US 9091533

I'll run searches for this patent number on the specified sources.

Let me check litigation/docket sources specifically.

Let me run one more targeted check on the Federal Circuit docket and USPTO PTAB.

US 9,091,533 — verified bibliographic summary

Search scope note: I ran web searches against public patent databases (Google Patents, Justia Patents, patents-review) and general Federal Circuit / PTAB coverage. I found no Federal Circuit 2026 docket entry, appeal, IPR/PGR, or district-court litigation naming US 9,091,533. I cannot treat that as proof of non-existence — my tooling does not query the CAFC docket system (PACER/CM-ECF) or USPTO PatentCenter docket directly. Statements below are grounded in the authoritative full text you supplied.

Abstract (as published)

The invention proposes an optical monitoring device for monitoring the activity of a tool in a monitoring area, having at least one camera (1). In order to improve the capture of the lateral regions of a three-dimensional object in the monitoring area (20), at least one mirror (2) is provided that is concavely bent in at least one partial region and that is disposed in the optical path between the monitoring area (20) and the camera (1).

Independent claims

There is exactly one independent claim — claim 1. Claims 2–14 all depend, directly or indirectly, on claim 1 (2→1, 3→1, 4→1, 5→1, 6→1, 7→6, 8→6, 9→8, 10→1, 11→1, 12→1, 13→1, 14→1). There are no other independent claims and no separate apparatus/method independent claim.

Claim 1, in plain language. An optical monitoring device for watching what a tool does in a monitoring area, made of:

1. at least one camera;
2. a first annular (ring-shaped) mirror that is concave-bent in at least part of its surface, placed in the light path between the monitoring area and the camera — i.e., the camera looks at the mirror, and the mirror looks at the work zone;
3. during operation the tool sits inside the hole/recess of the ring mirror, that recess being bounded by the mirror's inner contour (the mirror rings the tool like a collar);
4. the camera and ring mirror are arranged so the camera's field of view runs at least partly radially with respect to the tool's longitudinal axis;
5. the field of view looks into the monitoring area from outside it (a lateral/inward-looking view rather than a straight top-down view);
6. the ring mirror and camera are mountable on the tool itself, so the field of view is always aimed at the monitoring area as the tool moves — no separate tracking system needed; and
7. at least one component of the camera's line-of-sight axis travels radially relative to the ring mirror at the point where it strikes that mirror — a geometric restriction on the incidence direction at the mirror surface.

In short: one main claim that combines the annular concave mirror, tool-in-the-recess mounting, radially-directed viewing geometry, and tool-mounted (moving) deployment.

Dependent-claim coverage (brief). Circular/elliptical inner contour (2); mirror cross-section midpoint offset from the tool axis (3); circular/elliptical cross-section (4); radial, outward-directed line of sight after reflection (5); optional flat second mirror anywhere in the path (6); pivotable/displaceable mirrors (7); analysis device controlling camera/mirrors/illumination/light-line and analyzing images (8); 3-D measurement capability (9); ≥2 cameras with overlapping fields of view (10); flash or continuous illumination unit (11); light-line device for triangulation (12); line of sight directed radially outwards (13); camera positioned on the inner side of the ring mirror (14).

Points of analytical interest (flagging, not asserting invalidity)

• Drafting artifacts: claim 8 (via 6) recites "the illumination device" and "the light line device" without antecedent basis in that chain — those elements are only introduced in claims 11 and 12. Claim 13 largely restates the "directed outwards" idea already in claim 5. These are the kind of defects typically raised under §112(b) and could matter if the patent were ever asserted.
• Description inconsistency in the record: the "Detailed Description" opens "only one camera … is shown," then states the device "comprises two cameras 1," while the same passage discusses "the three cameras 1." The figure and count are internally inconsistent. Mirrored identical text appears in the pre-grant publication US 2010/0007733 A1.
• Closest cited art includes DE 203 07 305 U1 (same assignee — tool-mounted optical monitoring), EP 1 697 061 A1 and EP 1 701 801 A1 (Quiss GmbH, multi-camera recognition of an applied structure), DE 10 2006 008 552 A1 (tilted concave mirror with conical/part-spherical reflecting surface), DE 10 2007 031 835 A1 (Lensation, beam-deviating arrangement surrounding a tool), and non-patent literature on reflective-optics compound-eye cameras and spherical-image-sensor vehicle monitoring. These are the references examiners relied on; the claim-1 combination (annular concave mirror + tool in recess + radial strike geometry + tool-mounted) is what distinguishes it.

Uncertainty statement

• I have no authoritative confirmation of any 2026 CAFC activity involving 9,091,533. Absence of hits in open web sources is weak evidence and should be confirmed directly against PACER/CM-ECF and USPTO PTAB/PatentCenter dockets.
• Legal status ("Active," expiry 2032-12-17) is reproduced from the Google Patents legal-status field, which itself disclaims being a legal conclusion. It should be verified against USPTO PatentCenter.
• Minor housekeeping: your instructions give the current date as April 26, 2026, while my operating context states September 29, 2026. I have not adjusted any records based on either date, but any docket-currency conclusions depend on which date governs.

I'll search for litigation involving US patent 9091533.

Let me search additional litigation-specific sources.

Let me try additional targeted sources.

Findings: Litigation involving U.S. Patent No. 9,091,533

No litigation was found involving U.S. Patent No. 9,091,533 ("Optical surveillance device").

I searched for the exact patent number across sources that cover patent litigation (Unified Patents litigation portal, RPX Insight, CourtListener/RECAP, Google Patents legal-events data, and general web queries pairing the number with terms like "litigation," "lawsuit," "infringement," "defendant," and "IPR"). All searches for the exact number 9091533 returned only the patent's own bibliographic/prosecution record — no district court complaints, no IPR/PGR petitions, and no CAFC appeals.

What the record does show for '533

Adjacent proceedings that are NOT '533 litigation (do not confuse)

• The family's European counterpart EP 2 131 145 B1 is listed as "not-in-force," and the German application DE 10 2008 027 393 A1 is listed as "withdrawn"/not active. These are prosecution/status events, not litigation.
• The "Cited By" and "Families Citing this family" lists on the Google Patents page (e.g., Honeywell US 2015/0279099; Cryovac US 11,628,610; Glasstech US 11,867,630 / US 12,079,980; Recognition Robotics US 9,885,562) are later documents that cite '533 as prior art. A citation is not evidence of litigation against '533, and I found no case linking those parties to an assertion of '533.

Important caveats

• This is a negative finding, not a certification. I could not query Docket Navigator or PACER directly (paywalled/access-limited), and my search step budget was exhausted before I could run the full slate of planned queries. A definitive "never litigated" conclusion would require a direct PACER/Docket Navigator and PTAB (PTAB E2E) docket check.
• U.S. patents with expired/abandoned foreign counterparts and no assertion history are often simply never litigated, which is consistent with the clean record here, but I cannot represent that as certain.
• I did not auto-correct the identifier: every search used the literal 9091533 / 9,091,533. When a query for a different 7-digit number returned unrelated hits (as several did), I did not treat those as results for '533.

Bottom line: Based on available public sources, there is no known litigation — district court, PTAB, or Federal Circuit — involving U.S. Patent No. 9,091,533. If you need a certified clear, the next step would be a direct Docket Navigator / PACER and USPTO PTAB docket pull.

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

Let me run two more targeted checks to confirm the negative.

Proceedings overview

Zero AIA trial proceedings on file. The USPTO Open Data Portal structured block reports no IPR, PGR, or CBM petition ever filed against US 9,091,533, and independent web searching surfaced no PTAB trial, no Final Written Decision, and no Federal Circuit appeal arising from one. There is therefore nothing in the "claims invalidated / claims sustained / settled / institution denied" taxonomy to report — count of proceedings by status: active 0, claims invalidated 0, claims sustained 0, settled 0, institution denied 0. The bottom-line defensive posture for a defendant is the opposite of the "hardened patent" scenario: the patent has never been tested at the Board, which cuts both ways — there is no estoppel or adverse PTAB record to exploit, but there is also no free kill shot on the table, and all 14 claims stand as issued and presumptively valid.

No proceeding to describe

Because the structured data contains no proceeding number, I am not going to manufacture one. Drafting a heading like "### IPR20XX-XXXXX — Petitioner v. ISRA VISION AG" with invented panelists, grounds, and an institution date would be exactly the fabrication the constraints forbid. The correct deliverable here is the null result, stated plainly.

False positives to ignore. Searches for "9,091,533" surface a substantial volume of unrelated PTAB activity from similarly-numbered patents — most prominently petitions against US 9,058,153 (a multi-display user-interface patent, e.g., the petition at the PTACTS petition viewer, https://ptacts.uspto.gov/ptacts/public-informations/petitions/1556799/download-documents) and against US 11,035,885. Those filings involve a different patent number, different claims (directed to multi-display window management), and different parties (the briefing references Apple Inc. v. Fintiv, Inc., IPR2020-00019, and an Ericsson v. IP Bridge dispute). Do not let a defendant's counsel conflate the '153 patent with the '533 patent — the digit transposition is easy to make and I found no evidence connecting that litigation to ISRA VISION's optical monitoring patent.

Strategic summary

Claim status. All fourteen claims of US 9,091,533 remain in force exactly as granted on 2015-07-28. Independent claim 1 — the annular-mirror monitoring device with the tool seated in the mirror's inner-contour recess, radial field of view, and the "attachable to the tool" limitation — is untested and uncancelled. So are dependents claims 2–14, covering the circular/elliptical inner contour (claim 2), off-axis mirror cross-section midpoint (claim 3), circular/elliptical cross-section (claim 4), radially outward line of sight (claims 5 and 13), the optional flat second mirror (claim 6), pivotable/displaceable mirrors (claim 7), the analysis device and three-dimensional measurement (claims 8–9), multiple overlapping cameras (claim 10), the flash/continuous illumination unit (claim 11), the triangulation light line (claim 12), and the camera on the mirror's inner side (claim 14). There are no "surviving claims" in the narrowed sense — nothing has been narrowed.

Estoppel landscape. Section 315(e)(2) estoppel is a non-issue: it attaches only to petitioners who obtain a final written decision, and there has never been a petitioner. No IPR estoppel runs against anyone. That means a defendant sued today has the complete universe of § 102 and § 103 printed-publication art available in an IPR, and — critically — is not limited to the grounds that some earlier challenger already ran. The corollary is that a defendant has no ready-made § 315(e)(2) attack on a co-defendant or competitor, either. All prior art cited on the face of the patent (DE 203 07 305 U1, EP 1 697 061, EP 1 701 801, DE 102006008552 A1, DE 102007031835 A1, and the rest of the 19-item citation list) is still un-adjudicated at the Board, so both ex parte and AIA routes remain open.

Pattern signals. None. No petition has ever been filed by any party, so there is no repeat-petitioner pattern, no serial-IPR joinder fight, and no defensive aggregator (Unified Patents or similar) anywhere in this patent's chain. The patent owner, ISRA VISION AG, has not had to defend a single PTAB appeal for this patent — the only administrative-history events in the record are routine: the 2009-09-23 assignment of the inventors' rights to ISRA VISION AG, the 2015-07-08 notice of allowance/grant, and maintenance-fee payments on 2019-01-23 and 2023-01-19. The patent is listed as Active, expires 2032-12-17 (post-PTA adjusted).

The signal in the silence. A patent that has survived eleven-plus years and two maintenance-fee cycles without a single AIA challenge is unusual for a well-asserted patent — well-asserted patents attract IPRs. The reasonable inference is that US 9,091,533 has not been the subject of significant high-stakes infringement litigation, at least not against a party willing to fund a $300k–$1M PTAB campaign. (Its sibling EP 2131145 B1 shows as "not-in-force" in the family data — an EPO-status fact, not a PTAB one, and I would not read too much into it without the EP register.) A defendant receiving a demand letter citing this patent should treat the absence of IPR history as neutral-to-favorable for a challenger — no estoppel, no adverse claim constructions, a clean slate — while recognizing that the patent owner also has no loss narrative to overcome.

Recommended next steps

• Confirm the null result yourself before relying on it. Run the patent number through PTAB E2E (https://ptab.uspto.gov/) and the PTAB Decisions search at https://www.uspto.gov/patents/ptab/decisions, and check CourtListener's RECAP docket search (https://www.courtlistener.com/) for any district-court case in which this patent was asserted. The ODP ingest may lag; docket-level evidence of assertion is often the leading indicator that a petition is coming.
• No FWD to link. There is no Final Written Decision for US 9,091,533 to quote or hyperlink, and no disposition to characterize. Any document purporting to be a PTAB FWD on this patent number should be treated as a mis-numbered reference to a different patent (see the '153 / '885 false positives above).
• No trial milestones to track. With no instituted proceeding there is no statutory § 316(a)(11) one-year FWD clock running, no institution-decision deadline, and no oral-hearing date. Nothing to calendar on the PTAB side.
• If you are contemplating a challenge: the patent's own prosecution history is short on substantive rejections, and the cited art (DE 203 07 305 U1 in particular, the patent owner's own earlier device covering the tool-mounted camera with overlapping fields of view; and DE 102007031835 A1 to Lensation on beam-splitting and beam-path deviation around a tool) is the natural starting corpus. Given that claim 1 requires the combination of an annular mirror concavely bent in at least one partial region, the tool positioned within the mirror's recess, a field of view running at least partially radially to the tool's longitudinal axis, captured from outside the monitoring area directed into it, and the mirror-plus-camera attachable to the tool, a § 103 challenge would need art teaching that specific annular-mirror-with-central-tool-recess geometry — not merely a concave mirror somewhere in the optical path. I have not analyzed those references on the merits here, and I would not represent that they anticipate or render obvious claim 1 without doing so.
• One honest limitation: I could not verify the absence of a recently filed, not-yet-indexed petition. If counsel has specific reason to believe a competitor filed in the last few months, the PTACTS petition search at https://ptacts.uspto.gov/ is the authoritative check.

Bottom line for a defendant: this patent is un-challenged, not hardened. There is no cancelled claim to hide behind, no PTAB record to cite, and no estoppel working in your favor or against you. Every defense must be built from scratch — which is a clean slate, but not a free win.

I'll reconstruct the assignment record for US 9,091,533, cross-referencing the patent's own legal-events data with assignment records and corporate-status sources.

Let me verify the original assignee's corporate status and check for any successor entities or further transfers.

Let me check for any recorded post-issuance assignments (e.g., to Atlas Copco entities) and confirm the single-record status.

Inventors

Source: Justia patent page for 9,091,533 (inventor residences); Google Patents legal events (assignors of record on the 2009 assignment).

• All three inventors personally executed the assignment to ISRA VISION AG (Reel 023274/0535), which is the standard fingerprint of employee/obligated-assignor inventors rather than independent inventors licensing into a company.
• Darmstadt is ISRA VISION's headquarters, so the lead inventor's residence is consistent with direct employment; the other two reside elsewhere in Germany, which is normal for a multi-site German engineering company and is not by itself evidence of outside-inventor sourcing.
• Unusual-pattern check — all inventors departing within 12 months of filing: not determinable from the assignment record. There is no recorded inventor-to-inventor assignment, no employment-agreement assignment recorded separately, and no evidence in the public record of any of the three leaving ISRA around 2009. I found nothing that would support (or refute) a fire-sale-precursor pattern. Do not infer one.

Original assignee

Entity named on the issued patent: ISRA VISION AG (Darmstadt, Germany) — original and, as of the last recorded event, current assignee of record.

• Primary line of business: Industrial machine vision. ISRA VISION specializes in surface inspection (web/glass/foil) and 3D machine vision, specifically 3D robot guidance, quality inspection and 3D metrology. The '533 claims (camera + concavely bent annular mirror for monitoring a tool such as an adhesive-bead applicator) sit squarely in this product space — the patent's own specification frames it around robotically operated tools and adhesive-bead quality control.
• Did they ship a product embodying the claims? Unconfirmed at claim level. ISRA VISION commercially sells 3D robot-vision/robot-guidance and surface-inspection systems (e.g., the "PowerPick3D" sensor line appears under ISRA VISION AG's IP/trademark holdings). I could not verify a specific commercial product that practices the limitations of claim 1 (annular concave mirror with the tool in its recess, field of view directed into the monitoring area from outside). Treat "ships an embodying product" as likely true, not documented.
• Current status: Operating, and now part of Atlas Copco. In Feb 2020 Atlas Copco announced a voluntary public takeover at €50/share; settlement closed 2020-06-24 at 92.19%. In Aug 2020 Atlas Copco announced a merger of ISRA VISION AG into its takeover vehicle Atlas Copco Germany Holding AG, with a squeeze-out of minorities. That squeeze-out completed 2021-05-18: ISRA VISION AG became a wholly owned subsidiary of Atlas Copco Holding GmbH and delisted.
• Legal-entity subtlety worth flagging: the surviving corporation in that 2021 merger was the former Atlas Copco Germany Holding AG, which relocated its seat to Darmstadt and was renamed "ISRA VISION AG." So the name "ISRA VISION AG" on the patent record today denotes a different legal person than the 2009 assignee. Title to '533 would have passed to that surviving entity by universal succession (operation of law), which is generally not recorded as an assignment at the USPTO. This is a reasoned inference from the merger disclosures, not a recorded USPTO event — see the assignment timeline below.
• Not litigation-related: ISRA VISION AG appears in ISRA VISION, AG v. Burton Industries, Inc., 654 F. Supp. 2d 638 (E.D. Mich. 2009) — a supplier/contract dispute, not a patent case, and predating the '533 grant by six years. It is not evidence of patent assertion and should not be confused with an infringement suit.

Assignment timeline

Chronological — this is the complete set of recorded assignments I can substantiate for US 9,091,533.

• 2009-08-20 to 2009-08-24 (executed) / recorded 2009-09-23 — Reel 023274 / Frame 0535
  • Conveyance: Assignment ("ASSIGNMENT OF ASSIGNORS INTEREST")
  • Assignor: Bachem, Alexander; Westenhoefer, Matthias; Hohm, Karlheinz (individually, as inventors)
  • Assignee: ISRA VISION AG, Germany
  • Correspondent: Not exposed / not retrieved. The Google Patents legal-events feed for '533 reports the reel/frame, assignors, assignee and signing dates but does not publish the correspondent of record, and I was unable to open the Assignment Center detail page for a direct pull within this session. I am not supplying an attorney or firm name, because doing so would be fabrication. This field must be filled by a direct Assignment Center query.
  • Context: Initial acquisition / obligated assignment — the founding inventors assigning the application to their employer ~12 months after the 2008 priority filings and ~3 months after the 2009-06-03 US filing. Not a fire-sale, not a transfer-to-asserter.

No further recorded assignments exist on this patent. The remaining Google Patents legal events for '533 are administrative only and confer no title:

Two negative findings that matter:

1. No change-of-name or merger assignment was recorded at the USPTO when ISRA VISION AG was merged into the Atlas Copco vehicle in 2021. That is consistent with universal succession by operation of law (no instrument to record), and it explains why the face of the patent still reads "ISRA VISION AG." It does not mean ownership never changed — it means ownership changed without a recorded instrument.
2. No assignment to any LLC, IP-holding entity, or aggregator has ever been recorded. If the patent had been sold into an NPE structure, a reel/frame would exist. None does.

Caveat on completeness: I could not complete a direct Assignment Center / Patent Assignment Search pull for this patent number. The absence of post-2009 records is corroborated by the Google Patents legal-event table (which is derived from the same USPTO assignment data) and by the maintenance-fee payments continuing under the original owner name, but a definitive chain-of-title clear would require a direct query at the Assignment Center.

Timeline diagram

timeline
    title Ownership of US 9091533
    2008 : Priority filings in Germany
    2009 : US application filed 3 June
         : Inventors assign to ISRA VISION AG
    2015 : Patent issued 28 July
    2020 : Atlas Copco buys 92 percent stake
    2021 : Merger into Atlas Copco vehicle
         : Same ISRA VISION AG name retained

NPE / troll-pattern signals

1. Shell-entity transfer — not present. The only recorded conveyance is inventors → ISRA VISION AG at Reel 023274/0535 (recorded 2009-09-23). The assignee is the operating company that built the product, headquartered at a real corporate address in Darmstadt — not a registered-agent service address, not an "IP Holdings/Licensing/Ventures" suffix, not a single-purpose LLC.
2. Known asserter in the chain — not present. No assignee or prior assignee matches Acacia, Marathon, Intellectual Ventures, IPNav, Wi-LAN, Mosaid/Conversant, Vringo, Pendrell, Innovatio, MPHJ, Lumen View, Round Rock, Document Generation Corp, or any Spangenberg-linked entity. The chain runs ISRA VISION AG → (operation of law) → Atlas Copco group, i.e., a German machine-vision manufacturer and later a Swedish industrial conglomerate. Neither is on any Unified Patents / RPX high-frequency-plaintiff list.
3. Repeat correspondent across the chain — unclear / not assessable. The chain has only one recorded link, and the correspondent of record for Reel 023274/0535 was not retrievable in this session. With a single link there is no recurrence to detect regardless. Flagging as a gap, not a negative finding — a recurred correspondent is the single highest-value tell in anonymous-NPE chains, and this one was not checked at source.
4. Cascading transfers — not present. Zero chained assignments. One transfer in sixteen years, through an operating manufacturer, with no LLC-to-LLC hops.
5. Pre-litigation transfer — not present. No infra suit naming '533 has been identified (see the litigation section above), so there is no assertion to precede. The single assignment predates the 2015 grant by ~6 years and the 2008 priority by ~3 months — the opposite of a just-before-suit timing pattern.
6. Bankruptcy fire-sale — not present. No Chapter 7/11 proceeding involving ISRA VISION AG. Its 2020–21 event was a cash takeover and squeeze-out at €50/share, not a distressed sale, and the division continued to operate under the ISRA brand with a new Darmstadt headquarters completed ~2022/23.
7. Privateering — not present. Atlas Copco acquired ISRA as a strategic, all-cash operating acquisition (recorded intangibles/goodwill in Atlas Copco's 2020 annual report: total consideration ~MSEK 10,604 for ISRA; ISRA intangible assets ~MSEK 4,142), and continued the business with ~900 employees. No evidence — in SEC-equivalent filings or elsewhere — of Atlas Copco transferring ISRA patents to an NPE to assert against competitors.
8. Defensive aggregator (anti-NPE) — not present. The chain does not terminate at RPX, AST, LOT Network, Unified Patents, or OIN. No neutralization event.

Verdict

Insufficient data — and I mean this in the rubric's narrow sense: only the original assignment is recorded. The complete recorded chain is one instrument — inventors → ISRA VISION AG, Reel 023274/0535, executed 2009-08-20/24 and recorded 2009-09-23 — with no subsequent title records other than administrative grant/fee events. There is no NPE pattern to score (no shell transfer, no asserter assignee, no cascading LLC hops) and no defensive-aggregator termination, so the four NPE/defensive verdicts cannot be affirmatively selected. The substantive finding is affirmatively non-NPE: the patent has belonged since 2009 to the operating company that commercialized the technology, and it now sits inside Atlas Copco's ISRA VISION subsidiary by merger operation of law rather than by any recorded licensing-entity transfer. The one unresolved field is the correspondent of record on Reel 023274/0535, which I could not retrieve and therefore did not guess.

Verification link: https://assignmentcenter.uspto.gov/ — search by patent number 9091533 (or application 12/477,383); the legacy entry point https://assignment.uspto.gov/patent/index.html resolves to the same underlying records. A direct pull is required to (a) confirm no post-2009 records exist and (b) capture the correspondent name for Reel 023274/0535.

I'll verify the citation record and enrich a few key references, then build the §102 analysis. Let me run targeted searches.

Good confirmations. Let me enrich a few of the less obvious references.

US 9,091,533 B2 — Prior-Art Analysis (§102)

Scope, method, and literal-identifier compliance

Search performed. I attempted to pull patent 9091533 from authoritative sources. My available tooling does not include a direct USPTO PatentCenter/PAIR or USPTO Open Data API connector, so I relied on (a) the authoritative full text you supplied (the Google Patents record, which mirrors USPTO bibliographic and citation data) and (b) live web/patent-database lookups of individual reference numbers. Where a live lookup disagreed with the supplied record I say so.

No auto-correction. Every identifier below is reproduced literally from the record. I ran searches on the literal string 9091533 / 9,091,533 only. Searches that returned different (7-digit or unrelated) numbers were discarded and were not treated as results for '533. No similar-number result has been substituted.

Builds on prior sections. This section assumes the bibliographic table and claim-1 element breakdown already generated in the Patent summary, and the no-litigation finding in the Litigation summary (consequence: there is no IPR/PGR or court record supplying a different prior-art set; the only prior-art record is the examiner's citation list). I found no contradiction with those sections. The camera-count inconsistency and the claim-8 antecedent-basis defect flagged earlier are carried forward below because they affect claim mapping.

1. Governing law and critical dates

• Pre-AIA 35 U.S.C. §102 applies. Application 12/477,383 was filed 2009-06-03, i.e., before the AIA first-inventor-to-file provisions took effect (2013-03-16). So §102(a)/(b)/(e) (pre-AIA) govern, not §102(a)(1)/(a)(2).
• Priority: §119(a) foreign priority from DE 20 2008 007 630 U (2008-06-06) and DE 10 2008 027 393 (2008-06-09). Assumed effective date for §102(a)/(e) purposes: 2008-06-06.
• §102(b) bar date: measured from the U.S. filing date → one year before 2009-06-03 = 2008-06-03. Printed publications/patents dated before 2008-06-03 are statutory-bar art.
• §102(a) window: publications after 2008-06-03 but before the invention/priority date (2008-06-06) — practically closed.
• §102(e): a U.S. patent or U.S. application publication is art as of its U.S. filing date, even if published later. Foreign (DE/EP) documents get no §102(e) date — only their publication date counts.

Anticipation standard. A reference anticipates a claim under §102 only if it discloses every limitation of that claim. There is one independent claim (claim 1); claims 2–14 all depend from it and therefore incorporate all of claim 1's limitations. Consequently, a reference can anticipate a dependent claim only if it also discloses all of claim 1.

Let me label claim 1's limitations for the tables:

• C1-a at least one camera
• C1-b first annular mirror, concavely bent in ≥1 partial region, in the optical path between monitoring area and camera
• C1-c the tool sits in the mirror's recess (bounded by the inner contour) during operation
• C1-d camera + mirror arranged so camera field of view runs at least partly radially relative to the tool's longitudinal axis
• C1-e field of view captures the monitoring area from outside it, directed inward
• C1-f mirror + camera attachable to the tool so the FOV is always aimed at the monitoring area
• C1-g at least one component of the line-of-sight axis has a radial course relative to the annular mirror at the point of impact

2. Summary table — all 19 patent citations

3. Per-reference detail

Tier A — the four references that matter most

#9 — DE 203 07 305 U1 (ISRA Vision Systems AG) — 2003-05-09 / 2003-07-24 — §102(b)
Description (verified live against the DE document): "Optical monitoring device for monitoring the working area of a tool with at least one camera"; the device is fixable to the tool so the camera's field of view is always directed at the working area as the tool and/or workpiece moves; preferred versions have two, preferably three, cameras whose fields of view overlap. The document criticizes the alternative of tracking a fixed camera with dedicated control electronics.
Significance: This is the same-assignee ancestor of '533 and the record the '533 specification itself names as known art. It squarely discloses C1-a, C1-f (the "attachable to the tool / always oriented toward the monitoring area" limitation) and the subject matter of claim 10 (multiple overlapping cameras).
Anticipation verdict: Anticipates nothing in full. It discloses no mirror at all, hence no C1-b, C1-c (tool in the mirror recess), C1-d, C1-e, C1-g. It is a §102(b) reference for C1-f and claim 10 only, and the natural primary reference in a §103 combination — which is presumably how the examiner used it.

#15 — DE 10 2006 008 552 A1 (Kappner, Helmut A.) — pub. 2007-04-05 — §102(b) (family: EP 1 826 556 A2/A3)
Description (verified live): A testing device (Prüfeinrichtung) with at least one electronic camera whose optical axis is alignable coaxially with the symmetry axis of a workpiece (bottles, bottle caps, other hollow bodies), and a deflection mirror arranged in front of the objective that is formed as a hollow/ring mirror ("Hohl-/Ringspiegel") with a conical-surface or part-spherical reflecting surface ("kegelmantelförmige oder teilsphärische Reflexionsfläche"), concave, that reflects light emitted by the workpiece back toward the camera.
Significance: This is the only cited reference that discloses a concave annular ("ring") mirror in the optical path in front of the camera — i.e., the closest hit on element C1-b. The part-spherical variant is arguably a "circular section," touching claim 4's cross-section requirement.
Anticipation verdict: Anticipates nothing in full. It is a workpiece-inspection device, not a tool-activity monitoring device: the object in the recess is the workpiece being inspected, not the tool being monitored (C1-c fails); the imaging geometry is coaxial, not radial-to-tool-axis (C1-d, C1-g fail); and there is no tool-mounting (C1-f fails). Best characterized as a §102(b) reference for C1-b (and possibly claim 4), and a strong §103 partner to #9.

#16 / #17 — DE 10 2006 006 204 A1 and US 2007/0236565 A1 (Vision Tools Hard- und Software Entwicklungs-GmbH / inventor Hermann Tropf) — 2006-02-09 / pub. 2007-08-23 and 2007-10-11 — §102(b); US pub. also §102(e)
Description (verified live): Arrangement for optical monitoring of the working result of a tool and/or optical guidance of a tool, in which the tool (e.g., an application nozzle) lies approximately centrally in the beam path and a mirror arranged in a pupil of the lens system deflects the imaging beam path; a beam-splitting device and beam-deviating devices route image fields onto a working field so that the image fields enclose a field around the tool. The text expressly complains that prior multi-camera arrangements require several cameras and consume space transverse to the direction of travel.
Significance: These two are the same family (same priority day, same inventor/company), so they are one prior-art teaching. They hit the "look around the tool" concept — the functional goal of C1-c/C1-d/C1-e — but achieve it with pupil mirrors/beam splitters, not a concave annular mirror.
Anticipation verdict: Anticipates nothing in full — C1-b (concave annular mirror) is absent, and C1-g's impact geometry is not shown. Relevant as §102(b)/§102(e) art for C1-c/C1-e and claim 6 (a second, deflecting mirror), again a §103 partner rather than an anticipation.

#19 — DE 10 2007 031 835 A1 (Lensation GmbH) — 2007-07-07 / pub. 2009-01-08 — date-suspect
Description (verified live): "Arrangement for optical control of the working result of a tool and/or optical guiding of a tool" comprising a beam-splitting device and devices deviating beam paths onto a working field so that the image fields enclose a field around the tool. It shares substantial text with #16/#17 (it cites the same problem set: space requirements, shading, the need for a "Rund-Um-Blick").
Significance: Conceptually one of the closest disclosures of a 360°-around-the-tool viewing arrangement on a monitored tool, i.e., C1-c/C1-e.
Anticipation verdict: Anticipates nothing in full (no concave annular mirror → C1-b fails). And its §102 status itself is doubtful: it is a German publication dated 2009-01-08, i.e., after the 2008-06-06 priority/invention date and less than one year before the 2009-06-03 U.S. filing. A foreign published application gets no §102(e) date, so it is neither §102(a) (published after the invention) nor §102(b) (inside the one-year grace) unless the applicant's actual invention date is later than 2009-01-08 or the 2008 priority claim fails. I flag this because it is the kind of reference that looks like the best art but may not be available as §102 art at all. This should be checked against the file wrapper before any reliance.

Tier B — adhesive-bead / multi-camera monitoring family (§102(b))

#10 EP 1 697 061 A1 (Quiss GmbH) — 2003-12-23 / 2006-09-06; #11 US 2007/0292629 A1 (Quiss GmbH) — 2003-12-23 / 2007-12-20; #13 EP 1 701 801 A1 (Quiss GmbH) — 2003-12-23 / 2006-09-20; #12 US 2008/0024602 A1 (Jan Anders Linnenkohl) — 2003-12-23 / 2008-01-31.
Description: Method/device for recognizing a structure to be applied to a substrate (an adhesive bead/track) using several cameras directed at the applying unit and arranged around it (the '533 specification itself cites EP 1 697 061 B1 / EP 1 701 801 B1 as known). #12 covers automatic application and monitoring of the structure.
Claim mapping: C1-a, and the "cameras disposed around the applying unit" concept relevant to claim 10 (≥2 cameras) and C1-c/C1-e generally.
Verdict: No anticipation. No mirror (C1-b), no radial-strike geometry (C1-g); the cameras look directly at the applicator. These four share overlapping families (all priority 2003-12-23) and should be treated as one multi-camera prior-art teaching; art under §102(b) (and §102(e) for the US members).

Tier C — reflective/annular optics and metrology (secondary)

#14 — US 2008/0137088 A1 (OBE Ohnmacht & Baumgartner GmbH & Co. KG; inventor Christoph Wagner) — 2005-03-24 / 2008-06-12 — §102(e) (US filing 2006-03-23; granted as US 7,808,644 B2, later Cognex).
Description (verified live): Device for optically measuring shapes of objects/surfaces using an opaque scattering body and multiple switchable light sources (photometric deflectometry); the granted patent also claims conical interior/exterior specular mirrors that image the circumferential (side) area of an object into the camera, so a face and a lateral/circumferential area appear in one image.
Claim mapping: Conceptually relevant to C1-e (side capture) and to the "analyze the sides of a 3-D object" problem statement; mirror is conical, not a concave annular mirror in the claimed sense.
Verdict: No anticipation; §102(e) art as of its 2006-03-23 US filing (its 2008-06-12 publication postdates the §102(b) bar date, so only §102(e) is available).

#3 — US 6,775,012 B2 (Accu-Sort Systems) — 1995-06-02 / 2004-08-10 — §102(b). System for dimensioning objects using at least one light beam offset from a perpendicular to the object-supporting surface. Relevant background to claims 9 and 12 (3-D measurement / light-line triangulation); no mirror geometry. No anticipation.

#18 — DE 10 2006 036 586 A1 (Reiter, Mathias) — 2006-08-04 / 2008-02-14 — §102(b). Method for testing a three-dimensional, lengthwise-running contour (recess, irregularity) by recording longitudinal positions of the contour from different angles observed in side view toward the running direction. Relevant to the problem of lateral/side viewing (C1-e) and to 3-D measurement (claim 9). No anticipation.

Tier D — general optical/inspection background (weakest)

• #1 DE 400 0988 C1 (G.M. Pfaff AG) — 1990-01-16 / 1991-05-29 — §102(b). Sewing machine with stitch plate, needle bar and a scanning device having an optical sensor. C1-a only. No anticipation.
• #2 US 5,699,152 A (Alltrista Corp.) — 1995-04-03 / 1997-12-16 — §102(b). Electro-optical inspection system and method. No anticipation.
• #4 US 5,912,776 A (Mitsubishi Nuclear Fuel Co.) — 1996-10-24 / 1999-06-15 — §102(b). Interior image-receiving apparatus (remote/internal viewing). No anticipation.
• #5 DE 199 48 837 A1 (Mahlo GmbH) — 1999-10-11 / 2001-04-12 — §102(b). Continuous non-contact thickness measurement of an organic deposit using NIR at measurement and reference wavelengths. Measurement, not imaging/geometry. No anticipation.
• #6 US 2003/0020916 A1 (Shay Ghilai) — 2000-05-11 / 2003-01-30 — §102(b). Optical detection device. No anticipation.
• #7 US 2005/0099605 A1 (Georges Buchner) — 2000-09-15 / 2005-05-12 — §102(b). Interactive audio-visual system. Peripheral to the claim. No anticipation.
• #8 US 2010/0045988 A1 (SR-Instruments Oy) — 2002-11-05 / 2010-02-25 — §102(e) (US filing 2002-11-05). Synchronous optical measurement and inspection method and means. Publication postdates the bar date, so only its §102(e) date is available; it is still comfortably before 2008. Relevant to claims 8/9 (measurement/analysis), not the optics. No anticipation.

4. Non-patent literature (3 references)

5. Most relevant prior art — ranked, and the §102 bottom line

Ranked relevance to claim 1:

1. DE 203 07 305 U1 (same assignee) — the tool-mounted, always-aimed, multi-camera monitoring concept (C1-f, claim 10).
2. DE 10 2006 008 552 A1 / EP 1 826 556 A2 (Kappner) — the only concave ring/hollow mirror in front of the camera (C1-b, possibly claim 4).
3. DE 10 2006 006 204 A1 ≡ US 2007/0236565 A1 (Vision Tools/Tropf) — mirrors/beam-splitters arranged so the image fields enclose a field around the tool (C1-c/C1-e, claim 6).
4. DE 10 2007 031 835 A1 (Lensation) — same "around-the-tool" arrangement, but its §102 date is doubtful (published 2009-01-08, after the 2008 priority).
5. US 2008/0137088 A1 (OBE/Wagner) — reflective/annular optics for circumferential (side) imaging (§102(e), 2006-03-23).
6. EP 1 697 061 / EP 1 701 801 / US 2007/0292629 / US 2008/0024602 (Quiss/Linnenkohl) — multi-camera recognition around an applicator (claim 10).

§102 conclusion. On the citation record, no cited reference anticipates claim 1 — and therefore no cited reference anticipates any of claims 2–14, because all of them incorporate claim 1. Every reference is missing at least the core concave annular mirror in the optical path with the tool in its recess under the claimed radial-strike geometry (C1-b/C1-c/C1-d/C1-g). The cited art reads as a §103 combination set, most naturally:

• DE 10 2006 008 552 A1 (concave ring mirror in the optical path) as primary, in view of DE 203 07 305 U1 (mount the whole unit on the moving tool); and/or
• DE 203 07 305 U1 as primary in view of DE 10 2006 006 204 A1 / US 2007/0236565 A1 or DE 10 2007 031 835 A1 (deviating optics that create a field around the tool).

That is consistent with the patent having issued over these references.

Most likely §102-relevant (not merely §103) attacks, if any: none on claim 1 as a whole. The only genuinely anticipatory-flavored items are the individual limitations — #9 for C1-f, #15 for C1-b — which cannot anticipate a claim standing alone.

4b. Cross-references and inconsistencies carried forward

• Claim 8 antecedent basis (flagged in the Patent summary): claim 8 recites "the illumination device" and "the light line device," which have no antecedent in its dependency chain (6→1); those elements are only introduced in claims 11 and 12. Practically, claim 8's scope is ambiguous for §§102/103 mapping — art is measured against the later-introduced elements in claims 11/12 (flash/continuous illumination; light-line triangulation). This matters because the best "illumination" and "light-line" art in the cited set is generic (e.g., #3 US 6,775,012 for triangulation/dimensioning).
• Camera-count inconsistency (flagged in the Patent summary): the Detailed Description says "only one camera … is shown," then "comprises two cameras," then discusses "the three cameras." This does not affect prior-art dating but does affect any §§112/§103 argument that the claims cover multi-camera arrangements (claim 10).
• Date discrepancy already flagged (Apr 26, 2026 vs Sep 29, 2026): the calendar date does not affect any prior-art date above, since all dating is fixed by the 2008/2009 documents.
• Reference-date note on #15 (Kappner): the supplied citation table lists DE 10 2006 008 552 A1 with a first date of 2005-09-23 and publication 2007-04-05, while the European family member EP 1 826 556 A2 claims priority to DE 10 2006 008 552 with a priority date of 2006-02-22 (and EP 1 826 556 A2 published 2007-08-29). The publication date that controls §102(b) is 2007-04-05, which is comfortably before the bar date; the internal priority-date discrepancy does not change the analysis and is flagged only for accuracy.

6. Caveats / uncertainty (stated explicitly)

• I examined the citation lists, not the underlying office actions. I cannot state which claims the examiner actually rejected or how many references were combined, because that requires the file wrapper (USPTO PatentCenter / Image File Wrapper), which my tooling does not reach. The p

I'll ground this §103 analysis in the actual cited references. Let me pull what public sources show about the key prior-art documents named in the patent's citation lists.

§103 Obviousness Analysis — US 9,091,533 B2 ("Optical surveillance device")

Method / evidentiary note. Per the operating rules I treat the supplied full text of US 9,091,533 (claims, description, and the "Patent Citations (19)" / "Non-Patent Citations (3)" lists) as authoritative. For the substantive teachings of the cited documents I relied on the publicly available texts of those documents returned by search (identified by URL below). Where I only have an abstract-level or machine-translated statement, I say so. I do not have the file wrapper/Office actions for 12/477,383, so this is an independent obviousness analysis, not a reconstruction of what the examiner actually did or what the applicant argued. (Flagging the previously noted date discrepancy: my system context says 2026‑09‑29 while the task header says 2026‑04‑26; nothing here turns on it.)

1. Person Having Ordinary Skill in the Art (POSITA)

A POSITA here is a machine-vision engineer with a B.S. in electrical/optical/mechanical engineering (or equivalent) and roughly 2–3 years of hands-on experience designing industrial optical inspection or robot-guidance systems for automated dispensing/welding/joining, or a higher degree with less experience. That person is fully familiar with (a) central-projection camera geometry, (b) catadioptric/reflective imaging using ring and spherical mirrors, (c) laser light-section (triangulation) sensors, and (d) tool-mounted "online" sensor heads for adhesive-bead inspection — all of which are the standard toolkit of the field. This is a mature, predictable art; '533 itself describes the triangulation calibration as "known per se" (Description).

2. Claim 1 decomposed

The point of novelty is thus a combination: annular concave mirror + tool-through-the-recess + radial viewing + tool-mounted deployment.

3. Key reference teachings (primaries)

• DE 10 2006 008 552 A1 (Kappner; pub. 2007‑04‑05; EP 1 826 556 A2/A3) — The most structurally on-point reference. A single electronic camera whose optical axis is aligned coaxially with the symmetry axis of a rotationally symmetric workpiece; an "Umlenkspiegel … als Hohl- bzw. Ringspiegel" (hollow/ring mirror) with a coaxial mid-axis and a cone-mantle-shaped or part-spherical reflecting surface; the mirror's smallest diameter is smaller than the workpiece width, i.e. the workpiece sits within/through the ring opening (claims 1–2, 14–20 of the DE/EP text). It explicitly notes that hollow/ring mirrors had already been used to make a cylinder's entire lateral surface visible in one camera image ("ringsherum in einem Kamerabild sichtbar machen… über einen kegelstumpfförmigen Hohlringspiegel"), and it uses a mirror tilt angle and two mirrors of different tilt to capture different flanks. Images go to an image-processing system and are evaluated from ≥2 perspectives ([0002]–[0006], EP 1 826 556 A2). Source: patents.google.com/patent/DE102006008552A1/en, freepatentsonline.com/DE102006008552.html, patentimages…/EP1826556A2.pdf.
• DE 203 07 305 U1 (ISRA Vision Systems AG; 2003‑07‑24) — Same assignee as '533. Optical monitoring device fixable to the tool so the camera's field of view is always directed at the working/monitoring area even as tool and/or object move, with two, preferably three cameras whose fields of view overlap. (Cited on the face of '533 and quoted in the EP 2131145 B1 counterpart at [0007].)
• DE 10 2007 031 835 A1/B4 (Lensation GmbH; 2009‑01‑08) — Arrangement for optical control of a tool's working result / optical guidance of a tool, comprising a beam-splitting device and beam-deviating devices arranged so that the image fields "enclose a field around the tool" (umschließen ein Feld um das Werkzeug). Its background expressly states that for "online" control the sensor is mounted near the application nozzle and fixed relative to it, that a "Rund-Um-Blick" (all-round view) is required to avoid rotating the sensor on curved beads, that light-section triangulation with laser lines is the preferred height control, and that the known multi-camera solutions are bulky and suffer shading problems. Sources: patents.google.com/patent/DE102007031835A1/en, …/DE102007031835B4/en.
• US 5,699,152 (Alltrista) — Above-the-object inspection with an "upwardly concave truncated ellipsoidal" annular reflecting member (claim 11) that "unfolds" the three-dimensional circumferential surface into a two-dimensional image (claim 15) for a camera located above; annular LED illuminator (claim 4); planar image-splitting mirrors at ~45° (claims 6–8); vision computers/TV cameras (claims 3, 9). Sources: patents.google.com/patent/US5699152, everypatent.com/comp/pat5699152.html.
• US 5,912,776 (Mitsubishi Nuclear Fuel) — Ring-shaped reflecting mirror with a conical inner surface, whose section along the object axis may be concave or convex, mounted coaxially between an above-mounted camera and the object so a single camera sees the object's sides "in an instant" without rotation; circular illuminating means surrounding the image receiver, optionally a stroboscope/light-flash. Sources: patents.google.com/patent/US5912776, patentimages…/US5912776.pdf.
• Quiss EP 1 697 061 B1 / EP 1 701 801 B1 (and US 2007/0236565, Tropf, "Optical Guide and Online Control of a Tool") — Sensor unit with multiple cameras each directed at the application unit and arranged around it; image-processing/3-D evaluation. (Cited on the face of '533; summarized at EP 2131145 B1 [0008] of the same family.)
• NPL — Kinoshita et al., "Thin compound eye camera with a zooming function by reflective optics" (MEMS 2005) and Li, "Monitoring Around a Vehicle by a Spherical Image Sensor" (IEEE ITS, Dec. 2006) — reflective-optics and spherical-mirror 360° surround imaging; Asada et al. (IEEE IAS 1997) — factory-automation image monitoring.
• Supplementary (NOT on the face of '533): DE 20 2005 003 352 U1 — sensor arrangement with three cameras whose common field of view covers 360° around the tool, a lighting device around the cameras, and a passage (Durchlass) for the tool, with each camera axis inclined radially toward the passage axis. I flag this as extra-record art; content taken from the published description returned by search (Source: patentimages…/DE202005003352U1.pdf) and it should be verified before being relied on.

4. Grounds of obviousness

Ground 1 (primary): Kappner '552 + ISRA '305 + Lensation '835

Motivation (KSR‑compliant, drawn from the references themselves, not hindsight):

1. Same problem, same field. '533's own background states the problem: a single camera above the work zone cannot see the flanks of a 3‑D bead, so 360° coverage conventionally "requires the use of multiple cameras." Lensation states the identical problem for on-line bead/nozzle control, and proposes fewer cameras with beam-deviating optics to obtain the all-round view. That is an express teaching, in the prior art, to replace a multi-camera surround view with a mirror-based single/lesser-camera surround view — the very substitution claimed.
2. A ring mirror around a body's axis to see its sides 360° with one camera was admitted old art (Kappner's background; US 5,912,776; US 5,699,152). Substituting a known annular concave mirror for the cluster of cameras is a simple substitution of a known element to obtain a predictable result (KSR).
3. Tool-mounting was the assignee's own known solution (ISRA '305: fix the sensor to the tool so the FOV always tracks the working area). A POSITA seeking to improve the tool-mounted sensor of ISRA '305/Quiss by reducing camera count would naturally mount the known ring mirror coaxially about the applicator, since the applicator is an elongated member passing through the ring's opening — exactly the "Durchlass"/recess arrangement already known (ISRA '305/Quiss/de facto in DE 20 2005 003 352 U1).
4. Predictability/finite solutions. The mirror geometries available were few and known (frustoconical, spherical, ellipsoidal), as were the camera placements (on-axis vs. radially offset inside the ring). KSR step 6 (design incentives) and the "obvious to try" rationale apply.
5. No teaching away. Lensation criticizes multi-camera arrangements but points toward mirror/deviating-optics solutions, i.e., away from the prior multi-camera practice and toward claim 1.

Net: every element of claim 1 is disclosed or suggested by Kappner+ISRA, with Lensation supplying the express motivation to make the tool-mounted, mirror-based, all-round surround configuration. Claim 1 would be obvious.

Ground 2: US 5,699,152 (Alltrista) + ISRA '305 (+ Quiss)

Alltrista discloses an annular, upwardly concave, ellipsoidal/truncated-ellipsoidal mirror above the object that "unfolds" the object's 3‑D circumferential surface for a single above-mounted camera, plus 45° planar folding mirrors, an annular LED illuminator, and vision computers. ISRA '305/Quiss supply the tool-mounted, applicator-surrounded deployment. Motivation is again the explicit problem statement (single camera, all-round side view) plus space constraints near the nozzle. Points L1, L2, L3 (annular member surrounding the object), L4, L5, L6 are all met; L7 is met by the geometry of a radially reflective annular mirror (in Alltrista the manifold view is radial by construction) or by an obvious radial offsetting of the camera.

Ground 3: US 5,912,776 (Mitsubishi) + ISRA '305 + Lensation '835

Mitsubishi expressly discloses the ring-shaped mirror with conical inner surface, concave-or-convex axial section, coaxial mounting, ring illumination surrounding the receiver, and flash illumination, obtained "in an instant… eliminating rotation." Combined with ISRA '305 (tool-attached, always-aimed) and Lensation '835 (all-round view for a nozzle-mounted sensor), the claim-1 combination follows with the same motivations as Ground 1; Mitsubishi adds the concave-section and illumination teachings directly useful for dependent claims 4 and 11.

Ground 4 (secondary): Lensation '835 + Kappner '552 as a system

Lensation teaches the objective (all-round, single/nozzle-mounted optical control using beam deviation around the tool) and directs the skilled person to deviating optics; Kappner teaches the specific concave ring-mirror implementation. The "predictable results" and "known technique to improve similar device" rationales apply.

5. Dependent claims 2–14

6. What the patentee would argue, and the weak spots in my grounds

• "Different art / different problem." Kappner, Mitsubishi and Alltrista inspect workpieces at fixed stations; '533 inspects a moving tool's work product. Counter: ISRA '305 (same assignee) and Quiss/Lensation are squarely in the tool-mounted/adhesive-bead field and supply L6, so the combination is within one field of endeavour, and the motivation to reduce camera count near a nozzle is express.
• "The reference cameras are on-axis; claim 1 requires a radial strike on the mirror." This is the strongest non-obviousness argument for L7. It is weakened by (i) Kappner satisfying the mirror-image geometry of claim 5, (ii) the radially-arranged camera axes already known around a tool passage (DE 20 2005 003 352 U1 — extra-record, verify), and (iii) the absence of any claimed optical correction, implying the off-axis radial arrangement is a predictable design choice, not an inventive one.
• "Tool in the recess" vs. "workpiece in the recess." Minor: the recess/passage-for-a-nozzle concept is taught in the tool-monitoring art; the substitution is one of a known element with a known function.
• Caveats on my own grounds. (a) I have not read the EP 1 697 061/EP 1 701 801 claim sets in full, only the family's summary of them; (b) DE 20 2005 003 352 U1 is not on the face of '533 and its content is search-derived — verify before relying on it; (c) I have not identified a secondary-considerations record (no evidence of unexpected results, commercial success nexus, or licensing in the documents reviewed), so no rebuttal of the KSR rationale appears in the record I can see.

7. Bottom line

On the record reviewed, claim 1 most likely would have been obvious over DE 10 2006 008 552 A1 (Kappner) in view of DE 203 07 305 U1 (ISRA Vision) — the ring-shaped concave mirror for 360° side viewing around a body's axis, plus tool-attached, always-aimed mounting — and further in view of DE 10 2007 031 835 A1 (Lensation), which supplies the express motivation to replace multi-camera surround sensing with mirror/deviating-optics "all-round" sensing on a nozzle-mounted sensor head. Equivalent alternative grounds exist over US 5,699,152 or US 5,912,776 combined with DE 203 07 305 U1. Claims 2–14 are each rendered obvious by the same references plus the additional mapping in §5, with claims 5/13 and 4 being the most directly met (Kappner/Alltrista) and claim 7 being the most vulnerable to a "routine design choice" critique by a patentee.

Not asserted as a legal conclusion. This analysis is based on the face-of-record citations and the public texts of the references; a definitive §103 opinion (or an IPR invalidity position) would require the file wrapper, the full claims of EP 1 697 061 / EP 1 701 801, verification of DE 20 2005 003 352 U1, and a POSITA declaration.