US 12332352

I'll search for US patent 12332352 in the USPTO database and CAFC dockets.

Let me search for more authoritative details on the USPTO site and verify the CAFC docket search.

I have confirmed the bibliographic data. Now let me search for the claims text and abstract of the granted patent US12332352B2.

I have verified the bibliographic data through USPTO-adjacent sources (patent leaderboard listings and the HT Syndication/US Fed News issue notice). I was not able to retrieve the verbatim text of the granted claims within my search step limit, and the full patent text provided in the prompt is truncated before the claims section. I'll flag claim-level uncertainty explicitly below.

US Patent 12,332,352 (US12332352B2) — Verified Data

Abstract: The published application/specification describes a detector for identifying at least one material property (e.g., scattering coefficient, translucency, transparency, deviation from Lambertian surface reflection), configured in particular for optical detection of biological tissue such as human skin. It comprises a sensor element (matrix of optical sensors, e.g., CMOS) for recording a reflection image of a light beam originating from an object, an illumination source (e.g., a laser/DOE projecting an illumination pattern such as a point cloud), and an evaluation device that applies at least one distance-dependent image filter (e.g., depth-from-photon-ratio and/or depth-from-defocus) to derive a distance feature and at least one material-dependent image filter (e.g., luminance, spot-shape, smoothness/speckle, gray-level-occurrence contrast/energy/homogeneity, Law's energy, threshold-area filters) to derive a material feature, then determines the material property and/or longitudinal coordinate from those features.

Plain-Language Overview of Independent Claims — WITH UNCERTAINTY

Important caveat: I could not retrieve the granted claim set verbatim (Google Patents full text for this grant was truncated before the claims, and no claims database surfaced in my searches). The specification indicates multiple independent claim "aspects" (a detector, a detector system, a human-machine interface, an entertainment device, a tracking system, a camera, a scanning system, methods, and uses), which is typical for this trinamiX family (see EP 3 938 802 B1, same-title counterpart). Based on the specification's repeated statement of the invention, the independent claims most likely cover the following, but each should be verified against the USPTO full text before reliance:

1. Detector claim (likely claim 1): A detector for identifying at least one material property m, comprising (a) a sensor element with a matrix of optical sensors configured to record a reflection image of a light beam originating from at least one object (e.g., a reflection beam profile of an illumination feature); (b) at least one illumination source configured to illuminate the object (e.g., projecting an illumination pattern); and (c) an evaluation device configured to determine a distance feature by applying a distance-dependent image filter to the reflection image and to determine a material feature by applying a material-dependent image filter to the reflection image, and to determine the material property and/or the object's longitudinal coordinate based on these features — solving the problem that most beam-profile-analysis filters depend on both distance and material.

2. Detector system claim: A system comprising the detector plus at least one beacon device attachable to/holdable by/integratable into the object to direct a light beam toward the detector.

3. Human-machine interface / entertainment device / tracking system / camera / scanning system claims: Combinations incorporating the detector (or detector system) for position/gesture-based information exchange, entertainment functions, tracking, imaging, or scanning, with the detector also identifying a material property (e.g., distinguishing skin from non-skin for gesture/security applications).

4. Method claim(s): Corresponding method for determining at least one material property by recording a reflection image, applying a distance-dependent image filter to obtain a distance feature, applying a material-dependent image filter to obtain a material feature, and determining the material property (and/or longitudinal coordinate) therefrom.

5. Use claims: Various uses of the detector, e.g., in security technology, gaming, traffic, production, agriculture, cosmetics, medical technology, or sciences (including skin detection).

Recommendation: Treat the claim-by-claim wording as unverified. The bibliographic facts above are confirmed by multiple independent sources consistent with the Google Patents record provided. For authoritative independent-claim text, pull the full granted PDF from USPTO Patent Center (application US18/743,352) or Google Patents claim view for US12332352B2, which I could not fully access in this session.

I'll search for litigation involving US patent 12332352 across multiple sources.

Let me check dedicated patent litigation databases directly.

Based on my searches across general web, patent-litigation-oriented queries, and the Unified Patents portal domain, I found no known litigation involving US patent 12332352.

Findings

Patent identification (for context, not litigation):

• US 12332352 B2, titled "Detector for identifying at least one material property," assigned to TrinamiX GmbH, granted 2025-06-17 (inventors include Friedrich Schick, Peter Schillen, Patrick Schindler, et al.). This patent is very recently granted (June 2025), which is consistent with the absence of any filed litigation as of today's date (April 26, 2026).

Search results:

• Searches for "12332352" patent litigation, "US 12332352" lawsuit, "12332352" TrinamiX patent case, and "12332352" site:portal.unifiedpatents.com returned no litigation records.
• The only hits were patent-inventory pages (e.g., patentleaderboard.com listing the patent) and an unrelated Soviet patent number (SU 1233235A1) that merely resembles "12332352" numerically — I excluded that as it is not the target US patent.

Conclusion

No litigation (district court cases, ITC proceedings, IPR/PGR challenges, or CAFC appeals) is known to involve US patent 12332352 based on available public sources. Given the patent's June 17, 2025 grant date, it remains within the early post-grant window where litigation is uncommon. I cannot, however, rule out sealed or very recently filed complaints that may not yet be indexed in public search engines; a PACER or Unified Patents caselist query by patent number would be the definitive check.

Proceedings overview

Zero AIA trial proceedings are on file for US 12332352 B2. The USPTO Open Data Portal (the canonical structured source) reports no IPR, PGR, or CBM petitions associated with the patent, and targeted web searches (including PTAB-specific queries against the patent number, title, and assignee TrinamiX GmbH) surfaced no petition, institution decision, settlement, or Final Written Decision. There is no proceeding-by-proceeding breakdown because there are no proceedings to break down. For a defendant, the posture is neutral-to-cautious: the patent is young (granted 2025-06-17), fully intact, and completely untested before the PTAB — there is no IPR track record, favorable or unfavorable, to lean on.

Note on timing windows relevant to why this is unsurprising:

• Grant date: 2025-06-17.
• PGR window: 9 months from grant — i.e., through ~2026-03-17 — has now closed (today is 2026-09-04). Any post-grant challenge filed today must be an IPR (available from 2026-03-18 onward, subject to the § 315(b) one-year bar if a defendant has been sued).
• The Google Patents record lists an anticipated expiration of 2040-03-13, so the patent has a long remaining enforcement runway — which typically invites challenges once it is actually asserted at scale.

Strategic summary

Claim status: all CANCELED vs. SUSTAINED vs. UNTESTED. Every claim of US 12332352 — independent and dependent alike — is UNTESTED. No petition has been filed, no ground has been briefed, and no claim has been canceled or confirmed. The patent emerges from its post-grant window with its full claim set intact: claims covering the detector configured to record a reflection image, apply a distance-dependent image filter (depth-from-photon-ratio / depth-from-defocus) to derive a distance feature ƒ₁z, apply a material-dependent image filter (luminance, spot-shape, GLCM-based, Law's energy, threshold-area, etc.) to derive a material feature ƒ₂m, and jointly determine material property m and longitudinal coordinate z from those features.

Estoppel landscape. With no proceedings on file, § 315(e)(2) estoppel binds no one. No petitioner has been estopped, and no defendant facing assertion today is foreclosed from raising any ground. The practical consequence: the entire prior-art universe remains available to a future IPR petitioner — every § 102/§ 103 ground over the underlying 2019-03-15 priority chain (including the family members cited in the specification itself: US 2016/0206216 A1, US 2016/155006 A1, DE 198 46 619 A1, CN 108 363 482 A, US 2018/033146 A1, WO 2014/097181 A1, WO 2014/198629 A1) is free game. A challenger should still do a diligent search for better art than what the Examiner cited, but there is no estoppel constraint whatsoever.

Pattern signals. None exist yet: no repeat petitioner, no defensive-aggregator involvement (no Unified Patents or similar entity in any chain), and no signal on how TrinamiX GmbH litigates PTAB challenges. The only observable pattern is the assignee's active prosecution of this family internationally — e.g., the parallel Japanese family member JP 2025 505 947 A ("Improved material detection with stereo beam profile analysis") — indicating the "detector for identifying a material property" concept is commercially important and likely to be asserted. A well-asserted patent in a valuable family typically attracts IPRs within a year or two of first significant litigation; the absence of any PTAB activity today most likely reflects the absence of any high-stakes assertion to date, not weakness in the prior art.

Recommended next steps

• If you are a defendant facing assertion of US 12332352 today, there is no FWD to cite and no canceled claim to wield. Your leverage is procedural and structural instead:
  1. Confirm the § 315(b) clock. If you were served with a complaint more than 12 months ago, you are time-barred from petitioning; if you are inside that window (or have not been sued), an IPR petition is available and the PGR window having closed means IPR is the only PTAB route.
  2. Run a fresh prior-art search now. Because the patent is untested, the quality of the art — not PTAB precedent — will decide your challenge. Focus on 2018-and-earlier optical material-classification and beam-profile-analysis references (skin detection, depth-from-defocus, depth-from-photon-ratio, structured-light systems) suitable for § 103 combinations against the independent claims' filter-pairing and joint distance/material-determination language.
  3. Verify the claim set you face. If the asserted claims are the narrow material-dependent-filter embodiments, consider whether a § 101 challenge (abstract idea / patent-ineligible material-classification) is worth preserving in district court, since PTAB grounds would be limited to §§ 102/103.
• Monitor the PTAB docket at the USPTO PTAB End-to-End (E2E) system (https://ptab.uspto.gov) and the USPTO Open Data Portal's trial-proceeding API for any late-indexed petition. Given today's date, a petition filed now would have its institution decision due roughly 6 months after filing and a Final Written Decision due within 12 months of institution — but no such proceeding is pending or on file as of 2026-09-04.
• If no PTAB activity exists, say so plainly — as done here. The absence is itself informative: a June 2025 grant means the patent survived its 9-month PGR window unchallenged, so any future attack must come via IPR, and the first mover who files will define the art landscape that later petitioners (and the district court, via the FWD) will rely on.

No proceeding numbers are provided because none exist; nothing in this report should be read as identifying any IPR, PGR, or CBM — the USPTO ODP record and all searched sources are empty, and I have not fabricated any.

I could not complete direct USPTO Assignment Center queries within my search limit, so reel/frame numbers and correspondent-of-record data below are unverified — flagged explicitly rather than guessed. The three conveyances are documented in the Google Patents legal-events feed reproduced in the patent text supplied with the task.

Inventors

All 12 inventors are German (per the patent record). Employers at time of filing, as determinable from the recorded assignment events (recorded 2024-06-20), split into two groups:

Unusual pattern check: None. No inventor-departure or portfolio-fire-sale pattern is indicated. Group 2 is the classic "external engineering partner contributes rights through its own entity" pattern — Hengen Ingenieurbuero GbR (an Ingenieurbüro, i.e., an engineering consultancy in German civil-law partnership form) is the vehicle through which two named inventors channeled their rights before they vested in trinamiX GmbH the same day.

Original assignee

• Original assignee (per Google Patents "Original Assignee" field and chain of title): trinamiX GmbH, Ludwigshafen am Rhein, Germany.
• Line of business: trinamiX GmbH is a technology/operating company (BASF-group spin-off founded 2015) developing optical sensing — NIR/mobile spectroscopy, material identification, and 3D depth sensing based on beam-profile analysis (its "Hertzstueck" IR sensors are cited in the specification). It employs the 10-inventor group above and pursues product/partner commercialization rather than pure licensing. Confidence note: this business characterization rests on background knowledge; I could not re-verify it with live search this session.
• Product embodying the claims: Plausible but not verified in this session. The claimed detector (distance-dependent + material-dependent beam-profile filters) tracks trinamiX's commercial beam-profile-analysis and skin/material-detection roadmap; however, no specific commercial product mapping to these claims was confirmed.
• Current status: Operating (active BASF-group entity). No bankruptcy, dissolution, or acquisition signal in any source accessed.

Assignment timeline

The USPTO Assignment Center could not be queried directly in this session, so reel/frame numbers and correspondents of record are unverified. The three conveyances below are taken from the Google Patents legal-events feed (all bearing USPTO recordation date 2024-06-20, contemporaneous with the 2024-06-14 filing of US application 18/743,352 — a continuation in the 2019-03-15 priority family). Execution dates are not shown in that feed.

• execution date not shown / recorded 2024-06-20 — Reel/frame not retrievable this session

  • Conveyance: Assignment of Assignor's Interest
  • Assignor: Hengen Ingenieurbuero GbR
  • Assignee: trinamiX GmbH
  • Correspondent: not retrievable this session
  • Context: Entity-level assignment vesting the two external-inventor contributions in trinamiX GmbH (chain-of-title step at US filing).

• execution date not shown / recorded 2024-06-20 — Reel/frame not retrievable this session

  • Conveyance: Assignment of Assignor's Interest
  • Assignor: Lars Diesselberg; Heiko Hengen (individuals)
  • Assignee: Hengen Ingenieurbuero GbR
  • Correspondent: not retrievable this session
  • Context: Individual inventors → their engineering-partnership entity; the feeder step to the assignment above.

• execution date not shown / recorded 2024-06-20 — Reel/frame not retrievable this session

  • Conveyance: Assignment of Assignor's Interest
  • Assignor: Robert Send; Christian Bonsignore; Jakob Unger; Ingmar Bruder; Michael Eberspach; Christian Lennartz; Andre Schmidt; Friedrich Schick; Peter Schillen; Patrick Schindler
  • Assignee: trinamiX GmbH
  • Correspondent: not retrievable this session
  • Context: The ten trinamiX-affiliated inventors assigning their interests directly to their employer at US filing.

Net result: Title vests in trinamiX GmbH via three same-day, same-event chain-of-title recordations. There are no recorded post-issuance assignments (nothing after the 2025-06-17 grant), which ordinarily means the original assignee still owns the patent. If Assignment Center search by patent number 12332352 confirms only these three conveyances, that is the complete record.

Timeline diagram

timeline
    title Ownership of US 12332352
    2019 : Priority application filed
    2024 : US continuation filed
         : Inventor assignments recorded
         : Title vested in trinamiX GmbH
    2025 : Patent granted

NPE / troll-pattern signals

1. Shell-entity transfer — not present. The only transferees are trinamiX GmbH (a BASF-group operating company with R&D staff and product activity) and Hengen Ingenieurbuero GbR (a two-person engineering partnership of named inventors that assigned onward the same day, leaving no residual holder). No "IP / Holdings / Licensing" LLC, no registered-agent address, no single-purpose licensing entity appears anywhere in the 2024-06-20 recordation set. Caveat: I could not pull the underlying recorded documents to inspect addresses.

2. Known asserter in the chain — not present. Neither trinamiX GmbH nor Hengen Ingenieurbuero GbR matches any entity on the Acacia/Marathon/IV/IPNav/Wi-LAN/Conversant/Vringo/Pendrell/Innovatio/MPHJ/Spangenberg-type lists, nor any Unified Patents/RPX high-frequency-plaintiff roster I can identify. No litigation naming US 12332352 is known (see prior litigation summary — none found as of today).

3. Repeat correspondent across the chain — unclear. Correspondent-of-record data for the three 2024-06-20 conveyances could not be retrieved this session (USPTO Assignment Center not directly queryable within my search limit). This is the single most useful check for hidden NPE coordination, and it remains unverified — recommend pulling reel/frame records by patent number 12332352 at the USPTO Assignment Center to confirm whether one correspondent filed all three.

4. Cascading transfers — not present. Three conveyances recorded on one day (2024-06-20) at a single filing event are a chain-of-title recordation, not chained LLC transfers over <24 months. No sequence of assignee-to-assignee conveyances exists.

5. Pre-litigation transfer — not present. No infringement litigation naming this patent is known, so no transfer timed within six months of a first suit can be identified. The 2024-06-20 recordations align with the 2024-06-14 US filing, not with litigation.

6. Bankruptcy fire-sale — not present. Neither trinamiX GmbH nor Hengen Ingenieurbuero GbR shows any Chapter 7/11 or equivalent insolvency proceeding in the sources accessed.

7. Privateering — not present. No transfer from an operating company to an asserting NPE exists in the record; the direction of transfer is inventors → operating company, the opposite of privateering. No SEC cross-reference applies (BASF SE is not a US SEC registrant; the 10-K/8-K check is not relevant here).

8. Defensive aggregator (anti-NPE) — not present. The chain does not terminate at RPX, AST, LOT, Unified Patents, or OIN; it terminates at operating assignee trinamiX GmbH.

Verdict

Insufficient data.

Justification: The only recorded conveyances are the three standard inventor/entity-to-assignee assignments recorded 2024-06-20 at the filing of US 18/743,352, vesting title in trinamiX GmbH — an operating BASF-group company, not an NPE — and no post-grant transfer exists (recordation dates 2024-06-20; grant 2025-06-17). No NPE signal is affirmatively present, but because I could not verify reel/frame numbers, correspondent identities, or the underlying recorded documents from the USPTO Assignment Center in this session, the formal verdict is insufficient data rather than a clean operating-company/defensive classification. For verification: search by patent number at https://assignmentcenter.uspto.gov/ (patent 12332352) and confirm the three 2024-06-20 conveyances and their correspondents.

Scope, sources, and confidence caveat

Before the reference-by-reference analysis, two transparency notes, consistent with the earlier sections of this file:

1. I could not retrieve the official USPTO "References Cited" (front-page citation list) or the verbatim granted claims of US12332352B2 in this session — the Google Patents full text supplied in the brief is truncated before both sections, and my search budget expired before I could pull the USPTO Patent Center record. The claim text of the same application (US18/743,352, pre-grant publication US20240377533A1) was located via a secondary aggregator, so claim 1 below is quoted from the published application of this exact patent. Independent-claim wording should still be re-verified against the granted B2 text.
2. The most reliable surrogate for the citation list is the granted European family member EP3938802B1 / EP3938802A1 (same priority date 2019-03-15, same WO2020187719A1 PCT parent, same title), whose Google Patents record lists "Patent Citations (7)." The US B2 front page almost certainly mirrors this PCT search-report list (or a subset), but I flag this as an inference, not a verified USPTO extraction.

All references below have publication dates before the effective filing date relied on (2019-03-15) and therefore are available as prior art under AIA 35 U.S.C. § 102(a)(1) (and, where the underlying application was filed by another and later published, § 102(a)(2)). Note the commonly-owned-entity caveat for the BASF/trinamiX references (potential § 102(b)(2)(C) disqualification for the § 102(a)(2) prong only; they remain usable under § 103 and against non-common claims).

Claim anchor used for the anticipation analysis (claim 1 of US20240377533A1, the application that granted as US12332352B2):

"A detector for identifying at least one material property m comprising … at least one sensor element comprising a matrix of optical sensors … configured for recording at least one reflection image of a light beam originating from at least one object; … at least one illumination source …; and at least one evaluation device configured for determining the material property by evaluation of at least one beam profile of the reflection image, wherein the evaluation device is configured for determining at least one distance feature φ1z by applying at least one distance dependent image filter ϕ1 to the reflection image, wherein the distance dependent image filter is at least one filter selected from: a depth-from-photon-ratio filter; a depth-from-defocus filter; or a linear combination thereof; or a further distance dependent image filter ϕ1other which correlates … by |ρ| ≥ 0.40 …; wherein the evaluation device is configured for determining at least one material feature ϕ2m by applying at least one material dependent image filter ϕ2 to the reflection image, [and] for determining a longitudinal coordinate z and the material property m by evaluating the distance feature ϕ1z and the material feature ϕ2m."

Cited references (as listed for the EP/family record) and § 102 assessment

1. DE 198 46 619 A1 (DE19846619A1)

• Full citation: DE 19846619 A1, Byk-Gardner GmbH, "Structured surface appearance quality determining equipment" (Gerät zur Bestimmung der Erscheinungsqualität strukturierter Oberflächen)
• Priority/filing date: 1998-10-09 | Publication date: 2000-04-27
• Description: Evaluates an electrical measurement signal from a photosensor array to derive a "structure code" characterizing a structure-dependent characteristic of a measurement surface (surface-appearance/structure quality metrology).
• Anticipation under § 102: None of claim 1. It addresses surface structure/appearance quality, not identification of a material property from a reflection beam profile, and discloses neither a depth-from-photon-ratio nor a depth-from-defocus distance filter nor a material-feature filter combination. At most it is background for dependent-claim limitations reciting surface-type properties (e.g., roughness/surface-property features); it would not alone anticipate any full claim.

2. WO 2014/097181 A1 (WO2014097181A1)

• Full citation: WO 2014/097181 A1, BASF SE, "Detector for optically detecting at least one object" (US family member e.g. US 2015/0316692 A1)
• Priority date: 2012-12-19 | Publication date: 2014-06-26
• Description: Foundational trinamiX/BASF optical detector with a matrix of optical sensors and an evaluation unit deriving a longitudinal coordinate of an object from a combined (quotient) signal of edge vs. center information of a light spot — i.e., the depth-from-photon-ratio technique, including the sensor element and illumination source structure.
• Anticipation under § 102: Not claim 1 in full. It satisfies claim 1's structural elements (sensor matrix, illumination) and the distance-feature prong (depth-from-photon-ratio filter), but it does not disclose any material-dependent image filter or the determination of a material property m by evaluating a material feature together with a distance feature. It is a strong § 103 primary reference and would anticipate any (hypothetical) claim limited to distance measurement alone. (Common ownership with the applicant may also disqualify it under § 102(b)(2)(C) for the § 102(a)(2) prong.)

3. US 2017/0363465 A1 (US20170363465A1)

• Full citation: US 2017/0363465 A1, BASF SE, "Optical detector"
• Priority date: 2014-12-09 | Publication date: 2017-12-21
• Description: BASF optical-detector continuation in the same photon-ratio/distance-detection family. I could not retrieve the full text in this session; based on family context it concerns optical detection of an object (distance/position measurement using light-spot evaluation).
• Anticipation under § 102: Not claim 1 in full — same deficiency as reference 2 (no material-feature filter / material-property identification). Would not anticipate the claimed detector claim or any material-identification claim standing alone.

4. US 2018/0276843 A1 (US20180276843A1)

• Full citation: US 2018/0276843 A1, BASF SE, "Optical detector"
• Priority date: 2014-12-09 | Publication date: 2018-09-27
• Description: Later-published BASF "Optical detector" application in the same family as reference 3 (distance detection from a light beam/spot profile; full text not retrieved this session).
• Anticipation under § 102: Not claim 1 in full. Same analysis as references 2–3: discloses distance-feature technology only, lacks the material-dependent filter and material-property output limitation.

5. US 2018/0033146 A1 (US20180033146A1; cited in the specification in the variant form "US 2018/033146 A1" — same document)

• Full citation: US 2018/0033146 A1, Michael Bleyer et al. (Intel Corp.), "Reflectivity map estimate from dot based structured light systems"
• Priority date: 2016-07-27 | Publication date: 2018-02-01
• Description: Determines a depth map from a structured-light image by triangulation of projected dots and a reflectivity map from the same dots (performing additional analysis of the dot images in the captured structured-light image).
• Anticipation under § 102: Closest single reference to the claimed two-channel (distance + material-related) concept, because it derives both depth and reflectivity (a material-related optical quantity) from one structured-light reflection image. It still does not anticipate claim 1: (i) its depth channel is triangulation/structured-light, not a depth-from-photon-ratio or depth-from-defocus filter (nor a filter correlating at |ρ| ≥ 0.40 with them as claimed); and (ii) its reflectivity output is not the claimed material-dependent image filter set nor "material property m" identification. It is the most damaging § 103 combination reference and could anticipate narrower dependent claims that add a structured-light distance filter to a reflectivity/material evaluation — but only if the granted dependent claim language tracks Intel's disclosure (verify against the granted B2 dependent claims).

6. WO 2019/042956 A1 (WO2019042956A1)

• Full citation: WO 2019/042956 A1, trinamiX GmbH, "Detector for determining a position of at least one object"
• Priority date: 2017-08-28 | Publication date: 2019-03-07
• Description: trinamiX detector for position determination of an object using optical sensor elements (photon-ratio/depth-measuring technology; also referenced as prior art within the trinamiX detector lineage).
• Anticipation under § 102: Not claim 1 in full — distance/position detection without the material-property identification and material-dependent-filter limitations. Same common-ownership § 102(b)(2)(C) caveat as reference 2.

7. CN 108 363 482 A (CN108363482A)

• Full citation: CN 108363482 A, 江苏四点灵机器人有限公司 (Jiangsu Sidianling Robot Co., Ltd.), "A method of the three-dimension gesture based on binocular structure light controls smart television"
• Priority/filing date: 2018-01-11 | Publication date: 2018-08-03
• Description: Binocular structured-light method for smart-TV control: synchronously captures left/right images, reconstructs a 3D image, pre-processes, segments and recognizes 3D gestures, and converts recognized gestures into operation instructions for the television.
• Anticipation under § 102: Not claim 1 — no material-property identification, no depth-from-photon-ratio/defocus filtering, no material-feature filter. It is a relevant § 103 reference only for the human-machine interface / entertainment device / gesture-control claims (identifying an object such as a hand/user for information exchange), not for the material-identification detector claim.

Additional references disclosed in the specification's background (likely on the US B2 citation list, not in the EP list of 7)

The granted US specification (text supplied in the brief) expressly names two Philips skin-detection references that are directly relevant to the claimed purpose (material classification, especially skin vs. non-skin) and should be checked on the USPTO front page:

8. US 2016/0206216 A1 (Philips; skin detection via thermal + light sensor data)

• Device, system and method for skin detection using a thermal sensor input and a light sensor input, with an evaluation unit analyzing both to detect skin areas.
• § 102: Does not anticipate claim 1 — different sensor modality combination (thermal + visible light), no reflection-beam-profile dual distance/material filter structure.

9. US 2016/155006 A1 (as written in the specification; i.e., the Philips skin-detection family published as US20160155006A1)

• Device/method for skin detection that projects a predetermined illumination pattern onto a scene, acquires an image, and evaluates the imaged illumination pattern as reproduced in the image to detect skin and distinguish skin from non-skin.
• § 102: This is the conceptually closest prior art to the claimed material-classification-by-reflection-pattern idea (classifying material — skin — by analyzing how a projected illumination pattern appears in a recorded reflection image). It still does not anticipate claim 1 because it does not disclose the claimed two-filter architecture: a distance feature from depth-from-photon-ratio or depth-from-defocus (or ≥ 0.40-correlated filter), a material feature from the enumerated material-dependent filters, and joint determination of longitudinal coordinate z and material property m. It is the strongest § 103 obviousness anchor against the method/system claims and against any claim not requiring the specific filter combination.

Bottom line

• No single cited reference appears to anticipate claim 1 of US12332352B2 in full, because none discloses the required combination of (a) a depth-from-photon-ratio and/or depth-from-defocus distance filter (or a ≥ 0.40-correlated surrogate) and (b) a material-dependent image filter, jointly evaluated to output both a longitudinal coordinate z and a material property m. This is consistent with the claims having been granted.
• Strongest individual references: US 2018/0033146 A1 (depth + reflectivity from one structured-light image — closest in concept to simultaneous distance/material evaluation) and US 2016/155006 A1 (material/skin classification from a projected illumination pattern's reproduction in an image — closest in purpose). DE 198 46 619 A1 is essentially irrelevant to the claims.
• The BASF/trinamiX references (WO 2014/097181 A1, US 2017/0363465 A1, US 2018/0276843 A1, WO 2019/042956 A1) supply the distance-measurement and detector-hardware elements and are the primary § 103 building blocks; CN 108 363 482 A is relevant only to the gesture/HMI claims.
• Unverified items to close out: (i) the official USPTO "References Cited" list and granted claim set for US12332352B2 (Patent Center, application US18/743,352); (ii) whether the B2 dependent claims recite any structured-light/triangulation or skin-classification limitation that would change the § 102 exposure of references 5 and 9. I could not confirm either within this session's search budget and recommend a Patent Center pull before relying on claim-level conclusions.

Preliminary note on the record available to me

The "Prior Art section of this page" you reference is not present in the message I received. The materials I have are: (a) the Google Patents full text of US12332352B2 (truncated before the claims); (b) the previously generated patent summary, litigation, and PTAB sections; and (c) live search results I just ran. I therefore reconstruct the prior-art universe from the documents cited in the specification's own background section plus the verified search results, and I flag at the outset that the granted claim set of US12332352B2 was not retrievable verbatim in this session (consistent with the earlier sections of this analysis). What the searches did surface is the claim text of the closely related, same-title granted family member US 11,947,013 B2 (publication US 2022/0157044 A1), whose specification text is essentially identical to the US12332352B2 text I have. That claim text — quoted below — is my working model of independent claim 1. Any conclusion below is conditional on that claim model and should be re-run against the actual USPTO grant PDF (application US18/743,352) before litigation use.

1. Working model of the relevant independent claim

From the family-member record (US 11,947,013 B2 / US 2022/0157044 A1, "Detector for identifying at least one material property"), the likely independent claim reads:

A detector for identifying at least one material property m, comprising:

• at least one sensor element comprising a matrix of optical sensors, each having a light-sensitive area, wherein the sensor element is configured for recording at least one reflection image of a light beam originating from at least one object;
• at least one evaluation device configured for determining the material property by evaluation of at least one beam profile of the reflection image,
• wherein the evaluation device is configured for determining at least one distance feature φ1z by applying at least one distance dependent image filter Φ1 to the reflection image, wherein the distance dependent image filter is at least one filter selected from: a depth-from-photon-ratio filter; a depth-from-defocus filter; a linear combination thereof; or a further distance dependent filter correlating to those by |ρ| ≥ 0.40;
• wherein the evaluation device is configured for determining at least one material feature φ2m by applying at least one material dependent image filter Φ2 to the reflection image;
• wherein the evaluation device is configured for determining a longitudinal coordinate z and the material property m by evaluating the distance feature φ1z and the material feature φ2m.

Reduced to its inventive core, the claim asks for: (i) an imager that captures reflected light-spot/beam profiles from an object, (ii) a distance feature extracted from those profiles (via photon-ratio or defocus techniques), (iii) a material feature extracted from the same profiles (via texture/spot-shape/luminance-type filters), and (iv) joint use of both features to output both distance z and material property m — thereby de-con founding the known distance/material ambiguity of beam-profile analysis.

2. Person of ordinary skill in the art (PHOSITA)

A PHOSITA would be a team: an optical engineer and an image-processing/computer-vision engineer, with 2–4 years' experience in active 3D sensing, structured light, beam-profile/depth-from-defocus algorithms, and optical material characterization (skin detection, reflectivity estimation, surface inspection), aware of standard image-texture operators (Haralick GLCM features, Law's masks, speckle statistics, Gaussian/median smoothing, gradient norms) and of standard statistical filter-selection tools (correlation, ANOVA/F-tests). This profile matters because most of the individual "material-dependent filters" in the specification are textbook image operators that such a person would know.

3. Verified prior-art references and what each teaches

Two further documents cited in the specification (WO 2014/097181 A1; WO 2014/198629 A1) disclose transversal sensors and pixel-array optical detectors of the type used as the claimed "sensor element."

4. Obviousness combinations and motivation analysis

Combination A (strongest): US 9,947,099 B2 (US 2018/033146 A1) + trinamiX beam-profile distance art (WO 2014/198629 A1 / WO 2015/024871 A1 / WO 2016/120392 A1, all published 2014–2016)

Element mapping. Apple teaches: a matrix-of-pixels camera capturing a reflection image of projected dots (element 1); per-dot intensity-profile analysis to generate a reflectivity (material-related) map — a "material feature φ2m" from the beam profile (element 2's material side); and recognition that the dot profiles must be corrected for distance. What Apple does not teach is deriving the distance from the beam profile itself; it uses triangulation. The trinamiX publications teach exactly that missing piece: deriving a distance feature from the same reflected-spot profile via a photon-ratio combined signal Q or defocus analysis. The claimed step of determining "a longitudinal coordinate z and the material property m by evaluating the distance feature and the material feature" is the direct, foreseeable union of Apple's distance-corrected reflectivity-from-dots with trinamiX's profile-derived distance.

Motivation. Apple's specification itself identifies the exact problem the patentee states in its background — intensity profiles of projected dots vary with both distance and surface/material — and solves it with a distance-correction curve applied to intensity profiles. A PHOSITA wanting to improve material classification from reflection beam profiles (the patent's stated goal: "identifying a material from its reflection beam profile would be highly desirable") would have a concrete, documentable motivation to replace Apple's triangulation-supplied distance (which requires a calibrated projector-camera baseline) with a profile-derived distance (trinamiX's Q-signal), because that permits monocular distance+material extraction from the same image data, at the same wavelengths, with no extra hardware. The combination uses the same input (dot reflection images), produces no new physical structure (both are algorithmic filters over an image), and each piece performs its known function — the textbook KSR "known problem + known solution + predictable result" pattern. The Apple disclosure of "profile correction ... as a function of distance" (Fig. 3 of US 9,947,099 B2) is itself an implicit teaching that material-feature extraction requires distance normalization — i.e., a skilled artisan reading Apple is directed toward jointly estimating distance and material from dot profiles.

Combination B: US 2016/0155006 A1 (trinamiX skin detection) + trinamiX DPR/DFD art

Element mapping. The '5006 application is nearly a preview of the patent's material side: it projects an IR dot pattern, images it, and classifies skin vs. non-skin by measuring blur of dot edges and dot intensity — i.e., it applies material-discriminating features to the beam profiles of reflection features. It even teaches comparing imaged vs. projected pattern dimensions to estimate blur. What it lacks is a formal separation of the distance contribution: dot width and intensity are precisely the features that vary with distance for translucent materials (the patent's own motivating example). The trinamiX DPR/DFD publications teach extracting z from the same dots. The claimed invention's "distance feature + material feature → z and m" is thus an obvious refinement of trinamiX's own earlier skin detector: normalize the blur/intensity-based skin features using the simultaneously measured profile-based distance, exactly as the specification describes (determining m "subsequently after determining of the longitudinal coordinate z such that the information about z can be considered for evaluating φ2m").

Motivation. Same assignee, same application domain (live-skin detection for face unlock — see the later trinamiX publication US 2024/0037201 A1 surfaced in search, confirming the commercial program), same hardware class (dot projector + NIR camera). The '5006 document's known deficiency — distance sensitivity of blur/intensity thresholds ("lower and upper bounds suited to segment illumination pattern ... can be selected based on the distance," '5006 ¶[0058]) — provides an express reason to add a distance-measuring channel; the DPR/DFD art supplies that channel from the very same captured pattern. Combining two same-field references to remove a known confound is the paradigm of obvious improvement, not invention.

Combination C: US 9,947,099 B2 + DE 198 46 619 A1 + DPR/DFD art

DE 198 46 619 A1 independently shows that one can illuminate a surface, capture reflected light on a photosensor array, and derive both a structure/surface-quality code (a material-dependent output) and a distance characteristic value from the measured signals — the conceptual "joint z and material" determination predates the patent. A PHOSITA combining BYK's structure-code extraction (which the patent itself cites in its background as beam-profile analysis, "BPA") with Apple's per-dot reflectivity estimation and trinamiX's profile-based distance would arrive at the claimed two-filter architecture. Motivation: BYK establishes the desirability of a single compact reflective-mode surface-characterization instrument; Apple and trinamiX supply the algorithmic machinery to make it work on arbitrary dot patterns rather than BYK's fixed gonio geometry.

Combination D (weaker, corroborating only): CN 108 363 482 A + any of the above

The CN reference (binocular structured-light 3D gesture control) corroborates that structured-light dot imagers feeding classifiers (including skin/gesture discrimination) were conventional in the consumer-device context; it adds little structurally beyond Combination A/B but would be cited to defeat any argument that the claimed hardware context (projector + camera + evaluation unit) is non-obvious in a human-machine interface.

5. Why a PHOSITA would have been motivated (general rationales)

1. Same field and overlapping hardware. All primary references work in active optical sensing of reflected light from surfaces, using a camera (matrix of pixels) and optionally a dot/pattern projector. No incompatible hardware or conflicting design philosophy separates them.
2. The patent's own problem statement is the motivation. The specification admits that "most image filters used for beam profile analysis (BPA) yield features that depend on both distance and material such that reliable identification and classification of the material is not possible," citing the spot width of translucent materials as the canonical example. Apple's '099 patent already identifies the identical confound and applies a distance-dependent correction. Where the prior art itself identifies a problem and a partial solution, combining the complementary missing piece (a profile-derived distance) is obvious under KSR.
3. Predictable, purely algorithmic result. Both the distance feature (DPR/DFD) and the material features (luminance, spot-shape, gradient, standard deviation, smoothing/speckle, GLCM contrast/energy/homogeneity/dissimilarity, Law's energy, threshold-area — all conventional image operators the specification lists) are computed from the same reflection image. Running two known filters on one image and combining outputs by a predetermined function (lookup table or polynomial, as the specification permits) is routine engineering with a predictable result: separating distance and material contributions.
4. The specification's own selection metrics are conventional statistics. The |ρ| ≥ 0.40 correlation threshold and the ANOVA/F-test (p ≤ 0.075/0.05/0.025/0.01) "hypothesis testing" passages are standard filter-selection/validation methodology, not inventive algorithms; a PHOSITA selecting any of the listed texture filters for material discrimination would use precisely those tools.
5. Design incentive / market pull. Live-skin detection for face unlock (Qualcomm/trinamiX program, 2020, documented in the EE Times coverage surfaced in search) created strong incentive to fuse trinamiX's own earlier distance-from-beam-profile work with its earlier skin-from-pattern work — the Combination B path — because a spoof-resistant material check must work at arbitrary user distance.

6. Balanced assessment — where the obviousness case is weaker

• Joint determination of z and m from the two features. Apple's reflectivity map is albedo/intensity-based, not a full scattering/translucency classification, and Apple still needs triangulation for depth. If the examiner-granted claims require the specific two-filter joint evaluation as a closed loop (using φ1z to de-confound φ2m and vice versa), a defendant's expert must show the DPR/DFD art was combinable with material filters "with a reasonable expectation of success" in reflective-mode material ID at <1000 nm — the very regime the patent says was unreliable. This is the claim's strongest surviving technical point.
• Dependent-claim filters. Claims tied to the enumerated filter outputs (e.g., Law's energy convolutions E5×L5/L5×E5, GLCM homogeneity, speckle-noise variance normalization by z) may individually be supported by specific embodiments; a defendant should map each dependent limitation to a concrete disclosure in the combinations above or add a reference (e.g., standard image-texture literature: Haralick 1973; Laws 1980) for the texture operators.
• Correlation-threshold claim language (|ρ| ≥ 0.40) is functional/result-effective language; under § 103 it is vulnerable if the underlying filter families are shown, but it complicates a clean prima facie case.
• No secondary considerations on this record. I found no evidence of long-felt need, industry skepticism, or copying attributable to the claimed invention specifically; trinamiX's commercial success (Qualcomm program, smartphone design wins) is real but attaches to the whole BPA stack and could cut both ways.

7. Conclusion

On the working claim model, the most persuasive § 103 case is:

• Primary combination: US 9,947,099 B2 (US 2018/033146 A1) as the base reference (single structured-light image → depth map + distance-corrected per-dot intensity/reflectivity map) in view of trinamiX's depth-from-photon-ratio / depth-from-defocus publications (e.g., WO 2014/198629 A1, WO 2015/024871 A1, WO 2016/120392 A1), which supply profile-derived distance features; and
• Fallback combination: US 2016/0155006 A1 (trinamiX's pattern-based skin detection from dot blur/intensity) in view of the same DPR/DFD art — arguably the cleanest "same-assignee obvious improvement" narrative, directly addressing the specification's stated problem that blur-based material features are distance-coupled.

Both combinations map every element of the modeled claim 1, involve no incompatible technology, and are supported by express or implicit motivations in the references themselves (Apple's distance-correction curves; '5006's distance-dependent thresholds; DE 198 46 619 A1's combined structure-code + distance-value instrument). A PHOSITA would combine them with a reasonable expectation of success because the innovation is algorithmic — applying known distance filters and known image-texture filters to the same recorded beam profiles and merging their outputs.

Bottom line: I assess a moderate-to-high likelihood that independent claim 1 (as modeled from the family member) would be found obvious under § 103 over Combination A or Combination B, with dependent claims materially less vulnerable. This assessment is provisional: it must be re-run against the verbatim US12332352B2 claims (unavailable this session), the actual cited art in the file wrapper, and the specification's claim-construction terms before being used in any proceeding.

Key sources: Google Patents family records for US 11,947,013 B2 / US 2022/0157044 A1 (claim text quoted above); granted US 9,947,099 B2 PDF (patentimages.storage.googleapis.com); Unified Patents portal record for US 2016/0206216 A1; Google Patents/Patents-Review records for US 2016/0155006 A1; FreePatentsOnline machine translation of DE 198 46 619 A1; trinamiX family documentation of the FiP/quotient-Q DPR methods (e.g., US 11,415,661 B2 description); trinamiX press coverage of beam-profile material classification (EE Times, March 2020).