US 9190014

I'll search for this specific patent number across the sources you named.

Let me check for litigation, PTAB, and CAFC activity specifically tied to this patent number.

Let me run a few more targeted checks on litigation/docket activity and confirm no other patent number is being conflated.

US Patent 9,190,014 — Summary

Important scoping note first: My searches of a general web index surfaced several different patents that must not be conflated with the target: U.S. 9,185,268 (Dolby, subject of IPR2026‑00252 re InterDigital Madison), U.S. 8,191,157 (Intertrust), U.S. 9,137,190 (Xerox/E Ink), and U.S. 10,553,255 (Dolby). None of those is U.S. 9,190,014. The Google Patents record for US9190014B2 is the authoritative source for the data below.

Bibliographic data

Abstract (verbatim): "Methods for transmitting additional information together with image data involve replacing pixel values to take in-gamut points to out-of-gamut points. The additional information may constitute metadata describing the image data and/or useful for controlling display and/or processing of the image data; watermark information; or other information. In some embodiments the image data is restored to its original form after extraction of the additional information."

Independent claims (as granted — 16 claims total; independent claims 1 and 11)

Claim 1 — encoding method (video encoder). A method for encoding additional information in image data, where the image data has pixel values defining in‑gamut points for pixels. The method: (a) selecting one or more pixels; (b) mapping one or more pixel values of each selected pixel to corresponding mapped values. At least one mapped value defines an out‑of‑gamut point, and at least one mapped value corresponds to one or more bits of the additional information. Further required: at least one mapped value defining the out‑of‑gamut point corresponds to at least one of the pixel values; and that mapping is performed by copying a plurality of lower‑order bits of the pixel value into the mapped value. (Claim text reads literally "by video encoder" and "mapping one ore more pixel values… the method of comprising" — transcribed as written, per the no‑auto‑correction rule.)

Claim 11 — decoding method (video decoder). A method for decoding additional information encoded in image data, the data containing in‑gamut points plus one or more out‑of‑gamut points. Steps: identify pixel values of an out‑of‑gamut pixel defining an out‑of‑gamut point; identify one or more bits of additional information corresponding to those identified pixel values; output the bits. The claim then recites, in product‑by‑process form, that the additional information was encoded by selecting pixels of the original image data and mapping pixel values to mapped values, with the same limitations carried over from claim 1 (out‑of‑gamut mapped value; mapped value corresponds to bits of additional information; mapped value corresponds to a pixel value; and lower‑order‑bit copying). Note the preamble literally reads "encoded in image data by a video decoder," which is ambiguous on its face.

Dependent claims: 2–10 depend from claim 1 (pixel selection based on pixel values; matching pre‑determined pixel values; matching a non‑selected pixel such as an adjacent pixel, the first pixel of a frame, or the first pixel of a row; metadata; restoration to an in‑gamut point; processing controlled by the additional information). Claims 12–16 depend from claim 11 (metadata used by a processing apparatus to process/render/display; low‑spatial‑resolution image or luminance channel for local‑dimming/dual‑modulation backlight driving; use by a dual‑modulation imaging device; identification of a decoder protocol; use to restore an HDR image).

Plain‑language overview

The patent is a steganographic/data‑embedding scheme for color image and video data. Because a color format's legal coordinate ranges (e.g., 8‑bit U and V values from 0–255) include combinations that fall outside the reproducible color gamut ("out‑of‑gamut points"), those points would never legitimately appear in normal imagery. The inventors exploit that: an encoder deliberately rewrites selected pixels' coordinates so they land outside the gamut, and which out‑of‑gamut point (or which out‑of‑gamut region/box in color space) is chosen carries the embedded bits — typically metadata about the video (e.g., display‑management metadata, a low‑resolution luminance map for backlight modulation, or an HDR‑restoration aid).

Claim 1's specific contribution is that the embedding is not a wholesale replacement of the pixel value: the encoder preserves the low‑order bits of the original pixel value and copies them into the out‑of‑gamut mapped value, while altering higher‑order bits to push the point out of gamut. That makes the transformation reversible — a decoder can flag the out‑of‑gamut pixel, read out the payload (from lower‑order bits and/or from which region the point sits in), and then restore the original in‑gamut color. The specification's worked example is pixels on the "grey line" (chroma = 0), where the three highest‑order bits of each 8‑bit U/V coordinate are set to '1' to force out‑of‑gamut and the remaining five bits of each carry data.

Claim 11 is the mirror‑image decode claim, reciting detection of the out‑of‑gamut pixel, recovery of the bits, and output — with the encode‑side limitations imported as a product‑by‑process recitation. The disclosure also covers alternative schemes (one out‑of‑gamut point per in‑gamut point as a 0/1 flag; plural non‑overlapping out‑of‑gamut regions carrying multi‑bit values; encoding into runs/spatial blocks of uniform pixels; configurability via sideband control such as AVC/H.264 SEI messages), and the restoring/display apparatus of FIGS. 6–8.

CAFC 2026 docket / USPTO proceeding check — what I could and could not verify

• No CAFC 2026 docket involving US 9,190,014 was found. My searches returned no appellate case, opinion, or briefing referencing this patent number. I want to be explicit: this is a negative result from a general web index, not a direct query of the Federal Circuit's docket (PACER/CM‑ECF) or the CAFC's own case database, so I cannot represent it as exhaustive. If a 2026 appeal exists and is recent or sealed, it may simply not be indexed.
• No PTAB proceeding (IPR/PGR) naming US 9,190,014 was found either. A Dolby PTAB‑activity listing I retrieved (updated Aug 2026) showed IPR2026‑00448, IPR2026‑00252, IPR2024‑01354, IPR2022‑01508, IPR2021‑00275, IPR2020‑01273, IPR2020‑00665 and IPR2020‑01209 — none of which lists the '014 patent (IPR2026‑00252 targets U.S. 9,185,268, a different patent).
• Corroboration from search results: a Justia U.S. Patent Documents table for U.S. 10,553,255 lists "9190014 | November 17, 2015 | Messmer," and an Espacenet/Google family table lists "US9190014B2 | 2010‑09‑13 | 2015‑11‑17 | Dolby Laboratories Licensing Corporation" — both consistent with the record above.

Confidence and caveats

• High confidence on title, application/PCT numbers, priority, filing and issue dates, inventors, assignee, abstract, and the claim set — these come from the full authoritative patent text supplied, and are independently corroborated for the number/title/assignee/date by the Google Patents and citation listings retrieved.
• Medium confidence on "Active / expires 2032‑02‑19": Google Patents itself labels legal status an assumption and not a legal conclusion; it also notes an "adjusted expiration," implying PTA. Maintenance‑fee status and any later expiry adjustment should be verified on USPTO Patent Center rather than taken from this listing.
• Low/negative confidence on litigation: no infringement suit, ITC action, PTAB trial, or CAFC appeal tied to this specific patent was located. Treat this as "nothing found," not "nothing exists."
• Literal-text flags: several typographical/grammatical irregularities appear in the granted claims as transcribed ("by video encoder" without an article; "one ore more"; "the method of comprising"; and in claim 11 the phrase "encoded in image data by a video decoder," which may be a drafting artifact intended to read "by a video encoder" or to refer to decoding). I have not auto‑corrected these, and the claim‑scope reading may turn on how they were actually printed in the granted patent; a certified copy from USPTO Patent Center should be consulted before relying on the claim language.

I'll search for litigation involving US patent 9190014.

Let me search more targeted patent litigation databases.

Litigation Search: U.S. Patent No. 9,190,014

Result: No litigation specifically involving U.S. Patent No. 9,190,014 was found.

I ran targeted searches across patent-litigation sources (Unified Patents portal, PTAB/USPTO PTACTS document repositories, Docket Alarm, RPX Insight, and general litigation reporting) using the exact identifier "9190014"/"9,190,014." Every search returned either unrelated patents, unrelated cases, or the patent's own prosecution/publication record. No case in which U.S. 9,190,014 is asserted (or is the subject of a declaratory-judgment action, IPR, or other PTAB challenge) surfaced.

Important caveat on completeness

I want to be explicit rather than overstate certainty:

• I could not perform a direct, definitive query of a dedicated litigation database (e.g., Unified Patents caselist at portal.unifiedpatents.com/litigation, Docket Navigator, or PACER) for this specific number within this session. My finding is based on web-search coverage of those ecosystems, which may not be exhaustive.
• A negative result from these searches is strong but not conclusive. District court complaints (and especially settled/dismissed cases) are not always well-indexed. A definitive answer would require a paid database query or a PACER/Unified Patents docket search keyed to the patent number.

Contextual notes (not litigation of the '014 patent)

• Patent owner: Dolby Laboratories Licensing Corporation (per the Google Patents record). The '014 patent is a member of a family that includes EP2617184B1, JP5607831B2, KR101471830B1, and CN103098453B (all national/regional counterparts). It does not appear in the list of patents Dolby has publicly asserted in its active campaigns.
• Dolby's documented patent campaigns involve different patents and are worth distinguishing here so they aren't mistaken for '014 litigation:
  • Intertrust Technologies v. Dolby and Dolby v. Intertrust (e.g., N.D. Cal. 3:19-cv-03371 and related EDTex actions) — digital-cinema/DCI patents, not '014.
  • InterDigital v. Disney (C.D. Cal.) involving the '268 patent (color correction/color-gamut mapping), with a collateral Dolby v. InterDigital DJ action (C.D. Cal. 2:26-cv-02269, filed March 3, 2026). The '268 patent is a different patent with a superficially similar color-gamut subject matter.
  • Roku v. Dolby / Access Advance HEVC/SEP dispute (D. Mass.) — H.265 standard-essential patents.
  • The Brazilian Dolby v. Roku action — asserts BR 112014010839-0 and BR 112014010842-0 (HEVC).
  • The Unified Patents-sourced "Daignean Technologies" and "Ultravision" cases involve entirely unrelated patents.

None of the above asserts U.S. 9,190,014.

Bottom line

Based on the sources available to me, there is no known litigation involving U.S. Patent No. 9,190,014 as of April 26, 2026. Accordingly, I cannot provide plaintiff(s), defendant(s), jurisdiction, case number, filing date, or outcome for a case that does not appear to exist.

If you need a higher-confidence negative (or the "clean bill of health" you'd rely on for diligence), I recommend confirming via a direct Unified Patents portal search and a PACER Docket Navigator lookup keyed to the exact number, since those tools index complaints and PTAB filings in a way open web search does not fully replicate.

I'll start by verifying the structured block's "no proceedings" signal against web sources, since the task instructs me to flag anything the ODP ingest may have missed.

No PTAB hits yet. Let me run a few more targeted queries to be sure I'm not missing something the ODP ingest may lag on.

Let me do one last check against PTAB docket aggregators and Unified Patents before concluding.

Proceedings overview

Total AIA trial proceedings on US 9,190,014: 0. The structured "PTAB proceedings on file" block — the canonical list from the USPTO Open Data Portal — returns no IPR, PGR, or CBM proceeding for this patent, and independent web verification (Google Patents /en, free patents online, RPX Insight, and targeted IPR/IPR2019–IPR2026 queries) surfaced none either. Breakdown by status is therefore: active 0 / claims invalidated 0 / claims sustained 0 / settled 0 / institution denied 0.

Bottom-line defensive posture: There is no PTAB record to work from, which cuts both ways. On the positive side, a defendant gets no free roadmap — no institution decision, no FWD, no petitioner's expert declaration, and no estoppel-triggering record to inherit. On the negative side, there is no adjudicated invalidity of any claim: all 16 claims remain presumptively valid and untested at the Board, so an IPR-based defense must be built from scratch. Do not describe this patent as "hardened by IPR" — it has never been challenged. The correct framing is "un-challenged and un-narrowed, but also un-vindicated."

Proceedings

No proceeding entries. There are no AIA trial proceedings to list for US 9,190,014 — no proceeding numbers exist to report, and none will be invented here.

Strategic summary

Claim status — all claims untested. US 9,190,014 issued 2015-11-17 with 16 claims, and no claim has ever been canceled, confirmed, or even instituted for review. Independent claim 1 (encoder method) and independent claim 11 (decoder method) stand as issued, each carrying the same limiting tail: "wherein at least one of the one or more mapped values defining the out-of-gamut point corresponds to at least one of the one or more pixel values; wherein mapping the at least one pixel value ... comprises copying a plurality of lower order bits of the at least one pixel value to the at least one mapped value." Dependent claims 2–10 hang off claim 1 (selection criteria, metadata, restoration, controlled processing); claims 12–16 hang off claim 11 and are the commercially loaded ones — metadata used by a processing/rendering/display apparatus (12), a low-spatial-resolution version or luminance channel for driving a backlight in a local-dimming or dual-modulation display (13), use by a dual-modulation imaging device (14), an identification of a decoder protocol (15), and restoration of an HDR image (16). Because nothing has been canceled or narrowed, there is no surviving-claim list to exploit; the full issued scope is live. Assignee is Dolby Laboratories Licensing Corp, inventors Neil W. Messmer and Lewis Johnson, adjusted expiration 2032-02-19.

Estoppel landscape. Section 315(e)(2) estoppel is not a factor here, because no IPR or PGR was ever instituted against this patent — there is no petitioner, no privity chain, and no ground that anyone is barred from raising. That is genuinely good news for a defendant: the entire universe of prior art is available in a district court invalidity case, unrestricted by any Board record. It also means there is no Fintiv/Sotera discretionary-denial history, no § 325(d) "same or substantially the same art" estoppel against the Office, and no patent-owner amendment record to squeeze. The practical corollary is that a defendant must self-source art, and the portfolio-level art cited during prosecution is the obvious starting point — the '014 file history cites a deep Digimarc "out-of-phase color watermark" line (e.g., US 6,614,914; US 6,590,996; US 6,763,124; US 7,561,714 "Reversible watermarking") plus US 6,611,273 (Microsoft, encapsulating signed over-ranged color image data), Xerox US 6,439,465, and the non-patent literature Cheddad et al., "Digital Image Steganography: Survey and Analysis of Current Methods" (Signal Processing, 2010) and Specht, "Using Perceptually Uniform Color Spaces for Image Steganography" (2006). Note the teaching-away/anticipation tension embedded in claim 1's "copying lower order bits" requirement: a watermark reference that hides data by replacing chroma values may not read on that limitation, so a petitioner's art selection must be deliberate.

Pattern signals. None internal to this patent — no repeat petitioner, no serial IPR campaign, no defensive aggregator (Unified Patents or similar) on the chain. It is worth situating the '014 in Dolby's broader out-of-gamut/gamut-metadata family, because activity near this patent can be confused with activity on it. Two examples, both expressly not proceedings on US 9,190,014: (i) Dolby itself has appeared as a petitioner in IPRs against third-party gamut/color patents (e.g., the IPR against US 9,185,268, where Dolby challenged claims 1–30 on § 103 grounds over Jones, Ollivier, Osaki, Morović, Yeung, and Vetro); and (ii) InterDigital v. Disney in the C.D. Cal. involves Dolby Vision and the '268 color-correction patent, not the '014, with Dolby having filed a declaratory-judgment counter-action. Neither docket touches US 9,190,014. Also of note: foreign counterparts exist and can matter for parallel validity work — EP 2617184 B1, JP 5607831 B2, KR 101471830 B1, CN 103098453 B, and WO 2012/036947 A2. The '014 has attracted forward citations in the color-remapping/displays space, including Qualcomm US 10,368,099 (color remapping information SEI signaling for display adaptation) and various Samsung display/gamut patents — a hint that the claim space is crowded, which is useful for obviousness theories.

Recommended next steps

1. Treat "no PTAB activity" as confirmed, not as an excuse to skip a validity screen. I ran four independent query sets (patent number bare, number with "IPR2019–IPR2026," Dolby-assignee searches, and PTAB/docketalarm aggregator searches) and found zero proceedings. If your investigation surfaces a recently filed petition the ODP ingest has not yet indexed, flag it immediately — but on the current record there is nothing to distinguish. There is no FWD or institution decision to link, because none exists.
2. Build the invalidity case from the prosecution history art, not from a Board record. Start with the Digimarc out-of-phase watermark family and Cheddad 2010; map each against claim 1's requirement that the mapping "copy a plurality of lower order bits" to the out-of-gamut value — a genuine narrowing hook that the specification supports (the 8-bit chroma example sets the three high-order bits to '1' and writes 5 bits of data into the low-order bits). That limitation is the most promising § 103 battleground.
3. Calibrate expectations against the current institution climate. With Director Squires having asserted personal control over § 314/§ 324 institution decisions and 2025 practice (Fintiv plus the "settled expectations" factor) sharply reducing institution rates generally, a fresh petition on a patent in force since 2015 (over six years) faces elevated discretionary-denial risk regardless of merits. Weigh an IPR's estoppel and cost profile against preserving all invalidity arguments in district court.
4. If you are the defendant and the patent owner is asserting claims 12–16, focus there first. Those dependents target the commercially meaningful HDR / dual-modulation / local-dimming and decoder-protocol use cases; they are the claims most likely to be teed up, and they carry the same "copying lower order bits" backbone through claim 11, so an invalidity or non-infringement position on claim 11 propagates upward.
5. No active proceedings means no PTAB milestones to monitor — no institution deadline, no oral hearing, no statutory one-year FWD date. The relevant dates on this patent are the adjusted expiration (2032-02-19) and any district-court or foreign validity schedule you choose to run.

Caveat on sourcing: no PTAB proceeding numbers, panels, institution decisions, FWDs, settlements, or Federal Circuit appeals are reported above because none were found for US 9,190,014 in the ODP-provided structured data or in the searches run on 2026-09-28. Nothing in this report is fabricated; where I could not verify something, I have said so.

I'll reconstruct the assignment record for US 9190014 using the authoritative patent text provided plus verification searches.

Let me check uspto.report and assignment-specific sources for the recorded reel/frame data.

Assignment Record Reconstruction — US 9,190,014 B2

Patent: Data transmission using out-of-gamut color coordinates
Application: US 13/819,995 (national stage of PCT/US2011/050647) · Priority 2010-09-13 (US provisional 61/382,449) · Filed 2011-09-07 · Granted 2015-11-17
Source of record used here: Google Patents legal-events table for US9190014B2 (authoritative text supplied, fetched 2026-09-28) plus corroborating search.

Data caveat up front: I was able to confirm that exactly one assignment event exists in this chain, but the Assignment Center's reel/frame numbers and the correspondent of record were not retrievable in this session (the USPTO Assignment Center and uspto.report pages did not return the numeric reel/frame data). Per the operating rules I will not invent reel/frame numbers. Where a field is unknown I say so explicitly.

Inventors

• Both inventors are named on the issued patent and on the PCT/WO publication (WO2012036947A2).
• Corroboration that both are long-standing Dolby researchers, not a departing group: the same pair, together with Peter W. Longhurst, appears as co-inventors on the sibling Dolby filing Content Metadata Enhancement of High Dynamic Range Images (WO2012071235 / EP2643809, priority 2010-11-23) — see the INPI/EP register entries listing "LONGHURST, Peter W.; JOHNSON, Lewis; MESSMER, Neil W." And Messmer/Johnson appear again on Video Delivery and Control by Overwriting Video Data (US 2012/0315011 A1), assigned to Dolby Laboratories Licensing Corp.
• Unusual pattern: none detected. There is no evidence of inventors leaving the assignee within 12 months of filing; both continued filing Dolby applications for years afterward. No pre-fire-sale exodus signal.

Original assignee

Dolby Laboratories Licensing Corporation — the IP-holding/licensing subsidiary of Dolby Laboratories, Inc. (NYSE: DLB), San Francisco, CA (1275 Market Street, later confirmed in the register records above).

• Line of business: audio and imaging technology licensing and products. Dolby is an operating company, not a shell. It ships and licenses real product technology (Dolby Vision HDR, Dolby Atmos, cinema/consumer display pipelines).
• Product embodying the claims: the patent is directed to embedding metadata in image data via out-of-gamut chroma coordinates — squarely within Dolby's HDR / display-management / video-delivery technology stack (the additional-information examples in the spec — "low-spatial resolution version of the image data… for driving a backlight in a local-dimming or dual-modulation type of display," HDR reconstruction — are Dolby's own commercial subject matter).
• Current status: operating. Dolby Laboratories Licensing Corporation remains the current assignee per the Google Patents legal-events table. No acquisition, dissolution, or bankruptcy record found.

Assignment timeline

Only one assignment event is recorded against this patent. No post-issuance transfers exist.

• 2013-03-13 (recorded; execution date not surfaced by the available sources) — Reel not retrievable in this session
  • Conveyance: Assignment (Assignment of Assignors' Interest)
  • Assignor: Lewis Johnson and Neil Messmer (the named inventors)
  • Assignee: Dolby Laboratories Licensing Corporation
  • Correspondent: not retrievable from the sources consulted. No repeat NPE-aligned correspondent identified anywhere in this chain — irrelevant here because there is only one link and it runs to an operating-company assignee.
  • Context: original inventor-to-company assignment; standard "assignment of inventors' interest" taken as part of national-stage filing — not a fire-sale, reorg, securitization, or transfer-to-asserter.

No other recorded assignments. The chain terminates at Dolby Laboratories Licensing Corporation. There is no second link, no LLC, no holding entity.

The other dated entries in the legal-events table are not assignments: 2011-09-07 (application filed), 2013-06-27 (publication of US 2013/0162666 A1), 2015-11-17 (grant/publication), and 2032-02-19 (adjusted expiration — a term/status date, not a transfer).

Timeline diagram

timeline
    title Ownership of US 9190014
    2010 : Priority provisional filed
    2011 : US and PCT application filed
    2013 : Inventors assign to Dolby Licensing
         : Application published
    2015 : Patent granted
    2032 : Adjusted expiration

NPE / troll-pattern signals

Verdict

Insufficient data — more precisely, no NPE activity; ownership remains with the original operating-company assignee.

This is the "only the original assignment" case the rubric anticipates. The entire chain consists of a single assignment of the inventors' interest to Dolby Laboratories Licensing Corporation recorded 2013-03-13, and nothing since. Zero of the eight NPE signals is present: no shell LLC, no known asserter, no cascading LLC transfers, no pre-litigation transfer, no bankruptcy sale, no defensive-aggregator terminus. Both inventors (Messmer and Johnson) remained prolific Dolby co-inventors for years after filing, which weighs against the departing-inventor / fire-sale pattern. The additional-information uses described in the specification (local-dimming backlight metadata, HDR reconstruction) map directly onto Dolby's own commercial product lines.

Confidence: high on "no NPE signals"; the residual uncertainty is that I could not pull the numeric reel/frame or the correspondent of record, so I flag those as unverified rather than guessed.

Verification link (USPTO Assignment Center search by patent number): https://assignmentcenter.uspto.gov/ — query 9190014 (mirror: https://assignment.uspto.gov/patent/index.html). Google Patents legal events: https://patents.google.com/patent/US9190014/en.

I'll research this patent and its cited prior art references using live search.

I have substantial data. Let me research a few of the key cited references in more detail to properly characterize them.

Prior‑Art Analysis — US 9,190,014 B2

1. Reference point (as established in the prior sections)

Because the application was filed 2011‑09‑07 (before 16 March 2013), it is a pre‑AIA case. The effective priority date for art purposes is 2010‑09‑13; references published before 2009‑09‑13 are § 102(b) art, and references filed/published in the window up to the priority date are § 102(a)/102(e) art.

2. The anticipation test applied here

Under 35 U.S.C. § 102 a reference anticipates only if it discloses every limitation of a claim, arranged as in the claim (Net MoneyIN v. VeriSign; MPEP 2131). The claim set has only two independent claims:

• Claim 1 (encoding, "by video encoder"): (a) image data with in‑gamut points; (b) selecting pixels; (c) mapping selected pixel values to mapped values; (d) at least one mapped value defines an out‑of‑gamut point; (e) at least one mapped value corresponds to bits of the additional information; (f) the mapped value defining the out‑of‑gamut point corresponds to at least one of the pixel values; (g) the mapping copies a plurality of lower‑order bits of the pixel value into the mapped value.
• Claim 11 (decoding, "by video decoder"): identifying a pixel value defining an out‑of‑gamut point; identifying bits of additional information corresponding to it; outputting the bits; plus the negative limitations reciting how the information "was encoded" (selecting/mapping/out‑of‑gamut/copying lower‑order bits).
• Dependent claims 2–10, 12–16 add: value‑based pixel selection (2–7), metadata (8), restoration mapping (9–10), metadata/backlight/HDR/dual‑modulation/decoder‑protocol payload types (12–16).

The dispositive limitation is the out‑of‑gamut mapping (claim 1(d)/(f)/(g), claim 11). Almost every reference cited on the face of US 9,190,014 embeds data within the gamut (watermark modulation, LSB substitution, glyph printing). Such references are properly § 103 material, not § 102 anticipation. I flag each accordingly.

3. Reference‑by‑reference analysis

A. The two references the patentee itself treated as closest prior art

A1. WO 2008/096281 A1 — NDS Limited, "System for embedding data"

• Filing 2008‑01‑13 (PCT/IB2008/050104); priority 2007‑02‑05. Family: CN 101601068 A (also cited). US equivalent US 8,135,168 B2 / US 8,755,598 B2.
• Description: Embeds marking information in video frames by applying a positional dot‑product transform R′(p)=R(p)+<p,ω_R>, etc., and detects it via a "color mass center" difference. Nothing is mapped outside the gamut — the EP counterpart (EP 2 617 184 B1, background ¶[0004]) expressly states: "Both the original and the modified video frames comprise in‑gamut color elements only."
• § 102: No anticipation. It discloses pixel selection and mapping (claim 1(b)–(c)) but not limitations (d), (f) or (g) (no out‑of‑gamut point, no lower‑order‑bit copy). Closest § 103 reference for claim 1 and for claim 11 only if combined with an out‑of‑gamut‑carrier teaching.
• Source: https://patents.google.com/patent/WO2008096281A1; EP2115694B1.

A2. Specht, Michael, "Using Perceptually Uniform Color Spaces for Image Steganography," Otto‑von‑Guericke‑Universität Magdeburg, 15 March 2006 (XP055094013)

• Description: Replaces LSBs of each color channel with payload bits and subdivides the color space into "color classes" in a perceptually uniform (CIELab) space so that perceptually similar colors can be exchanged to carry the payload. Expressly an improvement on LSB substitution; it operates on in‑gamut colors (the paper notes some CIELab classes are empty "due to the shape of the color gamut" — i.e., gamut is treated as a constraint, not a carrier).
• § 102: No anticipation of claims 1 or 11 — LSB substitution is in‑gamut and does not use out‑of‑gamut points as the signalling mechanism. It does disclose "copying lower order bits" (claim 1(g)) and bit‑to‑payload correspondence (claim 1(e)), making it prime § 103 art against the lower‑order‑bit aspects of claims 1/11 and against EEE5/EEE19.
• Source: https://www.inf.ovgu.de/inf_media/downloads/forschung/technical_reports_und_preprints/2006/Using%20perceptually%20uniform%20color%20spaces%20for%20image%20steganography.pdf

A3. Cheddad, A. et al., "Digital Image Steganography: Survey and Analysis of Current Methods," Signal Processing, Elsevier, vol. 90, no. 3, 1 March 2010, pp. 727–752

• Description: A general survey of spatial‑ and transform‑domain steganography (LSB, masking/filtering, transform techniques). It is a taxonomy of in‑gamut data‑hiding; it does not teach signalling via out‑of‑gamut coordinates.
• § 102: No anticipation — a survey lacking the out‑of‑gamut limitation and any single‑reference disclosure of all claim elements. Useful only as § 103 background evidence of the general state of data‑hiding.

B. The Digimarc watermark family (≈20 of the 43 citations)

These share a common scheme: a watermark signal is embedded in color channels, often with inverted signal polarity between chrominance channels ("out‑of‑phase"), and detection exploits the polarity difference. The modifications stay within the reproduction gamut. Representative members:

Net: the entire Digimarc block fails limitations (d), (f) and (g) of claim 1 and the out‑of‑gamut limitation of claim 11. They are § 103 references at most, and several are individually weak because they are print/fluorescence‑oriented.

C. Gamut / color‑representation and video‑encoding references

D. Document/data‑hiding (glyph and tracking) references

E. Co‑pending Dolby applications cited (contextual, not anticipatory)

4. Bottom line for § 102

No cited reference anticipates claims 1–16. The out‑of‑gamut mapping of claim 1(d)/(f) and of claim 11, together with the lower‑order‑bit‑copy step of claim 1(g), is absent from every reference on the face of US 9,190,014:

• The Digimarc block, Specht, Cheddad, Xerox/US 6,439,465, Sharp and the NDS/WO 2008/096281 family all embed data inside the gamut (watermark modulation, LSB substitution, glyphs, or colour‑mass shifting) — the NDS reference is expressly distinguished in the EP counterpart as using in‑gamut color elements only.
• The gamut‑related references (Maurer US 2004/0170319, Demos US 2009/0201309, Microsoft US 6,611,273) touch out‑of‑range/out‑of‑gamut color values but use them to represent image content, not to carry metadata bits, and none restores them by the claimed lower‑order‑bit mechanism.

Most relevant art, ranked:

1. WO 2008/096281 A1 (NDS) / CN 101601068 A — closest § 103 reference on video‑frame data embedding generally; fails the out‑of‑gamut limitations.
2. Specht (2006) — closest on lower‑order‑bit substitution and bit‑to‑payload mapping; fails the out‑of‑gamut limitations.
3. US 2009/0201309 A1 (Demos) and US 6,611,273 B2 (Microsoft) — supply the "out‑of‑range color values" concept needed to bridge the gap under § 103.
4. US 6,590,996 B1 / US 7,391,880 B2 (Digimarc colour‑adaptive) — § 103 against claim 3 (value‑based selection).
5. US 7,561,714 B2 (Digimarc reversible watermarking) and US 8,135,168 B2 (NDS) — § 103 against claim 9 (restoration).

Caveat on confidence: the characterizations above for the bulk of the 43 citations are based on the citations' titles, publication data and the cited fragments retrievable through Google Patents/EPO/FPO (URLs given in § 3); I did not retrieve the full specification of every one of the 43 references, so for any individual reference the § 102 conclusion should be re‑verified against its complete disclosure before being relied on formally. The two references the patentee addressed as background (WO 2008/096281 A1 and the Specht article) were verified directly from the EP 2 617 184 B1 text.

I'll ground this analysis by verifying the key prior-art references listed on the patent page before building the §103 combinations.

Let me verify two more reference families that will anchor the dependent-claim analysis.

Obviousness Analysis — U.S. Patent No. 9,190,014 B2

"Data transmission using out-of-gamut color coordinates"
Assignee: Dolby Laboratories Licensing Corp. · Inventors: Messmer, Johnson
Priority: 2010‑09‑13 · Filed: 2011‑09‑07 · Granted: 2015‑11‑17 · App. 13/819,995
Source: https://patents.google.com/patent/[US9190014](/patent/US9190014)/en

Scope note / caveats. This is an independent analyst reconstruction of §103 grounds from the prior art listed on the patent's own face, not the examiner's record. I have not reviewed the actual Office Actions or any PTAB proceeding, so I cannot state what the examiner actually relied on. Because the application was filed 2011‑09‑07 (priority 2010‑09‑13), the pre‑AIA version of §§102/103 governs, including the pre‑AIA §103(c) common‑ownership exception — which matters a great deal here (see §5). Characterizations of references I verified by search are grounded with URLs; characterizations of references known to me only by their titles as listed on the patent page are flagged as such.

1. The claims to be evaluated

Only two claims are independent. Both recite a video codec context and — critically — the same narrow bit‑manipulation limitation.

Claim 1 (encoder) requires: selecting pixels of image data whose values define in‑gamut points; mapping pixel values to corresponding mapped values; where (i) at least one mapped value defines an out‑of‑gamut point, (ii) at least one mapped value corresponds to bits of the additional information, (iii) at least one mapped value defining the out‑of‑gamut point corresponds to at least one of the original pixel values, and (iv) "mapping … comprises copying a plurality of lower order bits of the … pixel value to the … mapped value."

Claim 11 (decoder) mirrors this, and additionally imports the encoder-side limitations verbatim (selecting pixels, mapping, out‑of‑gamut point, copied lower‑order bits) into the decode claim as a recitation of how the data "was encoded."

Dependents add: selection by pixel value (2), by match to predetermined values (3), by match to a non‑selected pixel that is adjacent / first‑in‑frame / first‑in‑row (4‑7); metadata (8); restoration mapping (9) and processing controlled by the data (10); and, in 12‑16, metadata for a processing/rendering apparatus, a low‑spatial‑resolution luminance channel for local‑dimming / dual‑modulation backlighting, use by a dual‑modulation imaging device, an identification of a decoder protocol, and use to restore an HDR image.

The patent's own specification frames the core insight as: out‑of‑gamut points "do not exist naturally in image data," so their presence can carry information, and they "map to in‑gamut points such that the image may be properly reproduced at a display" (spec., FIG. 1 discussion). The FIG. 4 embodiment thresholds U/V to classify out‑of‑gamut regions; the "grey line" embodiment sets the three high‑order bits of 8‑bit chroma to 1 and writes payload into the five low‑order bits — which is precisely what claim 1(iv) claims.

2. The prior art on the face of the patent, mapped to the elements

3. Proposed combinations and the motivation to combine

The following are the strongest §103 grounds a challenger would plead. All follow the KSR Int'l Co. v. Teleflex Inc., 550 U.S. 398 (2007) framework (known elements, predictable results, design incentives) and Graham v. John Deere (scope/content, differences, PHOSITA level, secondary considerations).

Ground A — Specht + Digimarc out‑of‑gamut family + Microsoft '273 (attacks claims 1‑3, 8, 9, 11)

• Specht teaches the POSITA to select color‑space coordinates (L, a, b / L, U, V) as the payload‑carrying medium and to prefer chrominance for capacity/imperceptibility — the claim‑1(i)(ii) selection and chroma‑only encoding.
• Digimarc supplies the well‑settled practice of replacing/augmenting selected pixel color‑channel values to carry bits, and — per the family's own abstract — the express contemplation of an embedded signal that is out‑of‑color‑gamut in a second color scheme. A POSITA reading Specht (color‑space payload) together with the Digimarc out‑of‑phase family (payload placed so as to be out of gamut) would arrive at "put the payload in coordinate values that fall outside the target gamut."
• Microsoft '273 supplies the missing mechanics: a color‑coordinate word may be partitioned so that low‑order/in‑range bits are preserved while over‑range bits (values >1 or <0) are carried separately, and those extra bits can be stashed where legacy processors ignore them.
• Claim 1(iv) ("copying a plurality of lower order bits of the … pixel value to the … mapped value") is met by the straightforward merger of Specht's payload‑in‑color‑coordinates with Microsoft '273's in‑range/over‑range word partitioning: reserve the low‑order bits for the value/payload and force the high‑order bits to a value that pushes the point out of gamut (exactly the patent's own "grey line" example).
• Motivation to combine: (1) the recognized need to send metadata without a separate channel, stated on the face of the '014 patent's Background; (2) Microsoft '273's own stated concern — backward compatibility with legacy in‑range systems — is directly achieved by a carrier that legacy in‑range displays will clip back into gamut automatically; (3) Robotic design choice: use the portion of the coordinate word that no conforming image renders.
• Reasonable expectation of success: high — each step is individually routine and the survey literature (Cheddad) confirms the field's maturity.

Ground B — Demos '309 + Digimarc (reversible watermarking / out‑of‑phase) + Maurer '319 (attacks claims 1, 9, 10, 11, 15)

• Demos '309 establishes that color coordinates and gamut extents are separately manipulable quantities and that "extra" color information can be preserved and later applied — supporting elements (i)(iii) and the "metadata for later processing" theme.
• Maurer '319 supplies the out‑of‑gamut detection and re‑mapping logic that claim 1(iii) implies and claim 9/11's restoration literally requires.
• Digimarc US 7,561,714 (Reversible watermarking) supplies restoration‑to‑original after payload extraction (claims 9, 11).
• Motivation: the '014 Background itself says the metadata must be delivered "carried by a standard image data transmission fabric"; a carrier that is (a) invisible because out of gamut, (b) ignored by in‑range legacy pipelines, and (c) reversible per Digimarc's reversible‑watermark teaching, is the predictable, advantageous design. Claim 15 (protocol identification via sideband, e.g. SEI) is a design choice analogous to Microsoft '273's "private tag … ignored by others."

Ground C — Ground A or B + spatial/pixel‑selection art (attacks claims 4‑7)

• Claims 4‑7 (select a pixel only when its value matches a non‑selected pixel that is adjacent / first‑in‑frame / first‑in‑row) are classic predictive/run‑length pixel‑run logic. Clark US 2005/0169544 (encoding/decoding video), Mitsubishi US 2009/0080769, Sharp US 6,640,005 ("tracking data in coded images"), Xerox US 6,439,465 ("arbitrary locations within an image"), and NDS WO 2008/096281 teach embedding in coded/compressed video and choosing embedding sites by local pixel relationships. Motivation: maximizing payload opportunities and minimizing visible artifacts — the '014 patent concedes these are the very reasons (spec.: choose a frequently occurring color "thus providing the possibility for carrying more additional information in a frame").

Ground D — Ground A/B/C + Dolby display‑metadata family (attacks claims 12‑16)

• Claims 12‑16 claim the content of the payload (display metadata, low‑res luminance for local‑dimming/dual‑modulation backlight, HDR restoration, dual‑modulation device, decoder‑protocol ID). That content is squarely the subject of US 2012/0321273, US 2012/0315011, US 2013/0076763, US 2013/0120656, and US 2012/0229495. If these count as prior art, claims 12‑16 reduce to "use a known metadata‑embedding channel to carry known display‑management metadata," a predictable combination.
• Motivation: the industry‑recognized need to ship tone/gamut/backlight metadata with the essence (a need the cited Dolby applications were themselves addressing) and to reconstruct HDR at the sink.

4. Why a POSITA would have found these combinations predictable

1. The problem was known and articulated in the field. Watermarking/steganography surveys (Cheddad) and the cited Digimarc portfolio predate the priority date by a decade; the '014 Background concedes both the need and the problem.
2. Every element is a known, enumerable option. Out‑of‑gamut values and over‑range words (Microsoft '273; Demos '309); out‑of‑gamut watermarking in a color‑scheme sense (Digimarc out‑of‑phase family); gamut detection/remapping (Maurer '319); reversible embedding (Digimarc '714); chroma‑only embedding (Specht). In re Keller / MPEP 2143 fusion analysis: the combination yields no more than the expected sum of these known functions.
3. A concrete, pre‑existing design incentive. Microsoft '273's explicit goal — keep legacy in‑range systems working while carrying extra color bits — is the same backward‑compatibility incentive that makes an out‑of‑gamut carrier attractive.

5. The genuine weaknesses in the §103 case (and what the patentee will argue)

I should be explicit that this is not a clean prima‑facie case; the patent issued, and the record suggests the examiner did not find an out‑of‑gamut carrier in a single reference. The exploitable gaps:

1. No listed reference expressly uses "out‑of‑gamut" as the payload channel in the '014 sense. Microsoft '273 concerns storage of over‑range data, not covert conveyance; its partitions are explicit data structures, not concealed carriers. Digimarc's "out‑of‑color‑gamut in a second color scheme" is asserted across color schemes (e.g., a signal valid in one rendering becoming out of gamut in another), which is arguably a different proposition from deliberately mapping an in‑gamut pixel to a point outside the display's gamut within one color space. The patentee will press this distinction hard.
2. Claim 1(iv) is bit‑level specific. "Copying a plurality of lower order bits of the … pixel value to the … mapped value" requires preservation of multiple LSBs of the original value in the mapped out‑of‑gamut value. Microsoft '273 keeps low‑order in‑range bits, but as a storage partition; whether that is "copying to the mapped value" is a claim‑construction fight the patentee can win. Ground A largely rests on this being an obvious mechanical instantiation — a contestable proposition.
3. The Dolby family is almost certainly disqualified under pre‑AIA §103(c). US 2012/0315011, US 2012/0321273, US 2013/0076763, US 2013/0120656 and US 2012/0229495 share the '014 applicant/assignee (Dolby Laboratories Licensing Corp.) and — as US application publications disclosing subject matter — may qualify as prior art only under pre‑AIA §102(e). Because they were commonly owned at the time the '014 invention was made, pre‑AIA §103(c) bars their use in an obviousness combination. That removes the strongest direct evidence for claims 12‑16, and a challenger would have to substitute a third‑party display‑metadata reference (e.g., the Chellomedia "Display of Enhanced Content" application, or Sony's CN 101188749 A video‑transmission art) — a weaker fit for "local‑dimming / dual‑modulation" specifically.
4. Secondary considerations. The patentee can marshal non‑obviousness evidence: the very commercial significance of Dolby's HDR/dual‑modulation ecosystem; licensing/adoption of embedded display metadata; and the argument that the prior art's out‑of‑gamut values were treated as errors to be clipped or avoided (gamut‑mapping art is about eliminating them), i.e., a teaching away from intentionally writing out‑of‑gamut values into delivered image data. That "why would anyone deliberately publish out‑of‑gamut pixels?" argument is the strongest non‑obviousness narrative available and should be anticipated.

6. Bottom line

• Strongest viable ground: Ground A / Ground C against claims 1‑7, 9‑11 — Specht (perceptually uniform color‑space steganography) in view of the Digimarc out‑of‑phase / out‑of‑color‑gamut color‑watermark family, further in view of Microsoft US 6,611,273 (signed, over‑ranged color‑coordinate word partitioning and preservation of in‑range low‑order bits), optionally with Maurer US 2004/0170319 and Digimarc US 7,561,714 for the detect/restore steps of claims 1(iii) and 9/11, and with Xerox '465 / Sharp '005 / Clark '544 / Mitsubishi '769 / NDS WO 2008/096281 for the dependent pixel‑selection claims 4‑7.
• Claims 12‑16 are the hardest to invalidate on this record, largely because the best‑fitting references are commonly owned Dolby applications that pre‑AIA §103(c) likely disqualifies; a challenger would need a non‑Dolby display‑metadata reference.
• Overall assessment: I would rate the §103 challenge as moderate — a plausible prima‑facie case exists for the two independent claims under a KSR "known elements, predictable result" theory, but it turns on (a) claim construction of "copying a plurality of lower order bits," and (b) whether a fact‑finder accepts that the out‑of‑gamut‑as‑carrier insight was a predictable design choice rather than a non‑obvious departure from gamut‑mapping's "clipping" orthodoxy. The dependent claims' payload content limitations (12‑16) are the most defensible.

Confidence: High on the text/scope of the '014 claims and on the identities/URLs of the verified references (Microsoft '273, Digimarc '959/out‑of‑phase family, Specht, Cheddad, Demos '309). Medium‑to‑low on my characterization of references I know only by their listed titles (e.g., the Dolby family, Clark '544, Mitsubishi '769, Sharp '005, NDS WO 2008/096281) — those characterizations should be verified against the full texts before being relied on in any filing. I have not seen the prosecution history and cannot report what the examiner actually relied on.