US 7265870

I'll search for litigation and docket information on this specific patent, then compile the summary.

US Patent 7,265,870 B2 — Analyst Summary

Sourcing note: The following is grounded in the Google Patents/USPTO record for US7265870B2 (https://patents.google.com/patent/US7265870B2/en) and the Justia mirror (https://patents.justia.com/patent/7265870). I performed targeted web searches for a CAFC 2026 docket and for USPTO PTAB/IPR activity on this specific number. I found no CAFC 2026 docket, appeal, IPR, PGR, or reexamination tied to US 7,265,870. I could not access a proprietary docket database (Lex Machina/Docket Alarm/PACER) in this session, so treat "no litigation found" as a search limitation, not a definitive negative. Critically, the patent's own legal-status record shows it expired for failure to pay maintenance fees with an adjusted expiration of 2025-08-07, which makes 2026 assertion/CAFC activity unlikely.

Bibliographic Data

Discrepancy flag (not auto-corrected): The specification states the application "claims the benefit of U.S. Provisional Application No. 60/336,634 filed Apr. 12, 2001," while the USPTO/Google record lists provisional US33663401P with a priority/filing date of 2001-12-04. These two dates conflict in the source record. I am reporting both literally rather than reconciling them.

Abstract (as published)

Separation of a color into values for colorants in colorant space is performed by splitting color correction and color separation into two separate, independently controllable steps, where the color is defined as a set of values in a "trajectory space," and trajectories include points for which ink behavior during color separation is known. Colorant values for the color are determined by interpolation using the color's values in trajectory space together with the known values and corresponding colorant values of the trajectory points. Trajectories are preferably color axes dividing the trajectory space into tetrahedra. Colorant values may be corrected using techniques such as colorant replacement or colorant splitting prior to color separation, and the separation process can be calibrated using a test target.

Plain-Language Overview of the Claims

Independent Claim 1 (the only independent claim)

A method for converting a color, expressed as a set of color values in a color space (e.g., sRGB/XYZ/L*a*b*), into colorant values (C1…CN) for N colorants (e.g., CMYK). The method has three operative steps:

1. Color-correction step — convert the input color values into a set of values locating that color in an intermediate "trajectory space."
2. Define at least two trajectories in the trajectory space, populated with trajectory points (px). Each trajectory point carries both (a) its set of values in trajectory space and (b) corresponding pre-known colorant values (C1x…CNx) for the N colorants.
3. Interpolate the target color's colorant values using (i) the color's trajectory-space values and (ii) the trajectory-space values and the corresponding colorant values of at least two trajectory points.

Conceptually, this is the core insight of the patent: it breaks one hard 3-D color→colorant inversion problem into (a) a separate colorimetric correction mapping and (b) a set of easy, one-dimensional ink-behavior curves defined along chosen axes, stitched together by interpolation. This is what the patent touts as giving simple, robust control over constraints like maximum total ink (e.g., 280%) and black-ink level.

Dependent Claims 2–21 (each narrows Claim 1)

• Claim 2 — The trajectory space is three-dimensional.
• Claim 3 — Seven trajectories are used (the preferred embodiment: neutral, red, green, blue, yellow, magenta, cyan axes).
• Claim 4 — At least one trajectory is a straight line.
• Claim 5 — The straight lines are one or more of: black–white, black–yellow, black–magenta, black–cyan, black–red, black–green, black–blue.
• Claim 6 — Specifies the interpolation mechanics: divide the trajectory space into non-overlapping sub-volumes spanned by the trajectories; locate the color p in one sub-volume; find at most one trajectory point per spanning trajectory; express the colorant values as a function of those trajectory points' colorant values.
• Claim 7 — That function is a weighted mean.
• Claim 8 — The weights are the barycentric coordinates of the color in the triangle formed by three trajectory points (one per spanning trajectory).
• Claim 9 — That triangle is parallel to the tetrahedron plane opposite the trajectories' common intersection point (i.e., the "R-Y-W plane" construction of FIG. 11).
• Claim 10 — Trajectories are straight lines and the sub-volumes are tetrahedra.
• Claim 11 — Each tetrahedron contains the black and white points in trajectory space.
• Claim 12 — The N colorants are CMYK.
• Claim 13 — The N colorants are C, M, Y, K + light magenta + light cyan (six-color, inkjet graininess reduction).
• Claim 14 — The N colorants are C, M, Y, K + orange + green (Hexachrome-style extended gamut).
• Claim 15 — Adds a colorant-replacement step: at least partially replace one of the N colorant values with a value of a replacement colorant (the "colorant component replacement"/Pantone-equivalent workflow).
• Claim 16 — The replacement of Claim 15 is color-dependent.
• Claim 17 — At least one determined colorant value is corrected by a one-dimensional gradation correction.
• Claim 18 — Color correction is based on triangulation of an ordered (regular-grid) set of calibration points in trajectory space.
• Claim 19 — Color correction is based on triangulation of a non-ordered calibration point set (e.g., Delaunay triangulation).
• Claim 20 — Color correction uses at least one polynomial predicting trajectory-space values from color-space values.
• Claim 21 — Color correction is obtained by inverting at least one polynomial that predicts color-space values from trajectory-space values.

Prosecution / Post-Grant Notes and Uncertainty

• Assignment chain of record: Agfa-Gevaert (2002) → Agfa Graphics N.V. (2007, with two corrective assignments recorded in 2009 fixing the naming of the assignor) → Agfa NV (2018, change of name). One 2007 assignment record oddly lists assignor "Mark Barrett," later corrected on 2009-08-13 to "Agfa-Gevaert N.V." — an apparent recording error in the assignment database, noted here literally.
• No litigation found. I located no district court or CAFC case naming US 7,265,870. The Agfa enforcement actions surfaced in my searches (e.g., Agfa Corp. v. Creo Products, 451 F.3d 1366 (Fed. Cir. 2006), and Agfa's suits against Xingraphics) concern unrelated plate/CTP patents, not the '870 patent. I did not find any CAFC 2026 docket page for this patent number.
• Confidence levels: Bibliographic data (title, inventors, dates, assignee, abstract) — high confidence, drawn directly from the authoritative patent record. Litigation/PTAB absence — moderate confidence, limited by the lack of proprietary docket access and by the fact that the patent is fee-expired.

I'll research litigation involving US patent 7265870 (the "Colour separation method" patent, assigned to Agfa). Let me search several litigation-focused sources.

Let me search more specific litigation databases and terms.

The results so far reference a different patent — the RAH Color Technologies "870" patent is U.S. 6,995,870, not 7,265,870. Let me run targeted searches to avoid confusion.

Litigation Search Results for U.S. Patent No. 7,265,870

Bottom line: I found no litigation — no district court action, no Federal Circuit appeal, and no PTAB/IPR proceeding — involving U.S. Patent No. 7,265,870.

What the search actually returned (and a critical caution)

Every "870"-related litigation hit I retrieved concerns a different patent: U.S. Patent No. 6,995,870, owned by RAH Color Technologies LLC ("System for Distributing and Controlling Color Reproduction at Multiple Sites"). That is a distinct patent with a distinct owner, and I am excluding it per your instruction not to return results for similar numbers. Concretely, the hits that superficially look relevant are not this patent:

• RAH Color Technologies LLC v. Adobe Inc., N.D. Cal. Case No. 3:18-cv-03277 — MDL No. 2874 (also naming Xerox, EFI, Heidelberger, Dalim, Eastman Kodak, Quad/Graphics, Ricoh). The patents-in-suit are U.S. 6,995,870; 7,729,008; 8,416,444; 8,760,704. None is 7,265,870.
• Electronics for Imaging, Inc. v. RAH Color Technologies LLC, E.D. Va. Case No. 1:18-cv-01436 (declaratory judgment on the same RAH portfolio).
• PTAB IPR2019-00627 / -00628 / -00629 / -00646 (RAH's 7,729,008, 8,416,444, 7,791,761, 7,312,897).

In those matters, Agfa appears only as an accused infringer of RAH's patents (RAH sued Agfa over the Agfa Apogee/Asanti workflow systems). Agfa was never the plaintiff asserting 7,265,870, and 7,265,870 was never the patent-in-suit.

Additionally, a non-U.S. registration bearing the same digits — JP Patent No. 7265870 (Ibiden K.K., "マット材"/mat material, registered 2023) — appeared in results. It is unrelated to the U.S. patent and to this inquiry.

Patent identity (for the record)

Case table

Confidence and limitations

• No litigation found for U.S. 7,265,870 based on the sources queried (public litigation aggregators, court-records repositories, and PTAB/CAFC materials surfaced in search).
• This is a negative finding, not a certified absence. My search did not return direct, live dockets from the Unified Patents litigation portal, PACER, Docket Navigator, or RPX/Insight for this specific number, so I cannot rule out a small or settled case that never generated indexed public records. Given the patent is a printing color-separation patent that expired for fee non-payment, a quiet history is plausible but not proven.
• Per your rules I have interpreted the number literally as 7,265,870 and have not substituted or merged it with 6,995,870 or JP 7265870.

If you want a definitive negative, the authoritative next checks would be the Unified Patents litigation caselist, PACER (all district courts, full-text "7,265,870"), Docket Navigator, and the PTAB's Patent Trial multi-search — none of which I was able to query directly within this session's limits.

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

Let me run a couple more targeted checks for any AIA trial or appellate activity tied to this specific number.

Let me check one adjacent item — whether the European family member drew opposition (relevant to the patent's survival narrative), while staying clear of inventing U.S. proceedings.

Proceedings overview

No AIA trial proceedings exist on U.S. Patent No. 7,265,870 — zero IPRs, zero PGRs, zero CBMs, zero derivations. The structured "PTAB proceedings on file" block from the USPTO Open Data Portal returns an empty set, and targeted web searching for the number in PTAB contexts surfaced only unrelated matters (see verification note below).

Bottom-line defensive posture: This is not a "hardened by surviving two IPRs" patent, and it is not a "claims canceled, troll has no case" patent either. It is an untested patent that was never worth challenging — because it expired for failure to pay maintenance fees with an adjusted expiration of 2025-08-07. That combination produces an unusual posture: a defendant today faces no live injunctive or ongoing-royalty exposure, but a fee-expired patent can still support a past-damages demand for infringement occurring before expiry within the § 286 six-year lookback (i.e., conduct from roughly 2020-09-27 onward). Because no PTAB trial ever ran, there is no IPR estoppel, no FWD to point to, and no claim that has been canceled or adjudicated — the full prior-art universe remains available in district court under §§ 102/103/112, unconstrained by § 315(e)(2).

Proceedings on file

None. There are no proceeding subsections to write, and I will not manufacture proceeding numbers, panels, or decisions. Per the operating rules, the absence of a proceeding is reported as an absence, not inferred into one.

Verification performed (so the negative is auditable)

False positives encountered and excluded (important, because these are the near-misses a hurried searcher would mis-cite):

• U.S. Pat. No. 6,995,870 (RAH Color Technologies) — a different patent entirely. IPRs on the RAH portfolio (e.g., IPR2019-00627/‑00628/‑00629/‑00646) concern 7,729,008; 8,416,444; 7,791,761; 7,312,897 — not 7,265,870. Agfa appears in that story as an accused infringer, not as patent owner asserting the '870.
• Agfa-affiliated IPRs on other patents — e.g., Eastman Kodak Co., Agfa Corp., Esko Software BVBA & Heidelberg v. CTP Innovations, LLC, IPR2014-00791, on U.S. 6,611,349 (Vogt et al.), with a companion denial in IPR2013-00474. Agfa is a petitioner on the other side of the table, challenging someone else's patent. Different patent, different role.
• JP Patent No. 7265870 (Ibiden K.K., マット材 / mat material, registered 2023-04-19) — a Japanese registration that merely shares digits. Unrelated.

This is consistent with, and corroborates, the previously generated Litigation summary section, which likewise found no district-court or CAFC action on this patent and flagged the same RAH/6,995,870 confusion risk. No contradiction between that section and this one.

Strategic summary

Claim-by-claim status of the 21 claims: every single one is UNTESTED. Claim 1 (the sole independent claim) has never been before the Board. Claims 2–21 — including the four-color CMYK claim (12), the six-ink light-cyan/light-magenta claim (13), the Hexachrome-style orange/green claim (14), the colorant-replacement claims (15–16), the gradation-correction claim (17), and the color-correction claims keyed to ordered triangulation (18), non-ordered/Delaunay triangulation (19), polynomials (20), and polynomial inversion (21) — have never been construed by an APJ panel, never been the subject of an institution decision, and never been canceled or confirmed. A defendant cannot say "claim 1 is dead," and a patent owner cannot say "claims 1–21 survived PTAB scrutiny." Neither statement is available; the claims are simply unadjudicated. The practical implication is that all validity work remains un-done, which cuts against the patent owner on cost and against the defendant on certainty — there is no PTAB record to borrow for claim construction or obviousness.

Estoppel landscape: § 315(e)(2) is a non-issue, and that is the single most useful fact here. Because no IPR or PGR reached a final written decision, no petitioner, real party in interest, or privy is estopped from anything. A defendant being asserted against today faces no "raised or reasonably could have raised" bar, no Sotera-style carve-outs to negotiate, and no risk of collateral estoppel from a Board construction. Every § 102/§ 103 ground based on patents and printed publications is available, as are § 101 and § 112 grounds — the latter two being unavailable in IPR anyway (§ 311(b) limits IPR to §§ 102/103 on patents and printed publications), which matters because a 2001-priority colorimetry claim is a plausible § 101 target that was never tested in any forum. Note, however, that a defendant who files an IPR now and later needs to rely on the same art in litigation would then create its own estoppel — for a fee-expired patent whose only exposure is past damages, an IPR is usually a poor spend compared to invalidity defenses in court.

Pattern signals: none, and the absence is probative. This is a pre-AIA patent (filed 2002-11-20, priority 2001-11-26), so IPR was available against it from 2012-09-16 onward with no nine-month PGR waiting period. It sat exposed to AIA trials for thirteen years and never drew a single petition. That tells you the market did not perceive it as an assertion threat: it was a printing-workflow patent held by a practicing operating company (Agfa) that appears never to have enforced it, and it lapsed for non-payment of maintenance fees. There is no defensive aggregator (Unified Patents or similar) in the chain, no serial petitioner, and no PTAB-to-CAFC appeal history — because there was no PTAB case to appeal. The only adjacent foreign-track item I located for the family is a UK IPO journal listing of EP1315366 (the EP member, App. No. 01000662.5, Agfa-Gevaert); I found no European opposition decision or Board of Appeal decision on EP1315366 in the material retrieved, and I am explicitly flagging that as an unchecked item rather than asserting a clean European prosecution history.

Recommended next steps

If you are a defendant receiving a demand letter citing any claim of 7,265,870:

1. Lead with the expiration, and pin down the damages window. The patent records an adjusted expiration of 2025-08-07 (status: Expired – Fee Related). There is no ongoing infringement to enjoin. The only exposure is past damages, capped by 35 U.S.C. § 286 at six years before filing — so the assertable window is conduct on or after approximately 2020-09-27. Ask for the accused-activity date range in writing; a demand covering post-2025-08-07 conduct is facially meritless.
2. Confirm the fee lapse on the USPTO Patent Center maintenance-fee page (https://patentcenter.uspto.gov/ , application 10/300,162) and check whether any petition to accept late payment was filed — a reinstatement would change the analysis, and I have not verified that. Treat the Google Patents "adjusted expiration" line as a strong but secondary indicator until you pull the fee record directly.
3. There is no FWD to cite, so do not look for one. If opposing counsel implies PTAB validation of these claims, that is unsupported — nothing exists. Conversely, do not represent that any claim has been invalidated; none has.
4. Invalidity is fully open but must be built from scratch. The specification itself supplies the best obviousness runway: it candidly frames the invention as a separation of color correction from color separation by interpolation along trajectories, and expressly situates itself against masking equations (Hunt; Yule) and Neugebauer-based inversion (Mahy & Delabastita, Inversion of the Neugebauer Equations, Color Research and Application, Vol. 21, No. 6, Dec. 1996 — cited on the face of the patent). The patent also cites U.S. 4,334,240 and the Kasson (Proc. SPIE 2170, 24 (1994)) and Kang (Proc. SPIE 2414, 104 (1995)) interpolation papers. Note that the EP search report for the family cited Kasson et al., "Performing color space conversions with three-dimensional linear interpolation," J. Electronic Imaging 4(3):226–249 (1995) and Johnson, "Methods for characterizing colour printers," Displays 16(4):193–202 (1996) — both are useful § 103 starting points that were considered but never litigated.
5. Consider § 101 and § 112 rather than IPR. IPR cannot reach either, and a claim reciting "determining colorant values using an interpolation technique" invites an Alice step-two fight over whether the ordered combination is more than a mathematical manipulation of data plus generic computer implementation. That theory has never been adjudicated.
6. If you nonetheless want an IPR (e.g., to force a stay or because asserted damages are large): IPR remains statutorily available against a pre-AIA patent even after expiry, and there is no § 315(b) one-year bar unless and until you are served with a complaint. But weigh the estoppel you would create against the fact that you are defending past damages only.

If no PTAB activity exists and your question was simply "what happened at the Board": the answer is that nothing happened, at any point, ever. Thirteen years of AIA availability, zero petitions. The absence is itself the signal — well-asserted patents attract IPRs, and this one was never asserted. Cross-reference the previously generated Litigation summary, which reached the same conclusion on the district-court and CAFC side.

Links for your own verification (I am citing these as search entry points, not as opinions I have read, because no proceeding or decision exists to link to):

• PTAB E2E / Patent Trial multi-search: https://ptab.uspto.gov/ (search "7,265,870" / application 10/300,162)
• USPTO Patent Center (fee status, application 10/300,162): https://patentcenter.uspto.gov/
• CourtListener docket search for US 7,265,870: https://www.courtlistener.com/?q=%227265870%22
• Google Patents family/legal-events record: https://patents.google.com/patent/[US7265870B2](/patent/US7265870B2)/en

Confidence: That the structured PTAB dataset shows no proceedings on this patent — high confidence (it is the designated canonical source and my independent searches found no contradicting proceeding). That the negative is complete as to any small, settled, or unindexed case — moderate confidence, limited by the absence of direct PACER / Docket Navigator / Unified Patents caselist access in this session. I have not fabricated any proceeding number, panel, or disposition to fill that gap.

I'll research the assignment chain and cross-reference ownership signals before writing up.

Let me verify the assignment records and Agfa's corporate status.

I have the authoritative assignment chain from the patent record. I could not pull live correspondent fields from Assignment Center in this session, so I flag that gap explicitly rather than inventing attorney names.

Assignment & Ownership Analysis — U.S. Patent No. 7,265,870 B2 ("Colour separation method")

Method note / sourcing. The chain below is reconstructed from the legal-events and reassignment records reproduced on the patent's own Google Patents record (https://patents.google.com/patent/[US7265870B2](/patent/US7265870B2)/en), which mirrors the USPTO assignment data. Two reel/frame numbers are exposed in that record and are cited verbatim: 019179/0836 and 019278/0553. The correspondent-of-record field (the attorney/agent who filed each recording) was not retrievable from the sources available in this session — I state that as a gap and have not substituted the assignor-side signatories for it. Corporate-status facts were corroborated against Agfa's own filings and the Belgian/Polish/UK registry change-of-name notices surfaced in search.

Inventors

Pattern check — no departure anomaly detected. Both inventors are career Agfa color-science personnel and both appear as assignors on the filing-date assignment to Agfa-Gevaert (2002-11-20). There is no record of either inventor re-acquiring, re-assigning, or separately monetizing this patent. There is no evidence of the "all inventors depart within 12 months → portfolio fire-sale" pattern; the patent simply stayed inside the Agfa corporate family for its entire life. Caveat: employment records confirming tenure beyond the filing date were not independently verified; the assessment rests on the assignment record and the inventors' publication history.

Original assignee

• Entity named on the issued patent (grant 2007-09-04): AGFA GRAPHICS NV, by virtue of the 2007 recorded transfer from its parent/affiliate Agfa-Gevaert (see timeline). At filing (2002-11-20) the inventors assigned to AGFA-GEVAERT.
• Current assignee of record: AGFA NV (Mortsel, Belgium), by change of name from Agfa Graphics NV recorded 2018-01-03 (USPTO) and registered across foreign jurisdictions in March–May 2018.
• Line of business: Operating company. Agfa Graphics (now Agfa NV) is a prepress and industrial-inkjet supplier — plates, RIP/workflow software, and wide-format inkjet systems. Agfa-Gevaert NV is the listed parent group (Euronext Brussels), historically organized into Agfa Graphics, Agfa HealthCare, and Agfa Specialty Products.
• Product embodiment: The claimed subject matter (color separation using a trajectory space) is the kind of function performed inside prepress color-management/RIP stacks. Agfa's graphics products include color-management functionality of this general type, but I have not verified that any specific Agfa product practices the claims of the '870 patent; treat "product embodying the claims" as unconfirmed.
• Current status: Operating (not dissolved, not in bankruptcy). The patent itself is Expired – Fee Related, adjusted expiration 2025-08-07.

Assignment timeline

Only two reel/frame numbers (019179/0836 and 019278/0553) are exposed in the accessible record. For the remaining links I report the event and date from the legal-events record and mark the reel/frame as not retrieved — I do not fabricate reel/frame values.

• 2002-11-20 (executed) / recorded 2002-11-20 — Reel not retrieved/not exposed

  • Conveyance: Assignment (Assignment of Assignors' Interest)
  • Assignor: Delabastita, Paul; Vande Velde, Koen
  • Assignee: AGFA-GEVAERT
  • Correspondent: not retrievable in this session — data gap
  • Context: Initial filing-date assignment — inventors convey rights to their employer, the original operating-company assignee.

• 2007-04-12 (recorded) — Reel not retrieved/not exposed (later corrected at Reel 019179/0836)

  • Conveyance: Assignment (Reassignment)
  • Assignor of record: "BARRETT, MARK" — an apparent recording error, corrected on 2009-08-13 to AGFA-GEVAERT N.V.
  • Assignee: AGFA GRAPHICS N.V.
  • Correspondent: not retrievable in this session — data gap
  • Context: Internal corporate reorganization — transfer of the graphics/prepress patent estate from Agfa-Gevaert to the Agfa Graphics business entity.

• 2007-05-07 (recorded) — Reel not retrieved/not exposed (later corrected at Reel 019278/0553)

  • Conveyance: Assignment (Reassignment)
  • Assignor of record: "Corporate IP Department, Manager, General, Theunis, Patrick" — also an apparent recording error, corrected on 2009-08-20 to AGFA-GEVAERT N.V.
  • Assignee: AGFA GRAPHICS NV
  • Correspondent: not retrievable in this session — data gap
  • Context: Internal corporate reorganization (second/companion housekeeping link in the same transfer).

• 2007-09-04 — (grant event, not an assignment) — Publication/grant of US 7,265,870 B2

  • Context: Patent issues to Agfa Graphics NV.

• 2009-08-13 (recorded) — Reel 019179/0836

  • Conveyance: Correction (corrective assignment — "correct the assignor from Mark Barrett to Agfa-Gevaert N.V., previously recorded on reel 019179 frame 0836")
  • Assignor (corrected): AGFA-GEVAERT N.V.
  • Assignee: AGFA GRAPHICS NV
  • Correspondent: not retrievable in this session — data gap
  • Context: Bookkeeping correction only — fixes a mis-recorded assignor name; no change in beneficial ownership.

• 2009-08-20 (recorded) — Reel 019278/0553

  • Conveyance: Correction ("correct the assignors from Patrick Theunis, General Manager, and Corporate IP Department to Agfa-Gevaert N.V., previously recorded on reel 019278 frame 0553")
  • Assignor (corrected): AGFA-GEVAERT N.V.
  • Assignee: AGFA GRAPHICS NV
  • Correspondent: not retrievable in this session — data gap
  • Context: Bookkeeping correction only — same defect as above, in the companion recording.

• 2018-01-03 (recorded) — Reel not retrieved/not exposed

  • Conveyance: Change of Name
  • Assignor: AGFA GRAPHICS NV
  • Assignee: AGFA NV
  • Correspondent: not retrievable in this session — data gap
  • Context: Change of name only — no change in beneficial ownership; corroborated by foreign registry name-change notices (Agfa Graphics NV → Agfa NV) filed March–May 2018.

• 2025-08-07 — (legal-status event, not an assignment) — Adjusted expiration; status Expired – Fee Related.

Recorded security interests, licenses, mergers or releases: none found. No Security Agreement, Merger, License, or Release is recorded against this patent in the accessible record.

Timeline diagram

timeline
    title Ownership of US 7265870
    2001 : Priority date 26 Nov
    2002 : Filed as app 10 300 162
         : Inventors assign to Agfa-Gevaert
    2007 : Patent granted 4 Sep
         : Recorded transfer to Agfa Graphics NV
    2009 : Two corrective assignments
    2018 : Name change to Agfa NV
    2025 : Expired for fee non-payment

NPE / troll-pattern signals

Verdict

Defensive / non-asserting.

Justification: the record shows a pure single-family corporate chain — inventors → AGFA-GEVAERT (executed/recorded 2002-11-20), → AGFA GRAPHICS N.V. (recorded 2007-04-12 and 2007-05-07, defective assignor names corrected on Reel 019179/0836 and Reel 019278/0553 in 2009), → AGFA NV by change of name (recorded 2018-01-03) — with no arm's-length transfer ever occurring and no NPE, shell LLC, security interest, license, or aggregation vehicle anywhere in the chain; combined with the absence of any asserted suit (cross-referenced to the earlier Litigation and PTAB sections) and the 2025-08-07 fee expiration, this patent was never used as an assertion asset. Caveat on the label: the taxonomy's strict trigger for this verdict (termination at a defensive aggregator) is not literally met — the patent is neutralized by ownership retention plus fee lapse, not by aggregation — so I flag the definitional mismatch rather than stretch it; the operative conclusion (non-NPE, non-asserting) is well supported.

NPE verdicts affirmatively rejected: "NPE — high confidence" and "NPE — moderate confidence" are not supported by any reel/frame entry; "Operating-company assertion" is also rejected because, although Agfa is an operating company, it is not suing anyone on this patent.

Verification link (Assignment Center search): https://assignmentcenter.uspto.gov/ (search by patent number 7265870 or application 10/300,162).

Residual data gaps I could not close in this session

1. Correspondent-of-record for each recording — unavailable; the recurrence test (signal #3) remains unresolved.
2. Reel/frame numbers for the 2002-11-20, 2007-04-12, 2007-05-07, and 2018-01-03 recordings — only the two corrective-assignment reel/frames (019179/0836 and 019278/0553) were exposed.
3. Whether any petition to accept late maintenance payment was filed before/after the 2025-08-07 expiration — not verified.
4. Foreign-family ownership (EP1315366B1) — the Agfa NV name change is consistent across jurisdictions, but I did not pull the EPO register entry directly.

Confidence: High that the US chain is a pure Agfa intra-corporate chain ending at Agfa NV with no NPE involvement (the assignment events and the two corrective reel/frames are drawn from the patent's own record, and the 2018 name change is corroborated by multiple foreign registries). Moderate on completeness, limited by the missing correspondent data and reel/frame numbers noted above. I have not fabricated any assignment, reel/frame, attorney name, or correspondent.

I'll research the prior-art citations for US 7,265,870 specifically, using USPTO/Google Patents citation data.

I found the "Referenced Cited" list for US 7,265,870. Let me retrieve the details of each cited patent to identify titles, dates, and relevance.

Prior-Art Analysis for U.S. Patent No. 7,265,870 B2 — "Colour separation method"

Source note / provenance. The following is built from the patent's own face record as mirrored on Google Patents (https://patents.google.com/patent/[US7265870B2](/patent/US7265870B2)/en) and Justia (https://patents.justia.com/patent/7265870), which reproduce the "References Cited" list entered during prosecution of application 10/300,162. I confirmed one entry independently (US 4,837,722, Sara) via its Google Patents page. The full-citation list below is drawn directly from that authoritative "Referenced Cited" record; where I could not independently open a given reference, I say so and describe it functionally rather than inventing a title. This builds on — and does not repeat — the earlier summary and litigation sections.

Important framing caveat. The references below are the references the applicant/examiner listed during prosecution. Being cited is not the same as anticipating: this patent issued with claim 1 intact over all of them, so no listed reference was treated by the examiner as a § 102 anticipation of the allowed claims. My § 102 mapping below is therefore an analytic triage of which claim each reference could most plausibly be argued against, not a finding of invalidity. Under pre-AIA § 102, a U.S. patent reference is prior art as of its issue date (§ 102(a)/(b)) or, if by another filed earlier, under § 102(e); the § 102(b) critical date for this patent is ≈ 2000-11-20.

1. U.S. Patent References Cited

2. Foreign Patent References Cited

Titles for these foreign documents were not independently verified in this session.

3. Non-Patent Literature Cited (all pre-date the 2001-11-26 priority date)

These are the most technically substantive references of record for the interpolation/characterization subject matter:

• Po-Chieh Hung, "Smooth Colorimetric Calibration Technique Utilizing the Entire Color Gamut of CMYK Printers," Journal of Electronic Imaging, Oct. 1994, Vol. 3, No. 4, pp. 415–424. → Nearest art for Claims 1, 18–19 (calibration + CMYK gamut).
• James M. Kasson et al., "Performing Color Space Conversions with Three-Dimensional Linear Interpolation," Journal of Electronic Imaging, Jul. 1995, Vol. 4, No. 3, pp. 226–249. → Nearest art for Claims 6–10 (3-D linear/tetrahedral interpolation).
• Po-Chieh Hung, "Colorimetric Calibration in Electronic Imaging Devices Using a Look-Up-Table Model and Interpolations," Journal of Electronic Imaging, Vol. 2, No. 1, 1993, pp. 53–61. → Claims 1, 18–19.
• Tony Johnson, "Methods for Characterizing Colour Printers," Displays, Vol. 16, No. 4, May 1996, pp. 193–202. → Claims 1, 18–19.
• Alexander R. Harbury, "A Color-Corrected Halftone Algorithm for Bilevel Printers," SID International Symposium, Baltimore, May 1989, pp. 90–93. → Claim 1 (halftone color correction).

Additionally, the specification itself expressly relies on the following (they are described in the body, not necessarily in the face "References Cited" list, but are the applicant's own acknowledged prior art):

• U.S. Pat. No. 4,334,240 — cited in the specification as an example interpolation technique (relevant to Claims 6–9).
• Kasson, "Tetrahedral Interpolation Algorithm Accuracy," Proc. SPIE 2170, 24 (1994); and H. R. Kang, "Comparisons of Three-Dimensional Interpolation Techniques by Simulation," Proc. SPIE 2414, 104 (1995) — both cited for LUT/interpolation (relevant to Claims 6–10).
• Mahy & Delabastita, "Inversion of the Neugebauer Equations," Color Research and Application, Vol. 21, No. 6, Dec. 1996 (note: Delabastita is a co-inventor of the '870 patent) — cited for Neugebauer modeling/inversion (relevant to Claims 20–21).
• Yule, Principles of Color Reproduction; R. W. G. Hunt, The Reproduction of Colour (1987) — general color-reproduction background (relevant to Claim 1).

4. Analyst Assessment — Which reference maps to which claim, and how strong

• The strongest § 102 candidate is the interpolation family — Sara (US 4,837,722), Kasson (US 5,390,035 + the Kasson NPL), and the Hung papers. All disclose 3-D interpolation between tabulated color values to perform a color-space conversion. Their natural target is the interpolation mechanism of claim 1 and the sub-volume/barycentric mechanics of claims 6–10.
• The color-correction/calibration family — the Hung NPL, Johnson, and the Kakutani/Kanamori references — target claims 18–21 (triangulation of ordered/non-ordered calibration points; polynomial color correction).
• The colorant-replacement and multi-colorant references (Perumal; and generally the CMYK/Hexachrome background) target claims 12, 14–16.
• No single reference of record discloses the full claim-1 combination — namely, (i) a separate color-correction step mapping the color into an intermediate "trajectory space," (ii) at least two trajectories each carrying pre-defined colorant values, and (iii) interpolation using both the color's trajectory coordinates and the trajectory points' corresponding colorant values. That separation of color management from colorant management is the stated point of novelty, and it is why claim 1 issued over the cited art. The interpolation references operate directly in a device/color space rather than via a separately defined ink-behavior trajectory space.

5. Confidence and limitations

• High confidence: the identity and dates of the references in the "References Cited" list (taken verbatim from the patent's face record) and the confirmed content of US 4,837,722 (Sara).
• Moderate confidence: the functional descriptions of US 5,121,196, US 5,241,373, US 5,268,754, US 5,390,035, US 5,504,821, US 5,581,376, US 5,748,195, US 5,917,994, US 6,304,671, US 6,571,010, and US 6,697,520. I did not open each of these individually (tool-step limit reached mid-search). I have deliberately not fabricated titles for them; I describe them at the level supported by their citation context and inventor/assignee lineage.
• Not verified: exact filing dates of US 6,571,010 and US 6,697,520, which are needed to determine whether they qualify as § 102(e) art (both issued after the 2001-11-26 priority date).
• Methodological caveat: a reference cited in a patent can be prior art under § 102, § 103, or under the "background" of the specification; listing alone does not establish a § 102 anticipation. To convert this triage into a formal invalidity opinion, the next checks would be (a) confirming the § 102 date of each reference against the exact filing dates, and (b) pulling the full text/claims of US 5,390,035 (Kasson) and US 4,837,722 (Sara), which are the two references most likely to be asserted against the broad interpolation/color-conversion concept of claim 1.

I'll ground the obviousness analysis by verifying the key references from the Prior Art section before building combinations.

I have strong grounding, including the EP1315366A1 search report (which shows examiner X/Y designations) and the content of the Hung and Kasson references. Let me verify the extended-ink-set and colorant-replacement prior art relevant to dependent claims 13–17.

I have sufficient grounding, including a key piece of prosecution evidence — the EP1315366A1 search report (the family member of this patent), which shows how the European examiner actually rated these references. Note: I hit the session's step limit before completing the last three verification searches (Hung's US 5,121,196; Hexachrome; six-color inkjet prior art), so those specific points are flagged as lower-confidence below rather than asserted.

Obviousness Analysis of U.S. Patent 7,265,870 under 35 U.S.C. § 103

Scope note. This analysis builds on, and does not repeat, the earlier sections (summary, litigation, PTAB, assignment, prior art). It uses the prior-art set identified in the Prior Art section and cross-checks it against two primary-source search-report tables I retrieved this session: the EP1315366A1 search report (the European family member of this patent) and, for context, the EP0969413A3 search report (a contemporaneous Canon case). All reasoning is grounded in the references of record; I do not invent reference content.

A. Legal framework

Obviousness is a question of law reviewed on the Graham v. John Deere factual inquiries:

1. the scope and content of the prior art;
2. the differences between the prior art and the claims at issue;
3. the level of ordinary skill in the pertinent art; and
4. secondary considerations (objective indicia).

Under KSR Int'l Co. v. Teleflex Inc., 550 U.S. 398 (2007), a claimed combination is obvious if it was "obvious to try," if the improvement was a "predictable use of prior art elements according to their established functions," or if the combination is "the product of ordinary creativity, not of a genius." A motivation to combine may be found in the prior art, in the "design incentives and other market forces," or from the background knowledge of a PHOSITA — it need not be an express teaching.

This is a pre-AIA patent (effective filing 2002-11-20; priority 2001-11-26), so pre-AIA § 103(a) and the pre-AIA § 102 critical date (≈ 2000-11-20 for § 102(b)) apply.

Critical procedural fact that shapes everything below: the applicant allowed claim 1 over the cited art after those references were before the examiner (US face record; see Prior Art section). But the examiner's allowance does not immunize the claims: (a) the examiner addressed references individually against the allowed set, not necessarily the full combination with an explicit KSR rationale; and (b) the strongest obviousness theory here is a combination-of-references theory that need not have been the examiner's ground. So the analysis is worth doing, with the caveat that it must overcome an issued claim.

B. Level of ordinary skill in the art (PHOSITA)

A PHOSITA here would be a color-imaging engineer with a bachelor's degree in electrical/computer engineering (or equivalent) and 2–5 years' experience in device characterization, color separation (RIP/LUT construction) and halftone printing, familiar with:

• the CIE colorimetry chain (RGB/XYZ and device-independent spaces);
• the ICC-profile architecture (source → PCS → destination), which by the late 1990s was the industry-standard way to separate color management from device control;
• Neugebauer modeling and its inversion (Mahy & Delabastita, cited on the face of this very patent);
• UCR/GCR and black-generation curves;
• 3-D LUTs with trilinear/tetrahedral interpolation.

The named inventor Delabastita is himself a co-author of the "Inversion of the Neugebauer Equations" article cited in the patent's background — i.e., the applicant's own technical literature defines the ordinary-skill baseline. This is an admission that the relevant modeling skill was routine.

C. Qualifying prior art and dates

All of the technically central references post-date-free and pre-date the critical date. US 6,571,010 and US 6,697,520 should not be relied upon in a § 103 theory absent proof of their pre-critical-date filing dates.

D. The specification's own admissions — the single best § 103 runway

The Background section concedes the following as known art, which is dispositive of large parts of the claim set:

1. UCR/GCR and black generation as a curve along the neutral axis. The patent states that black is added "when a color is to be reproduced of which the density is too high," that "many variations exist" in how the black amount is calculated, and that UCR/GCR reduce total colorant by "replacing this neutral component by an equivalent amount of black." It also states the requirement that "the selected amount of black ink should fall in between" the minimum and maximum. → The "trajectory ink behavior" concept is not the point of novelty; it is admitted background.
2. The trade-offs (gamut vs. graininess/moiré, ink-limit, metameric robustness) are recited as a known "list of considerations."
3. The two-stage separation (colorimetric management separated from colorant management) — the patent's stated advantage — was the standard ICC architecture of the late 1990s.
4. Splitting a colorant into multiple densities ("light cyan and light magenta colorants are sometimes used… to reduce graininess") and extended ink sets (Hexachrome; the Pantone Matching System's CMYK-equivalents) are expressly described as pre-existing practice — admissions directly against dependent claims 13–16.
5. Delaunay triangulation and Neugebauer polynomials are described as conventional tools — admissions against dependent claims 19–21.

Under KSR, an admission in the specification that a feature is known art is strong evidence of obviousness when that feature is combined with the remaining teachings.

E. Claim 1 — the primary obviousness case

Claim 1 (recap, unamended)

(a) a color-correction step converting color-space values into a "trajectory space"; (b) at least two trajectories with trajectory points, each point having trajectory-space values and corresponding pre-defined colorant values (C1x…CNx); (c) interpolation of the color's colorant values using the color's trajectory-space coordinates and the trajectory points' coordinates and their colorant values.

Ground 1 (primary): Hung '93 + Hung '94 + Kasson + Johnson

Why a PHOSITA would have combined these (KSR rationale): All four references are in the same field (device color characterization/separation), address the same problem (converting colorimetric/device-independent values into colorant values), and are cross-cited in the literature of the day. Hung '94's own text supplies the missing motivation link — it says its smoothed-black method is "suitable for a lookup table and interpolation technique." That is a teaching to combine a black-generation/ink-behavior definition (Hung '94) with an interpolation engine (Hung '93/Kasson/Johnson). Choosing an intermediate 3-D space, defining ink behavior along paths through it, and interpolating is the "predictable use of prior art elements according to their established functions."

The alleged point of novelty evaporates on close reading. The patent frames "trajectory space" as if it were a special construct. But claim 1 does not require the space to be anything other than a 3-D intermediate space (claim 2 confirms it is 3-D), and the "trajectories with corresponding colorant values" are, functionally, black-generation/GCR curves along the neutral, primary and secondary axes — exactly what Hung '94 (black amount) and Johnson (GCR level) describe. The interpolation is ordinary 3-D interpolation (Kasson/Sara). Under KSR, merely naming this collection a "trajectory space" and relocating the LUT nodes onto paths through the space does not confer patentability.

Ground 2 (secondary): Sara + Hung + Johnson

US 4,837,722 (Sara) discloses a 3-D interpolation machine for "transformation from one color space to another," expressly for "graphic arts quality printing system[s], in which images must be color corrected prior to printing," using lookup tables and tetrahedral/trilinear interpolation (confirmed from the source PDF this session). Combined with Hung '94 (ink-behavior/black curves) and Johnson (printer characterization; GCR), the same claim-1 subject matter follows. Sara is the strongest hardware/systems reference because it frames the problem as "color correct prior to printing" — the exact claim-1 architecture.

Ground 3 (polynomial-alternative framing)

US 6,304,671 (Kakutani), US 5,241,373 / 5,504,821 (Kanamori) and Johnson's polynomial-fitting discussion supply claim-1's color-correction limb if one prefers a polynomial implementation, again combined with Kasson/Sara for the interpolation limb.

Why the examiner's allowance is not decisive

The US examiner allowed claim 1 with these references of record. But the strongest theory is the combination — Hung '94's smoothed max/min-black curves serving as the "trajectory" ink definitions feeding a Kasson/Sara interpolation engine — a KSR "predictable-use" rationale that the record does not show was expressly articulated. A § 103 challenge should lead with the express Hung '94→interpolation teaching to combine.

Cross-reference / flag: the Prior Art section correctly identified Kasson and Hung as the nearest art for claims 1 and 6–10. No contradiction. One caution I verified this session: an "X Kasson / X Hung" designation against broad color-conversion claims appears in the search report of EP0969413A3 (Canon), not in EP1315366. Do not attribute the Canon search report's X-ratings to this patent's family — a plausible conflation trap given both involve tetrahedral interpolation.

F. Dependent claims — reference mapping

Aggregate: every dependent claim is either (i) inherent in an identified reference, (ii) a predictable design choice/refinement of a recited element, or (iii) an element the patent itself admits was known.

G. Motivation to combine — the KSR showing

1. Same field, same problem, same solution type. All references are color-imaging/printing characterization references addressing how to map measured/colorimetric values to colorant values. KSR permits combination where references are "addressing the same problem."
2. Express cross-teaching. Hung '94 says its method is "suitable for a lookup table and interpolation technique in practical conversions" — a direct pointer to the interpolation references.
3. Market/design pressure. Johnson (1996) diagnoses the exact problem claim 1 solves: for CMYK "there is no unique solution… some additional definition is required… the GCR level." A PHOSITA was under market pressure to make black generation controllable and interpolation fast (speed/memory) — the very incentives KSR credits.
4. Predictable results. Combining a known 3-D interpolation engine with known GCR/black curves yields only predictable results (faster separation, controllable black) — "the product of ordinary creativity."
5. "Obvious to try." Even if one doubted predictability, a finite number of identified, predictable solutions (choose intermediate space; choose axes; choose interpolation) is the classic KSR "obvious to try" posture.

H. Rebuttal considerations (what the patent owner will argue) and their strength

Best claim for the patent owner: claim 9 (the specific "parallel to the opposite tetrahedron plane" geometry) and, secondarily, the breadth/specificity of claim 3's seven-axis scheme. These are the narrowest structural limitations; a validity challenge should not lead with them.

I. Bottom-line assessment

• Claim 1: Reasonably strong § 103 attack via Hung '93 + Hung '94 + Kasson ('035 and/or NPL) + Johnson, with Sara '722 as an alternative interpolation engine. The Hung '94→interpolation teaching supplies the motivation; the specification's own admissions supply the "trajectory/GCR/ink-limit" elements. Confidence: moderate (must overcome an issued claim and the examiner's implicit consideration of these references).
• Claims 2, 4, 6–8, 10, 12, 15, 17–21: Strong — read directly onto Kasson/Sara (interpolation mechanics), Hung (calibration/polynomial), Johnson (polynomial fitting, GCR/color-component replacement), and Harrington (gradation). Several are effectively single-reference or express-teaching combinations.
• Claims 3, 5, 11, 13, 14, 16: Moderate — predictable design choices, supported by the patent's own admissions (ink sets, Pantone, Delaunay) rather than by a single clean reference.
• Claim 9: Weakest ground; the parallel-plane geometry is a stated preference, so KSR/design-choice arguments are available but contestable.

Net: the claims are more likely than not invalid under § 103 on the strength of the interpolation + black-generation combination — a conclusion that does not depend on any reference whose filing date I could not verify, and that rests on references the patent's own family search report treated as the relevant art.

Confidence and limitations

• High confidence: the identity/dates of the references used (verbatim from the record and the family search report); the content of US 4,837,722 (Sara), US 5,390,035 (Kasson), Hung 1994, and Johnson 1996 (retrieved this session); the specification's admissions.
• Moderate confidence: the exact mapping of some EP-search-report claim designations to this patent's US claim numbers — the EP independent claim is claim 18, whereas the US independent claim is claim 1; the two claim sets are not numerically aligned, so I used the search-report designations only as evidence of the examiner's art selection, not as claim-by-claim findings.
• Not verified (step-limit cutoff): US 5,121,196 (Hung patent) content; a specific documentary reference for six-ink light-cyan/light-magenta inkjet separation (claim 13); a specific Hexachrome prior-art citation (claim 14). For those two claims I rely on the patent's own admissions and identify them as the weaker links.
• Methodological caveat: a reference cited in a patent may be art under § 102, § 103, or merely "background." Listing alone is not anticipation. To convert this triage into a formal invalidity opinion, the next steps are (1) pull the full text/claims of US 5,390,035 (Kasson) and US 4,837,722 (Sara); (2) obtain the complete EP1315366A1 search report to read every X/Y designation in full context; and (3) verify the pre-critical-date filing dates of US 6,571,010 and US 6,697,520 before relying on them.

Key sources for verification: EP1315366A1 search report (http://patentimages.storage.googleapis.com/9b/6e/64/c3dc97f8ba46d6/EP1315366A1.pdf); US 4,837,722 (https://patents.google.com/patent/[US4837722A](/patent/US4837722A)/en); US 5,390,035 (https://patents.google.com/patent/US5390035); Hung 1994 abstract (https://www.semanticscholar.org/paper/827bc05d61b4c974c931fdcec70d120c54cbb4f7); Johnson 1996 (DOI 10.1016/0141-9382(96)01013-x); US7265870B2 (https://patents.google.com/patent/US7265870B2/en).