correctly noted that the patent's "Citations," "Families Citing this family," and "Family Cites Families" entries are patent citations, not lawsuits. That stands; nothing here contradicts it.

---

Consolidated § 102 conclusion

No cited reference, taken alone, anticipates independent claim 1. The two limitations that none of the twelve references discloses as arranged are:

1. "the detector having a radiation collection surface perpendicular to a normal of the surface of the flat panel under test" (claim 1) — the perpendicular-downward detector geometry that the specification presents as the adaptability improvement; and
2. the claimed "beam trimmer" structure — a first optical member (reflecting prism / first+second reflectors) that routes scatter to the detector and a second optical member (outer radiation-reflecting area / third reflector) that routes the specular reflection away — as opposed to the mask/spatial-filter (US 4,598,997) or beam-block/pinhole (US 2011/0019197) approaches.

Most relevant prior art (for § 103 and prosecution-history purposes):

• US 4,598,997 A (RCA) — pass-scatter / block-specular mask; closest structural antecedent of the "separate reflection from scatter" function.
• US 2011/0019197 A1 (KLA-Tencor) — "Scattered Light Separation" via collector + beam block + pinhole; closest functional/optical antecedent of the "beam trimmer."
• US 2008/0174771 A1 (Yan / 3i Systems) — flat-panel, dark-field, TDI-camera inspection; most relevant to the flat-panel environment and to claim 3 (and the TDI/line-scan aspects behind cl. 6/11).
• US 4,464,050 A (Olympus) and US 4,391,524 A (RCA) — secondary; establish the long-known practice of separating/blocking specular light from scatter.

Anticipation exposure by claim (summary):

• Claim 1: no § 102 anticipation from any single reference; best § 103 challenge would combine, e.g., US 4,598,997 (mask rejection of specular) + US 2011/0019197 (collector-based separation).
• Claim 3 (TDI camera): US 2008/0174771 A1 is the closest § 102/§ 103 reference (expressly an imaging path + TDI sensor for flat panels).
• Claims 4–6 (three reflectors + geometry) and 7–11 (reflecting prism + geometry): no cited reference discloses these structures/equations; these dependent claims are the strongest from a validity standpoint.

Uncertainty statement and limitations

• I could not fully retrieve the internal disclosures of US 4,794,264, US 4,794,265, JP H07-297248, US 2001/0048761, US 2002/0122174, US 2007/0229833, and US 2016/0293052. The § 102 characterizations for those entries are therefore provisional and flagged inline; a definitive anticipation opinion requires full-text review of each reference against each claim.
• The above is a technical prior-art assessment, not a legal opinion or a validity conclusion.

Sources used: Google Patents full text for US 10,345,247 B2 (furnished); Justia patent record (https://patents.justia.com/patent/10345247); Google Patents/patentimages full text for US 4,598,997, US 4,464,050, US 4,391,524 (referenced within US 4,598,997), US 2011/0019197 (and US 7,907,269), and US 2008/0174771 (and US 7,714,996 / WO 2008/110061).

I'll research the key prior-art references cited on this patent's page to ground the obviousness analysis.

Obviousness Analysis — U.S. Patent No. 10,345,247 B2

"Apparatus for detecting degree of particulate contamination on flat panel"

35 U.S.C. § 103 / Graham v. John Deere framework

---

0. Scope, sourcing, and caveats

This analysis builds on the previously generated sections (bibliographic summary, claim overview, litigation report) and does not repeat them. I treat the issued claim text in the authoritative Google Patents full text (https://patents.google.com/patent/[US10345247](/patent/US10345247)/en) as controlling, and I use the prior-art references listed on that same page — the 12 "Patent Citations" and the 12 "Family Cites Families" entries — as the § 103 reference pool.

Caveats up front (not auto-corrected, flagged):

1. For four references I retrieved only abstract/[Summary]-level or partial text (US 4,314,763; US 4,398,997's parent; US 2008/0174771 A1; JPH07297248A). My element mappings against those are therefore provisional and should be confirmed against the full specifications. Where I have the full text (US 4,598,997, US 2011/0019197 A1 / US 7,907,269, CN105372256B, CN105675617A/CN205538567U) the mappings are firm.
2. The two claim-drafting discrepancies I flagged earlier remain operative: claim 5 renders "θ5=180°1.5×theta1" (missing the minus sign present in the specification) and "h2=d2/tan(theta1/alpha1)" (specification: tan(theta1 − alpha1)). These are § 112(b) vulnerabilities; they do not alter the § 103 analysis, because a POSITA would read claim 5 against the specification's correct geometry. I note below how they interact with claim scope.

---

1. The governing legal framework

• § 103 test (Graham): (a) scope and content of the prior art; (b) differences between the prior art and the claims; (c) level of ordinary skill; (d) secondary considerations.
• KSR Int'l v. Teleflex (2007): a claim is obvious if the improvement is a predictable use of prior-art elements according to their established functions, or a combination of known elements yielding no more than expected. Any articulated reason suffices (design incentive, market pressure, "obvious to try"), and the reference need not be from the same field if the problem is one the inventor faced (KSR; In re Kliby).
• "Mere optimization of a known design parameter" is obvious (In re Boesch; KSR), which governs the geometric/equation claims 5 and 9.
• Motivation to combine may arise from: the references themselves, the knowledge of a POSITA, the nature of the problem, or the desirability of a known solution (In re Kahn, KSR).

---

2. Level of ordinary skill in the art (POSITA)

A POSITA here is a degreed optical or mechanical engineer (B.S. + ~2–4 years, or M.S.) in dark-field optical inspection / lithography contamination metrology, familiar with: oblique laser illumination; collection NA and "angle of collection"; spatial filters and Fourier-plane masking; fold mirrors/prisms; line illumination with TDI cameras; and grating scattering physics (zero-order = specular reflection, ±1st orders = scatter). This level supports treating the claimed beam-trimmer geometries as routine optical engineering.

---

3. The inventive core, in one paragraph

Stripped of boilerplate, claim 1 requires: (i) an oblique illuminator producing scatter + specular reflection at a panel; (ii) a detector whose radiation-collection surface is perpendicular to the panel normal (i.e., the camera "looks straight down," decoupled from the collection angle); and (iii) a "beam trimmer" with a first optical member routing scatter → detector and a second optical member routing specular reflection → away from detector. The specification's own stated rationale (Background/DETAILED DESCRIPTION) is that this decouples detector orientation from collection angle, which is exactly the problem-and-solution that the cited prior art already addresses with masks, prisms, and beam blocks.

---

4. Element-by-element mapping — Combination A (primary, targets claims 1, 3, 7, 8, 9, 10, 11)

US 4,598,997 A (Steigmeier et al., RCA, 1986) — primary reference — in view of US 2011/0019197 A1 / US 7,907,269 B2 (Meeks, KLA-Tencor).

Claim 3 (imaging path + TDI camera): TDI line-scan sensors are standard in dark-field inspection; US 2011/0019197's family and the KLA-Tencor area-mode / TDI-mode disclosures (e.g., US 2013/0016346, cited by third parties in the same classification space) make "imaging optical path converging scatter onto a TDI camera" an obvious engineering choice for a line illumination scheme.

Claims 7–10 (reflecting prism, trapezoid, right trapezoid, geometry): US 4,314,763 (the parent of US 4,598,997, incorporated by reference therein) discloses "a series of prisms 18 and 20" where "Prism 20 acts as a spatial filter for the specularly retroreflected light along the axis of beam 12." That is a prism performing the reflected-radiation-rejection function of claim 7's "outer radiation-reflecting area." Selecting a trapezoidal/right-trapezoidal prism cross-section and computing facet angles from the law of reflection is routine design optimization, not invention (see § 8).

Claim 11 / claim 6 (linear FoV): Line illumination is the conventional partner of TDI collection; US 2011/0019197 discusses rectangular/line-like imaged spots.

---

5. Element-by-element mapping — Combination B (targets claims 1–5, 6)

US 4,598,997 or US 4,314,763 (dark-field + specular suppression) in view of US 2011/0019197 A1 (Meeks/KLA-Tencor "Scattered Light Separation") in view of US 2013/0016346-style multi-sensor collection for the discrete-reflector embodiment of claims 4–6.

• Claim 4 requires the trimmer to be three reflectors: first reflector (scatter in) → second reflector → detector; third reflector diverts reflected radiation away. This is a folding-mirror relay — an exceptionally well-known optical construction. US 4,314,763 uses prisms (i.e., reflecting surfaces) in the illumination/collection train; US 2011/0019197 uses reflective optical elements (e.g., "reflective optical element 910/914/916," "folding mirror," "apertured mirror"). Substituting a pair of fold mirrors for a prism face to (a) steer scatter to a fixed detector and (b) deflect the specular beam to a dump is the predictable, established function of mirrors — a textbook KSR combination.
• Claim 5 recites angle relations θ5 = 180° − 1.5·theta1 and θ6 = 180° − theta1 plus minimum mirror lengths 2·d1·tan(alpha1), 2·(d1+d3)·tan(alpha1), and (0, 2·(h1−h2)]. These are closed-form geometric consequences of (i) the reflection law at the chosen mirror angles and (ii) ensuring the mirror spans the object-side angle of view alpha1. A POSITA dimensioning fold mirrors to capture a specified object-side FoV would arrive at these expressions without invention. In re Boesch / KSR ("a court must ask whether the improvement is more than the predictable use of prior art elements according to their established functions").
• Claim 6 (linear FoV) — as above.

---

6. Element-by-element mapping — Combination C (same-field, same-problem combination; targets claims 1, 7, 9)

CN 105 372 256 B (Shanghai Micro Electronics Equipment (Group) Co., Ltd.; CN priority 2014-08-20, granted 2019-01-18) — a family-cited reference — in view of US 4,598,997 and US 2011/0019197 A1.

CN105372256B is highly probative because it discloses a surface-detection system for exactly the accused problem-space: an illumination source, a detection unit (探测单元) that detects scattered light from the surface, a recovery unit (回收单元) that collects the specularly reflected light (镜面反射光), a light-absorbing unit (吸光单元) that absorbs residual specular reflection, and a polarizing beam splitter that reflects the secondary specular reflection to the absorber so it does not reach the detector (claim 1; see the granted claims reproduced in the CN105372256B PDF).

• This reference teaches the second-optical-member function verbatim in substance — collect/divert the specular reflection so that scatter reaching the detector is not corrupted (here by beam-splitter + absorber, i.e., an optical member redirecting reflected radiation away from the detector).
• Method substantially the same, field substantially the same, problem substantially the same → strong KSR motivation. The only difference is the particular optical member (absorber/PBS vs. trapezoidal prism vs. three mirrors) — an obvious substitution of known equivalents.
• CN105675617A / CN205538567U (Dongxu) show, in the same "sheet-glass surface particulate" space, that light-path structures with first/second reflecting mirrors and a beam-splitting coating were well-known building blocks for flat-panel particle metrology.

---

7. Motivation to combine (the KSR rationales, stated)

1. Same field of endeavor / same problem. All primary references are dark-field optical inspection of flat or patterned substrates (wafers, masks, glass, flat panels) for particles/defects — precisely the problem in the '247 Background (crosstalk from mirrored particles and bottom-surface pattern degrades SNR). KSR; In re Kliby.
2. The references themselves supply the motivation. US 4,598,997 states the problem and solution outright: "Specular reflections from the surface as well as diffracted beams from the patterns are blocked so that the detector is exposed to substantially only diffusely scattered light." US 2011/0019197 states the equivalent goal ("separates the scattered light… before the scattered light… can attain the sensor"). CN105372256B motivates collecting the specular reflection and redirecting it so it is not detected. A POSITA confronting exactly the '247 Background problem would look to exactly these solutions.
3. Predictable result / no unexpected behavior. Beam-blocking, spatial filtering, fold mirrors, and prisms have established functions (reject on-axis specular, relay off-axis scatter). Combining them produces the expected SNR improvement — the specification's only asserted benefit (avoiding reflected-radiation crosstalk) is the predictable result of the combination.
4. Design incentive to fix the detector orientation. The '247 Background itself concedes the prior-art drawback: the imaging module's optical axis had to be aligned with the collection angle, and "the supporting, assembly and clamping of the imaging module 20 must be adjusted whenever there is a change in the angle of collection 60. Thus, the compatibility of this solution is inferior." Fixing the camera and steering light with folding optics is the most conventional response to that design constraint — a KSR "design incentive" rationale that comes from the patent's own admission of the problem.
5. "Obvious to try." With a finite number of known beam-steering alternatives (mirrors, prisms, masks, beam blocks, aperture plates, PBS+absorber), choosing one to route specular light away from a detector is the paradigm "obvious to try" case.

---

8. The dependent geometric claims (5 and 9) — separate obviousness ground

Claims 5 and 9 are product-by-formula claims whose equations are the algebraic solution of well-known reflection/geometry constraints:

• Claim 9 (θ1 = 90° − (90° − theta1); θ2 = 180° − 1.5·theta1; θ3 = 1.5·theta1; θ4 = 180° − theta1; l = 2·d1·tan(alpha1)). Note that the first equation is an algebraic identity (it reduces to θ1 = theta1) and the remaining relations follow from the specification's own derivations (2·beta1 = 90° − theta1, so beta1 = theta1/2). These are the ordinary angle bookkeeping of a trapezoidal prism chosen to reflect an on-axis scattered beam to a downward-facing sensor — the predictable output of applying the law of reflection to an arbitrary prism geometry selected by the designer.
• Claim 5 similarly recites the mirror-angle/length equivalents; the l = 2·d1·tan(alpha1)-type expressions are just the width of the object-side FoV footprint at a stated distance — a standard first-order optics calculation.

Under In re Boesch and KSR, "mere optimization of a known design parameter" does not confer patentability. Claim 9's "where d1/alpha1/theta1 are design inputs" framing confirms these are convenient parameterizations of a known optical arrangement, not a new result.

⚠️ Cross-reference to earlier flag: the grant text of claim 5 contains θ5=180°1.5×theta1 and h2=d2/tan(theta1/alpha1). The specification supplies 180° − 1.5×theta1 and d2/tan(theta1 − alpha1). If one reads claim 5 literally as issued, the missing operators make the claim ambiguous/indefinite (§ 112(b)) — which cuts against the patentee (a claim of indefinite scope cannot be saved by the specification where the claim is the measure), and in any event does not create a § 103 distinction, since the correct (specification) formulas are themselves conventional geometry.

---

9. Known weaknesses, counter-arguments, and what would defeat the analysis

I flag these honestly — an obviousness conclusion is only as strong as the weakest link:

1. The "perpendicular collection surface" limitation. Combination A locates the orthogonal-detector geometry in the parent US 4,314,763 (normal-incidence, detector on the beam axis) rather than in US 4,598,997. If the target construction is oblique incidence with a downward-facing camera, the mapping to A/B/C is proxy-level unless a reference expressly shows oblique incidence + normal collection + trimmer. The closest express teaching I found is the motivation (the '247 Background's own admission of the assembly/compatibility drawback), not a reference figure. I could not verify with the retrieved text that any single reference shows all three simultaneously. This is the patentee's best non-obviousness argument.
2. "Optical member" vs. "mask/block." In US 4,598,997 the light-rejecting element is an opaque aperture mask, and in US 2011/0019197 it is a beam block. A patentee may argue a mask/block does not "direct" reflected radiation as claimed, and that the claim requires an element that optically re-routes (reflects/refracts) the beam. Combination A's second-member mapping is thus vulnerable; Combination B (mirrors/prisms) and Combination C (PBS + reflector) are stronger because those elements actively redirect the specular beam. Recommend resting the rejection primarily on B and C for claim 1's second-member limitation.
3. Breadth of "beam trimmer." Because claim 1 uses functional language ("configured for directing…"), it is broad — which increases § 103 exposure. Broad functional claiming invites prior-art coverage; conversely it is easy to design around.
4. Secondary considerations. I found none in the record: no unexpected results, no industry praise, no long-felt-but-unmet need with evidence, no licensing-nexus evidence, and no commercial-success evidence tied to the claim. (The only post-grant event is the 2025 assignment to Amies Technology Co., Ltd..) If the patentee later proffers secondary evidence, it must show a nexus to the claimed combination; the "compatibility/orientation" benefit is a design convenience, which is weak objective evidence.
5. Teaching away — low risk. Nothing in the cited art teaches away from rejecting specular light; to the contrary, all of A/B/C embrace it. A "teaching away" defense is not supported by these references.
6. Foreign-language references (CN105372256B, CN105675617A/CN205538567U, JPH07297248A, JP5787261B2). Under § 102(a)(2)/§ 103 they are usable prior art if published before 2017-02-28 (CN105372256B published 2016-03-02; CN105675617A published 2016-06-15; CN205538567U published 2016-08-31 — all pre-priority). I have translated claim/abstract text for these, and my mappings rely on that; certified translations should be obtained before citing them in a formal office action or petition.
7. Same-assignee art muddies motive. CN105372256B shares the original assignee (Shanghai Micro Electronics Equipment (Group) Co., Ltd.) with the '247 patent. That is not a § 103 defense (common ownership is not an exception to § 103 absent § 103(c)-type statutory safe harbors that do not apply to issued-claim invalidity), but it does support a "known to the inventor/same laboratory" narrative that can strengthen the obviousness case (the solution family was evidently in the assignee's own prior toolbox).

---

10. Bottom line

Using only the references on the patent's own page:

• Claim 1 is, at minimum, prima facie obvious over US 4,598,997 in view of US 2011/0019197 A1 / US 7,907,269, and is more robustly obvious over CN105372256B in view of US 2011/0019197 A1, each optionally combined with US 4,314,763. Each limitation is disclosed or is an obvious substitution of known optical elements performing established functions, with a motivation supplied by the references and by the patent's own admission of the prior-art drawback.
• Claim 3 (imaging path + TDI) and claim 6 / 11 (linear FoV) are obvious design choices in dark-field line-scan inspection.
• Claim 4 (three-reflector trimmer) is obvious over the prism-based spatial filter of US 4,314,763 viewed as a mirror equivalent.
• Claims 5 and 9 (angle/length equations) are obvious as mere optimization of known design parameters / predictable geometry, subject to the flagged § 112(b) defects in claim 5.
• Claims 2, 7, 8, 10 (perpendicular incidence; reflecting-prism trimmer; trapezoid; right trapezoid) are obvious as particular known geometries/optical elements.

Most defensible single combination to assert: US 4,598,997 A + US 2011/0019197 A1 (US 7,907,269 B2) + CN 105 372 256 B, with US 4,314,763 as corroborating evidence of the prism/spatial-filter and normal-collection geometry, and US 2008/0174771 A1 (Zheng Yan) as secondary evidence that flat-panel substrate inspection was a known application of exactly this dark-field architecture.

---

Sources relied upon

• Authoritative patent text: https://patents.google.com/patent/US10345247/en
• US 4,598,997: https://patents.google.com/patent/US4598997 ; PDF https://patentimages.storage.googleapis.com/3e/a8/ae/a868a9ce912240/US4598997.pdf
• US 4,314,763 (RCA, parent/incorporated by reference in '997): https://patentimages.storage.googleapis.com/51/49/56/027d8c94187537/US4314763.pdf
• US 2011/0019197 A1 and US 7,907,269 B2 (KLA-Tencor, "Scattered Light Separation"): https://patents.google.com/patent/US20110019197A1/en ; PDF http://patentimages.storage.googleapis.com/49/5f/f7/31b0cc2c929e14/US7907269.pdf
• CN105372256B (Shanghai Micro Electronics Equipment (Group), Surface detection system and method): https://patentimages.storage.googleapis.com/15/ef/31/16fd5b69d73f57/CN105372256B.pdf
• CN105675617A / CN205538567U (Dongxu Technology, glass-surface particle light-path structure): https://patentimages.storage.googleapis.com/fa/f8/1f/b5b42f41b82492/CN105675617A.pdf
• US 4,394,524; US 4,464,050; US 4,794,264; US 4,794,265; US 2001/0048761 A1; US 2002/0122174 A1; US 2007/0229833 A1; US 2008/0174771 A1; JPH07297248A — as listed in the "Citations"/"Family Cites Families" tables of https://patents.google.com/patent/US10345247/en (mappings for the ones I could not open in full are flagged in § 0 and § 9).

This is a patent-analyst obviousness assessment, not a legal opinion or a validity determination; formal opinions should obtain certified translations of the CN/JP references and confirm the flagged mappings against full specifications.