US 5675717

I'll search for current information on this patent number, including any litigation or CAFC docket activity.

Let me check specifically for any litigation or court docket activity involving this patent.

US Patent 5,675,717 — Analyst Summary

Litigation / docket check (USPTO + CAFC, as of April 26, 2026): My searches returned no Federal Circuit appeal, district court case, ITC action, or PTAB proceeding involving US 5,675,717. The only third-party litigation hits that surfaced for the string "5,675,717" were unrelated (the Taction v. Apple haptics verdict and a SmartSky v. Gogo matter concerning a different "717 patent" — U.S. 10,257,717). Note the distinction: 5,675,717 is a 1997 Canon patent, whereas the "717 Patent" in SmartSky v. Gogo is 10,257,717. I found no evidence of any 2026 CAFC docket tied to 5,675,717. I cannot confirm a negative exhaustively from web search alone, so treat "no litigation found" as high-but-not-certain confidence. A 2026 appeal is also temporally implausible given the patent expired in 2015 (see below).

---

Bibliographic data

Data discrepancy to flag: The Google Patents record (the authoritative full text supplied) gives priority 1994‑01‑31, filing 1995‑01‑25, issue 1997‑10‑07. The Unified Patents portal lists priority 1994‑01‑30, application 1995‑01‑24, grant 1997‑10‑06 — a one-day offset in each date, likely a time-zone/normalization artifact or related to the 1998 certificate of correction. I am reporting the Google Patents values as authoritative per the source hierarchy, but noting the discrepancy rather than reconciling it silently.

Related family members: US 5,960,162 A (divisional of 08/378,371, filed 1997‑05‑20, issued 1999‑09‑28); JP 3491998 B2; JPH07254060 A.

---

Abstract (verbatim)

"In an image processing apparatus for previewing a color image, by using a color output device for forming a preview image, processing is decentralized, the load on a host computer is reduced, and a preview-processing program can be easily formed. The apparatus includes a communication device, e.g., a bidirectional interface, for communicating with a host computer, an output device for outputting image information for image formation to an image forming device for forming an image on a recording material based on input color-image information from the host computer, and a color conversion device for previewing an image which generates color-preview-image information for correcting the image formed by the image forming device on a color monitor. The color-preview-image information is transmitted to the host computer by the communication device."

---

Plain-language overview of the independent claims

Claim 1 — Apparatus (the "printer does the preview" claim). An image forming apparatus (i.e., the printer itself) that:

1. inputs color-image information from a host computer over a bidirectional communication line;
2. forms an image on a recording material (paper) based on that color-image information;
3. generates color-preview-image information — data representing how the printed image will look — for reproducing the image on a color monitor; and
4. sends that preview information back to the host computer over the same bidirectional line.

The inventive point is architectural: rather than the host PC simulating the printer's color pipeline (the prior art approach), the printer generates the preview and ships it back to the PC, decentralizing processing. The dependent claims (2–8) add: per-image-type color processing (natural vs. CG images); preferred color matching for natural images vs. color-difference-minimizing matching for CG images; a determination of whether preview is to be performed at all; host-supplied color-conversion information; storage of color-conversion information in the apparatus; and generation of preview data matched to the printer's number of gradation steps.

Claim 9 — Method (the "reduce resolution then send" claim). A method of previewing a color image on a color monitor before printing it, comprising:

1. outputting color image data for the image to image-forming means that will print it on a recording material;
2. generating color-preview-image information for reproducing the printed color image on a color monitor;
3. reducing the resolution of the generated color-preview-image information; and
4. transmitting the resolution-reduced preview information to a host computer, which forwards it to the color monitor for preview.

The added limitation relative to Claim 1's concept is the resolution-reduction step — the specification ties this to the 360‑dpi print resolution being down-converted to 90‑dpi monitor resolution so the preview transfer is smaller and faster without visible degradation. Dependent claims 10–16 mirror the Claim 2–8 subject matter (image-type-specific color processing, natural vs. CG matching, preview-determination step, host-supplied conversion info, storage of conversion info, gradation-step matching).

Claims 2–8 are dependent on Claim 1; Claims 10–16 are dependent on Claim 9. Two independent claims total.

---

Scope caveat

Only 16 claims are present, and all four disclosed embodiments (a binary-printer variant with printing memory 24; a preview-dedicated-memory variant 41; a CG-vs-natural-image variant with separate color-processing memories; and a combination of the latter two) fall within the two independent claims' scope. The claim language does not itself recite resolution conversion in Claim 1 (that appears only in Claim 9), nor does it recite CG/natural image-type discrimination (only dependent claims 2–4 and 10–12). A reviewer should not read those features into the independent claims. I have no authoritative information indicating this patent was ever asserted, licensed, or drawn into any 2026 CAFC docket.

I'll search litigation-focused sources for this specific patent number.

Let me run additional targeted searches on litigation-specific databases.

Litigation Search Results — US 5,675,717

Bottom line

No litigation involving US 5,675,717 is known. My searches returned no district court case, ITC Section 337 investigation, PTAB proceeding, or CAFC appeal in which US 5,675,717 (application 08/378,371; Canon; Yamamoto) was asserted, challenged, or named. Because no case exists in the sources I could reach, I cannot supply plaintiff, defendant, jurisdiction, case number, filing date, or outcome for any case — there is no case to populate those fields with.

This is a negative finding, not a confirmed absence. I can state with high confidence that no publicly indexed litigation exists; I cannot prove a negative exhaustively from web search.

---

What I searched

---

Look-alike identifiers I saw and excluded (per the strict-literal rule)

These surfaced repeatedly and are not US 5,675,717:

1. Taction Technology, Inc. v. Apple Inc. — the "5.7 billion" figure that keeps matching a 5,675,717-style query. $5.7bn jury verdict, S.D. Cal. (San Diego), jury verdict returned Sept. 25, 2026 (per Burford Capital's Sept. 28, 2026 press release). Patents asserted: US 10,659,885 and US 10,820,117 — haptic feedback. Filed 2021; Fed. Cir. revived after a 2023 loss. Completely unrelated to a 1997 Canon color-preview patent.
2. SmartSky Networks, LLC v. Gogo Business Aviation, LLC, D. Del. 1:22‑cv‑00266 — the "717 Patent" there is US 10,257,717 (held indefinite as to Claim 1 on summary judgment, Aug. 7, 2025). Different patent, different decade.
3. JP 5675717 B2 (published 2015‑02‑25) — a Japanese patent to Hitachi Automotive Systems for an air physical-quantity detection device (G01F 1/684). Same digits, unrelated jurisdiction and subject matter.
4. EP 1,312,043 B1 — this one does involve our patent, but as prior art cited against it, not as litigation: the EP specification states "US‑A‑5 675 717 discloses an image processing method for previewing a color image on a color monitor prior to forming the color image onto a recording material." That is a citation in prosecution/opposition, not a lawsuit involving the '717 patent.

---

Why the absence is expected

• Expired. The Google Patents record shows status "Expired – Lifetime," with anticipated expiration 2015‑01‑25. A patent out of force since 2015 cannot be asserted for damages accruing after expiry and generates no live 2026 docket.
• Prosecution/ownership profile. Original and current assignee is Canon Kabushiki Kaisha throughout; the only recorded post-issuance events in the legal events log are fee payments (4th, 8th, 12th year) and a 1998 certificate of correction. No assignment to any NPE, no reexamination, no reissue, no terminal disclaimer, no patent-security-agreement trail.
• Family. The division US 5,960,162 (from 08/858,507, filed 1997‑05‑20) similarly shows no assertion in the sources checked.

---

Two flags on the record

1. Date inconsistency in the corpus. Your prior section is dated April 26, 2026, but the live search results contain material dated September 25–28, 2026 (the Taction verdict and Burford release), and the patent text was fetched 2026‑09‑24. So the "as of" date in the prior summary is earlier than some evidence now on hand. This does not change the litigation conclusion (none of that later material concerns the '717 patent), but the analyst should not treat "April 26, 2026" as the current search horizon.
2. Prior-section carryover is consistent. The earlier summary's finding of "no litigation" is reproduced, not contradicted, by this round of searching. No contradiction to report.

---

How to close this out definitively (sources I could not query directly)

Web search cannot substitute for a docket-level negative. To convert "not found" into "none," query:

• PACER / CourtListener RECAP — full-text docket search on "5,675,717" and "5675717"; RECAP's archive is the most complete free federal-filing corpus (free.law).
• Unified Patents litigation caselist (portal.unifiedpatents.com/litigation/caselist) — filter by patent.
• Docket Navigator and Lex Machina — the commercial patent-litigation databases (Lex Machina's annual patent report is the standard source for assertion counts).
• USPTO Patent Center / Assignment and PTAB E2E — for any reexam, IPR/PGR, or assignment that would signal assertion.
• ITC EDIS — search Section 337 investigations for the patent number.

Each is a subscription, login, or API-gated resource I do not have access to here, which is precisely why my conclusion is stated as a high-confidence negative rather than an affirmative clearance.

I'll verify the ODP result with targeted web searches before concluding.

PTAB / AIA Trial Proceedings — US 5,675,717 (Yamamoto / Canon)

Bottom line up front: there are no AIA trial proceedings on this patent. Zero. The structured "PTAB proceedings on file" block (USPTO ODP) returns no IPR, PGR, or CBM for US 5,675,717, and my independent searches found nothing to contradict it — no petition, no institution decision, no Final Written Decision, no appeal, no settlement. A document-by-document walkthrough is therefore not possible, and I will not manufacture one.

---

Proceedings overview

Total AIA trial proceedings on US 5,675,717: 0 — no active proceedings, no claims invalidated, no claims sustained, no settlements, no institution denials, no FWDs, no CAFC appeals. One sentence on the defensive posture: because no tribunal has ever construed or tested these claims, the patent is not "hardened" by surviving IPRs — it is untested, and the correct defensive read is that the claims carry their original 1997-issue scope with no PTAB-populated invalidity record to work from, but also with no § 315(e)(2) estoppel binding the petitioner side, which means a defendant today is free to build an IPR from scratch.

There is one important triple-negative to state plainly: (1) no AIA proceeding, (2) no ex parte reexamination, and (3) no Federal Circuit appeal surfaced in my searches. All three are "not found," not "proven absent."

Verification performed (2026-10-01)

Searching a bare number like this reliably surfaces the wrong patent, because several live, heavily-litigated patents are also called "the '717 patent." A defendant researching 5,675,717 will hit these and must not conflate them:

None of the above is a proceeding on US 5,675,717. Do not cite any of them as such.

---

Why this patent never attracted an AIA trial — and why that isn't a clean bill of health

The absence is explainable and largely structural, not evidence of strength:

1. Expiration precedes the AIA trial boom. Anticipated expiration is 2015-01-25 (Google Patents; Unified Patents shows 2015-01-24 — the one-day offset already flagged in the prior section). The IPR regime started 2012-09-16. There was roughly a 28-month window in which an IPR was even theoretically available, and the patent's remaining life was short enough that defendants had little incentive to spend on a petition rather than ride out the term.
2. The one-year § 315(b) bar likely did the rest. Any defendant served with a complaint on this patent would have had 12 months to petition — and there is no evidence of such service in the record I reviewed.
3. PGR was never available. PGR applies only to patents with an effective filing date on or after 2013-03-16. This patent's priority is 1994-01-31.
4. CBM was never available. CBM covered financial-services business-method patents; this is a color-preview/printer-controller patent. CBM also sunset on 2020-09-16.
5. No defensive aggregator stepped in. Unified Patents maintains a profile page for this patent (https://portal.unifiedpatents.com/patents/patent/US-5675717-A) with a "Patent Art (4)" tab, but there is no Unified-filed challenge associated with it. No RPX involvement either.

The patent was, however, respected as prior art: EP1312043B1 (Transact Technologies, two-color thermal POS printing) states at [0009]: "US-A-5 675 717 discloses an image processing method for previewing a color image on a color monitor prior to forming the color image onto a recording material." That is an examiner-driven citation in a foreign prosecution, not a PTAB trial, but it tells you the disclosure was treated as material prior art by a different office. (EP1312043B1)

Near-miss check: the divisional

The prior section correctly identifies US 5,960,162 A as a division of 08/378,371 (filed 1997-05-20, issued 1999-09-28), titled "Image processing apparatus which converts multi-value density data and outputs the converted data." I searched for it in the same pass and found no PTAB proceeding on the '162 patent either — but I did not run a dedicated, exhaustive docket pull on the '162, so treat that as "none located," not "confirmed none." If you are clearing both members of this family, run a fresh ODP query against 5,960,162.

---

Strategic summary

Claim status across the entire patent. All 16 claims are UNTESTED. Nothing is canceled; nothing has been adjudicated patentable. Claims 1–8 (apparatus) and 9–16 (method) stand exactly as issued in 1997, subject only to the 1998-04-14 certificate of correction noted in the legal-events record. There is no PTAB record to narrow or knock out claim 1, and therefore no "the troll's claims are already dead" argument available. Conversely, there is no adverse PTAB construction or FWD for a patent owner to wield as a validity shield.

Estoppel landscape — this is the good news for a defendant. Because no IPR/PGR/CBM was ever instituted, § 315(e)(2) estoppel is a blank page. No petitioner and no privy of any petitioner is barred from raising any ground. A defendant today may petition on any § 102/§ 103 combination it can assemble from the 1994-era printing, color-management, and host-based preview art — including the four references the examiner did cite, which are available to you precisely because they were never litigated in a trial:

• US 5,081,529 A (Eastman Kodak, 1992-01-14) — color and tone scale calibration for a printer using electronically generated input images
• US 5,146,323 A (Hitachi, 1992-09-08) — white balance adjustment in a color video printer
• US 5,210,600 A (Fuji Xerox, 1993-05-11) — film image parameter extraction
• US 5,276,511 A (Fuji Photo Film, 1994-01-04) — image processing conditions based on finishing information

Also unencumbered: DE 3347049 C2 (Rudolf Hell GmbH, 1986-07-17), "Method and circuit arrangement for simulating multicolor printing on a color monitor." That is a startlingly on-point reference for the core "simulate the printer on a monitor" concept and sits on the face of the record as a family-cited document. Any petitioner should start there.

Practical reality check on whether a petition is even worth filing. The patent expired 2015-01-25. Under 35 U.S.C. § 286, damages are recoverable only for infringement occurring within six years before suit; all possible infringement here predates 2015-01-25, so a complaint filed today that sought back damages would be reaching conduct more than ten years stale and, for the pre-2015 window, would run straight into § 286 laches-adjacent limitations. In practice an IPR on an expired patent is a low-value exercise: the Board can and does institute on expired patents, but there is nothing left to enjoin and no live royalty base. The more efficient defense is the expiration-plus-§ 286 argument, not a petition. Reserve IPR for the scenario where a patent owner asserts the patent anyway and you need a merits kill for fee-shifting purposes.

Pattern signals. No petitioner has ever filed on this patent, so the "repeat petitioner" question is moot. Canon never had to defend it at the Board, so there is no history of aggressive PTAB appeals by the patent owner. No defensive aggregator is in the chain. The one behavioral signal worth flagging is prosecutorial, not litigious: this patent and its divisional (5,960,162) both issued cleanly in 1997 and 1999 respectively, with an unusually thin four-reference citation set, and the patent was later picked up as prior art by Transact's European counsel.

---

Recommended next steps

If you are a defendant and have received a demand or complaint citing US 5,675,717:

1. Lead with expiration, not invalidity. Anticipated expiration 2015-01-25. Any infringement theory must target pre-2015 conduct. Under § 286 the recovery window is six years pre-filing — which for a 2026 suit reaches back only to 2020, after the patent died. Say so in the first responsive letter. Ask opposing counsel to identify (a) the specific pre-2015 accused acts and (b) the damages model that survives § 286. Most demand letters citing this patent will not have an answer.
2. Demand the claim-by-claim theory. Force them to state which of claims 1–16 they assert. Independent claim 1 requires (i) input from a host over a bidirectional line, (ii) forming an image on a recording material, (iii) generating color-preview-image information, and (iv) transmitting it back to the host. That "the printer generates the preview and returns it" architecture is the whole inventive point. Many modern products do preview on the host — which is the prior art the patent criticizes — and therefore fall outside claim 1. Independent claim 9 adds an express resolution-reduction step; a product that sends a full-resolution preview does not practice claim 9.
3. Do not cite an FWD — there isn't one. If anyone hands you a "PTAB invalidated this patent" representation, check the number. They are almost certainly talking about US 8,648,717 (M2M Solutions) or US 10,257,717 (SmartSky). Both are different patents. Correct the record immediately.
4. If you do petition, file on the Rudolf Hell and Kodak references. DE 3347049 C2 (monitor simulation of multicolor printing) and US 5,081,529 (printer color/tone calibration) are the two most promising starting points, neither of which has ever been in front of the Board on this patent.
5. Verify freshness before relying on this memo. ODP ingest lags and web search is not a complete negative. Re-run the number against PTAB E2E, the PTAB Decisions page, and CourtListener immediately before you file anything. A newly-filed petition would change the § 315(e)(2) analysis only for the petitioner that filed it — but it would also be an early signal that someone believes there is still recoverable value here, which would be worth understanding before you set your own posture.

If you are the patent owner (or evaluating acquisition): the absence of PTAB activity is genuinely favorable, but it reflects the patent's age and expiration, not the strength of claims 1–16. There is no live enforcement runway — the term ended 2015-01-25 — so the remaining value is defensive/portfolio rather than licensing.

Honest confidence statement: I am highly confident there is no PTAB proceeding on US 5,675,717, based on the ODP block plus four independent search vectors. I cannot prove a universal negative from search alone, and I found no CAFC opinion, district court decision, or § 285 fee order arising from this patent. Treat "no litigation and no PTAB activity" as high-confidence but not exhaustive.

I'll verify the assignment record before reconstructing the chain.

Let me check for any post-issuance transfer and confirm the certificate of correction details.

Assignment & Ownership Analysis — US Patent 5,675,717

Verification sources consulted: Google Patents legal-events record for US5675717A (authoritative full text supplied), Unified Patents portal (https://portal.unifiedpatents.com/patents/patent/US-5675717-A), FreePatentsOnline record, and USPTO Assignment Center / Patent Assignment Search (https://assignmentcenter.uspto.gov/, https://assignment.uspto.gov/patent/index.html). Web-search access to the Assignment Center UI is indirect; findings below rest on the recorded-assignment data surfaced in the patent record's legal-events table plus family cross-checks.

---

Inventors

Pattern check — no anomaly. A single inventor who assigned to his employer before filing (execution date 1995-01-19, i.e., six days before the 1995-01-25 filing date) is the textbook corporate-capture pattern for a Japanese operating company. There is no evidence of inventor departure, no co-inventor exodus, and no post-filing inventor-side assignment activity. The "all inventors depart within 12 months" fire-sale precursor signal is not present.

Note the related JP priority application JP 6-316563 (filed 1994-12-20) matured as JP 3491998 B2, and the same family produced US 5,960,162 A (divisional of US 08/378,371, filed 1997-05-20). Whether the same inventor is named on the divisional is a separate patent and is outside this record.

---

Original assignee

Canon Kabushiki Kaisha (Canon Inc.) — the entity named on the issued patent and, per the record, the current assignee.

• Products embodying the claims: Yes, in the ordinary commercial sense. The specification describes a color printer controller that generates preview image data and ships it back to a host PC over a bidirectional Centronix interface; the disclosed embodiments (electrophotographic, ink-jet, and film-boiling droplet-discharge heads) map directly onto Canon's commercial BJ/LBP color printer lines of the mid-1990s. This is a genuine product-side patent, not a paper asset.
• Primary line of business: Imaging and optical products — printers, copiers, cameras, semiconductor lithography equipment, medical imaging.
• Current status: Operating. Publicly traded (TSE: 7751); no bankruptcy, no dissolution, no restructuring event touching this patent. The patent itself is Expired – Lifetime, anticipated expiration 2015-01-25.

---

Assignment timeline

One (1) recorded assignment exists. There are no post-issuance ownership transfers.

• 1995-01-19 (executed) / recorded 1995-01-25 — Reel 007351 / Frame 0182
  • Conveyance: Assignment
  • Assignor: Yamamoto, Tsuyoshi (individual inventor)
  • Assignee: CANON KABUSHIKI KAISHA, Japan
  • Correspondent: Not surfaced in the sources I could access. The Google Patents legal-events table records the assignor, assignee, reel/frame, and effective date, but does not expose the attorney/agent correspondent of record. I could not retrieve the PTO-1595 cover-sheet correspondent field through available search. I am not fabricating a name here. For completeness: the JP-family representative of record on Canon's parallel filings of this era was the firm Beresford & Co. (Keith Denis Lewis et al.), but that is an EPO/JP prosecution representative and is not evidence of the US assignment correspondent — do not treat it as such.
  • Context: Original corporate capture — inventor-to-employer assignment executed pre-filing, standard employment-obligation transfer. Not a fire-sale, not a reorg, not a securitization.

Non-ownership recordal (not an assignment, recorded here for completeness): 1998-04-14 — Certificate of Correction (legal-event code CC). A certificate of correction corrects clerical/typographical error in the printed patent; it does not change title. It is not an assignment link.

Maintenance-fee events (title-neutral): 1999-12-03 (FEPP, payor-number assignment), 2001-03-15 (FPAY, 4-yr), 2005-03-09 (FPAY, 8-yr), 2009-03-11 (FPAY, 12-yr). No 3.5-year payment appears in the record after 2009, consistent with the stated 2015-01-25 anticipated expiration and the "Expired – Lifetime" status.

Summary

Chain terminates at the original assignee. There is no second link.

---

Timeline diagram

timeline
    title Ownership of US 5675717
    1994 : JP priority applications filed
    1995 : Inventor Yamamoto assigns to Canon KK
         : US application 08378371 filed
    1997 : US 5675717 issued to Canon KK
    1998 : Certificate of correction recorded
    2009 : Final maintenance fee payment
    2015 : Patent expired lifetime

---

NPE / troll-pattern signals

1. Shell-entity transfer — Not present. No assignee other than Canon Kabushiki Kaisha appears anywhere in the record. There is no "IP / Patents / Licensing / Holdings / Ventures" successor, no registered-agent address, no single-purpose LLC. Reel 007351/0182 is the only assignment and it runs toward an operating company, not away from one.

2. Known asserter in the chain — Not present. Neither Canon Kabushiki Kaisha nor Tsuyoshi Yamamoto matches Acacia, Marathon, Intellectual Ventures, IPNav, Wi-LAN, Mosaid/Conversant, Vringo, Pendrell, Innovatio, MPHJ, Lumen View, Round Rock, Document Generation Corp, or any Spangenberg-affiliated entity. Unified Patents' portal page for US-5675717-A lists the Parent Company as "Canon Inc" with no downstream asserter.

3. Repeat correspondent across the chain — Unclear / not assessable. A single-link chain cannot exhibit recurrence by definition. More importantly, the correspondent field was not retrievable from the sources available (see above). I will not infer a correspondent. The signal is inapplicable rather than negative.

4. Cascading transfers — Not present. Zero transfers post-issuance; nothing to cascade. No chained LLCs, no shared correspondent addresses, no common principals.

5. Pre-litigation transfer — Not present. Consistent with the prior section's litigation check: no district court case, ITC action, PTAB proceeding, or Federal Circuit appeal naming US 5,675,717 was found. There is no suit for a transfer to have preceded.

6. Bankruptcy fire-sale — Not present. Canon has never filed Chapter 7/11 in connection with this asset; no Kodak/Nortel/Polaroid-style sale event touches this patent. The patent simply ran to its natural 2015 term.

7. Privateering — Not present. No transfer to a third-party NPE asserting on Canon's behalf. No SEC 10-K/8-K disclosure or EFF/Patent Progress coverage ties this patent to a proxy assertion campaign.

8. Defensive aggregator (anti-NPE) — Not present. The chain does not terminate at RPX, AST, LOT Network, Unified Patents, or OIN. It terminates at Canon. (Unified Patents indexes the patent, which is not the same as owning it — do not conflate the portal listing with a defensive-aggregation assignment.)

---

Verdict

Insufficient data (only the original assignment).

Justification: the complete recorded chain for US 5,675,717 is a single pre-filing inventor assignment to Canon Kabushiki Kaisha, executed 1995-01-19 and recorded 1995-01-25 at Reel 007351 / Frame 0182; the only subsequent title-adjacent recordal is a 1998-04-14 Certificate of Correction, which does not convey ownership. Every NPE signal that can be evaluated is not present — there is no shell transfer, no asserter in the chain, no cascading LLC chain, no pre-litigation transfer, and no defensive-aggregator terminus. Substantively this is a clean, single-owner, operating-company patent that expired on its own terms; the "insufficient data" label reflects the absence of any multi-link chain to analyze, not a suspicion of concealed ownership.

Discrepancy to carry forward: the Unified Patents portal lists priority 1994-01-30, application date 1995-01-24, and grant 1997-10-06 — each one day earlier than the Google Patents values (1994-01-31 / 1995-01-25 / 1997-10-07). Per the source hierarchy, the Google Patents dates are authoritative. This is the same one-day offset already flagged in the prior summary section; it does not affect the assignment analysis, since the only executed/recorded dates that matter (1995-01-19 / 1995-01-25) come from the assignment record itself and are not in conflict.

Verify at: https://assignmentcenter.uspto.gov/ or https://assignment.uspto.gov/patent/index.html — search patent number 5675717.

I'll start by searching for this specific patent and its cited prior art.

Now let me look at the other cited references and check the JP/DE family citations.

Now the remaining cited references — US5276511, DE3347049C2, and JPS5438921B2.

Scope of this search

I queried for the specific number 5675717 (US 5,675,717 A, "Image processing apparatus and method for previewing a color image generated by a color printer and displayed at a host computer," Canon K.K., app. 08/378,371). I did not retrieve results for neighboring numbers (5,675,716 / 5,675,718 / 5,675,731 / JP 5675717 B2 — the last of which is an unrelated 2015 Japanese patent on an air-flow sensor that surfaced in search and is not the same document). The reference set below is taken from the front-page citation records of US 5,675,717 as reported by Google Patents (the authoritative full text supplied to me) and cross-checked against Unified Patents, Justia, EveryPatent, and the EPO publication server.

Source: https://patents.google.com/patent/[US5675717A](/patent/US5675717A) ; https://portal.unifiedpatents.com/patents/patent/US-5675717-A

---

The complete citation set on the face of US 5,675,717

The record separates citations into two groups, which matters for the §102 analysis:

Legal framework. The application was filed 25 January 1995, so pre‑AIA 35 U.S.C. §102 governs. Every one of the four US references published more than one year before the 25 Jan 1995 filing date (earliest being 14 Jan 1992, latest being 4 Jan 1994), so each is at minimum §102(b) prior art as a printed publication/patent. The DE and JP family cites are likewise §102(b) printed publications (1986 and 1979).

Anticipation test applied. Under §102 a reference anticipates a claim only if it discloses every limitation, arranged as in the claim. Two independent claims exist — claim 1 (apparatus: input color-image info from a host computer over a bidirectional line → form image on recording material → generate color-preview-image information in the printer → transmit that preview information back to the host over the bidirectional line) and claim 9 (method: output color image data to image-forming means → generate color-preview-image information → reduce its resolution → transmit the reduced-resolution preview to a host computer that forwards it to the monitor). Dependent claims 2–8 and 10–16 do no more than add features onto those two bases.

---

Reference-by-reference §102 analysis

1. US 5,146,323 A — Hitachi, Ltd.

Full citation: US 5,146,323 A, "Signal processing circuit including a white balance adjusting circuit in a color video printer apparatus," Hitachi, Ltd.; priority 3 Oct 1988; issued 8 Sept 1992. Inventors associated with the Hitachi record: Kentaro Hamma, Yasunori Kobori, Yoshiaki Mochimaru. URL: https://patents.google.com/patent/US5146323

Description. A color video printer that A/D-converts input video (NTSC or Y/C), stores a still frame in image memory, D/A-converts it, converts to RGB, applies white‑balance adjustment (manual or automatic, optionally skin-color based) to the RGB signals, and then color-prints. Notably, the same processed signal path feeds a monitor television through an encoder/matrix so the operator can observe the still image while adjusting white balance; overlay signal insert circuitry is also disclosed. Claim 1 of that patent is directed to first A/D means, frame memory, D/A means, RGB converting means, white-balance adjusting means, second A/D means, and printing means. Corresponding EP 0202536 (Hitachi, "Signal processing circuit for a color video printer").

§102 relevance to 5,675,717: Does not anticipate any claim. It discloses elements of claim 1's second and third prongs in a generic sense — the printer performs color processing (white balance) on image data and there is a monitor display of the printer-bound image — but it is missing all of the distinguishing limitations:

• no host computer and no bidirectional communication line (input is a video signal, not digital color-image information from a host over a bidirectional line);
• no generation of color-preview-image information by the printer for reproducing the image it will form on a monitor;
• no transmission of preview image information back to a host;
• no resolution reduction step (claim 9).

It is at most §103 background showing that a printer-side controller may use a monitor to view the image it will print. Verify against the actual Hitachi disclosure before relying on it either way.

2. US 5,210,600 A — Fuji Xerox Co., Ltd.

Full citation: US 5,210,600 A, "Extraction of film image parameters in image processing apparatus," Fuji Xerox Co., Ltd.; priority 8 Jan 1990; issued 11 May 1993. URL: https://patents.google.com/patent/US5210600

Description. Concerns a digital color copier with a film projector input. During a pre-scan, color-decomposition data at sampling points are summed/averaged line-by-line (with shading correction), and from those averages the apparatus computes a density correction amount and image-processing parameters that are set in a line memory for the subsequent main scan. The patent is about automatically deriving color/density correction parameters for film originals.

§102 relevance to 5,675,717: Does not anticipate any claim. This is a color-correction-parameter extraction reference with no preview function at all — no host computer, no bidirectional line, no preview-image generation, no preview transmission, no resolution reduction. It is, at best, §103 art for the general proposition that a printer/copier derives its own color-processing parameters, which is expressly acknowledged in the 5,675,717 background section ("The color printer, in some cases, generates the color correction parameters").

3. US 5,081,529 A — Eastman Kodak Co. (the closest of the four)

Full citation: US 5,081,529 A, "Color and tone scale calibration system for a printer using electronically-generated input images," Eastman Kodak Co. (inventor Robert P. Collette, Pavilion, NY); filed/app. no. 629,516, 18 Dec 1990; issued 14 Jan 1992. Family: EP 0 515 665 B1 (granted 6 Mar 1996). URLs: http://www.everypatent.com/comp/pat5081529.html ; https://patents.google.com/patent/US5081529

Description. A calibration utility operates on stored digital image data from an input memory. Until the print matches the monitor, a printer tone-scale algorithm and printer color algorithm compute correction terms that are written into the printer's tone-scale LUT and color matrix; simultaneously, inverse corrections are applied to the monitor's LUT/color matrix so the operator gets interactive feedback ("drive the monitor to look as bad as the print"). When the monitor looks like the print, the monitor reverts and the printer is calibrated. Claim 1 is directed to printer calibration means + display calibration means + commonly-controlling interactive means producing a monitor adjustment substantially inverse to the printer adjustment.

Why it is the closest art but still not anticipatory. It shares with 5,675,717 the core problem (a printed image differs in color/tone from the monitor image) and the core technique (colorimetric matching between a printer's output characteristics and a monitor's display characteristics). It also shows the host side sending color-conversion/correction information to the printer — relevant to the subject matter of dependent claims 6 and 14.

However, it is missing each of the limitations that defines the independent claims of 5,675,717:

• the printer generates the preview-image information — in Kodak, the calibration utility on the host side computes both the printer corrections and the monitor corrections; the printer stores and applies a LUT/matrix, it does not generate an image for display;
• transmission of preview-image information from the printer back to a host over a bidirectional line — data flow in Kodak is host → printer (calibration values and image data), never printer → host (preview image data);
• resolution reduction of preview data (claim 9).

Verdict: no §102 anticipation of claims 1–16. Potentially citable under §103 in combination with a reference showing printer→host return of image data, and it is the best of the four for arguing the colorimetric-matching concepts that appear in dependent claims 4/6/7 and 12/14/15.

4. US 5,276,511 A — Fuji Photo Film Co., Ltd.

Full citation: US 5,276,511 A, "Method of and apparatus for processing image by setting up image processing conditions on the basis of finishing information," Fuji Photo Film Co., Ltd.; priority 21 Feb 1991; issued 4 Jan 1994. Family: EP 0 502 369 A3 / B1. URL: https://patentimages.storage.googleapis.com/d7/ba/3f/a90e40a6695009/US5276511.pdf

Description. In a scanning/platemaking system, an image is displayed on a CRT display unit and the operator sets "finishing information" (e.g., "fine skin," "beautiful sky," "make uniform gray"). Hue-space data corresponding to that finishing information are determined, particular spots and main subjects are extracted, weights are computed, and image-processing conditions (Y/M/C correction values) are set so that the selected hue region becomes the optimum hue value, including skin-color handling and preferential weighting of the main subject. A related sibling, US 5,475,509, continues the same finishing-parameter approach.

§102 relevance to 5,675,717: Does not anticipate any claim. It has no printer-side preview generation, no host-computer bidirectional transmission, no return of preview image data, and no resolution reduction. Its only substantive overlap is with the dependent claims that recite preferred color matching / accurate reproduction of an important or dominant color such as skin color — i.e., claim 4 and claim 12 ("performs preferred color matching for the natural image information"). Because claims 4 and 12 depend from claims 1 and 9 respectively, and because US 5,276,511 cannot satisfy those independent claims, it does not anticipate claims 4/12 outright; it is §103 material for the "preferred/importance-weighted color reproduction" limitation.

5. DE 3347049 C2 — Dr.-Ing. Rudolf Hell GmbH (family cite; conceptually the most on-point preview art)

Full citation: DE 3347049 C2, "Method and circuit arrangement for simulating multicolor printing on a color monitor" (Verfahren und Schaltungsanordnung zur Simulation eines Mehrfarbendrucks auf einem Farbmonitor); applicant Dr.-Ing. Rudolf Hell GmbH, 2300 Kiel; priority 24 Dec 1983; A1 published 11 July 1985; C2 granted/published 17 July 1986. (Recorded in the Google Patents front page as a family citation, i.e., cited in the JP counterpart rather than in the US prosecution; the same document also appears as a citation on the Eastman Kodak US 5,081,529 page.)

Description. As the title states, a method and circuit for simulating a multicolor print on a color monitor — i.e., electronic soft-proofing. This is the closest conceptual ancestor of "preview," and it is a §102(b) printed publication (1985/1986, well over one year before the 1995 filing).

§102 relevance to 5,675,717: Does not anticipate any claim of 5,675,717. It is a prepress reproduction-apparatus disclosure; it does not show an image forming apparatus that prints onto a recording material in response to color-image information received from a host computer over a bidirectional communication line and that generates and returns preview-image information to that host. It also has no resolution-reduction-and-return architecture. Its proper role is as §103 art establishing that monitor simulation of a printed image was known long before the 1994 priority date — which undercuts any attempt to read the result (a monitor image resembling the print) as novel, leaving the architectural limitations (printer-generated preview returned to the host) as the point of novelty. That is consistent with the patent's own framing in its "Description of the Related Art."

6. JP S54‑38921 B2 (family cite — unresolved)

Full citation as recorded: JP S54‑38921 B2; priority date listed 31 May 1974; publication 24 Nov 1979; no title, no assignee, no inventor in the Google Patents "Family Cites Families" entry for US 5,675,717.

Honest status: I could not retrieve the subject matter of this document, and I will not guess at it. One caveat worth recording: the same numeric string in the kokai (A) series, JPS5438921A, is indexed elsewhere as a 1979 publication titled "Production of high density yarn" (Keigo Matsumoto), i.e., a textile case — which is either a database linking artifact or an indication that the B2 entry is being mis-associated. Do not treat JP S54‑38921 B2 as color-image-processing prior art until the actual JP‑B publication (or its JPO/Espacenet abstract in English or Japanese) is pulled and read. If it is genuinely an examined Japanese color-reproduction case, it would be high-value §102(b) art given its 1979 publication date; that must be verified, not assumed.

---

Bottom line on §102

No reference cited on the face of US 5,675,717 anticipates any of its 16 claims under 35 U.S.C. §102, singly or as the examiner applied them — consistent with the fact that the application issued on 7 Oct 1997 with all claims intact after the examiner considered these four US references.

• US 5,081,529 (Kodak) is the closest citation: it addresses the same monitor-vs-print matching problem and even sends correction parameters host→printer, but it never has the printer generate preview-image information, never transmits preview data printer→host, and has no resolution reduction. Missing at least three limitations of claim 1 and two of claim 9.
• DE 3347049 C2 (Hell) is the closest preview-concept citation (simulating multicolor print on a color monitor) but is a prepress simulation disclosure lacking the host/bidirectional/return-path architecture.
• US 5,146,323 (Hitachi) supplies only the notions of printer-side color processing and monitor observation of the printer-bound image.
• US 5,210,600 (Fuji Xerox) and US 5,276,511 (Fuji Photo Film) are color-correction-parameter references; the latter touches the preferred/skin-color matching limitation of dependent claims 4 and 12 but lacks every limitation of independent claims 1 and 9.
• JP S54‑38921 B2 is unresolved; do not rely on it without reading the document.

The claims that these references can realistically be used against are therefore the dependent ones, and only in a §103 combination — most plausibly claims 4 / 12 (preferred vs. color-difference-minimizing matching) on US 5,276,511 plus US 5,081,529, and claims 6 / 14 (host supplies color-conversion information) on US 5,081,529. The independent claims 1 and 9 stand or fall on the printer-side preview generation, the bidirectional return path to the host, and (claim 9 only) the resolution-reduction step, none of which any cited reference discloses.

---

Notes, caveats, and cross-checks against the earlier sections

Consistent with the prior summary (bibliographic data, claims structure, divisional US 5,960,162, expiry 2015): nothing here contradicts it. The one-day date offset flagged earlier (Unified Patents shows priority 1994‑01‑30, app. 1995‑01‑24, grant 1997‑10‑06 vs. Google Patents' 1994‑01‑31 / 1995‑01‑25 / 1997‑10‑07) reappears identically in my checks and is carried forward as an unresolved source discrepancy rather than reconciled.

Two caveats a reviewer must carry forward:

1. I could not run a native USPTO PatentCenter / Patent Public Search query in this session; the citation set above is the front-page record as published and mirrored by Google Patents, Unified Patents, Justia and EveryPatent. The examiner-cited set ("Patent Citations (4)") is what was considered during prosecution; the two "Family Cites" are citations made in the Japanese family and are not evidence that the US examiner considered them.
2. The dates in the citation tables are the priority dates as labeled by Google Patents (which for US 5,081,529 happens to equal its filing date, 18 Dec 1990, confirmed independently by EveryPatent). Where a reference's actual US filing date differs from its priority date, the §102(b) status is unaffected here because every publication date is more than one year before 25 Jan 1995, but a formal invalidity chart should use the filing dates.

Sources consulted: https://patents.google.com/patent/US5675717A (authoritative full text); https://portal.unifiedpatents.com/patents/patent/US-5675717-A ; https://patents.google.com/patent/US5146323 ; https://patents.google.com/patent/US5210600 and https://patentimages.storage.googleapis.com/48/77/f1/b7a2c3c1f89d05/US5210600.pdf ; https://patents.google.com/patent/US5081529 and http://www.everypatent.com/comp/pat5081529.html and http://data.epo.org/pise-server/rest/collections/lgpi/EP0515665B1.pdf ; https://patentimages.storage.googleapis.com/d7/ba/3f/a90e40a6695009/US5276511.pdf ; https://patents.google.com/patent/DE3319941A1 (for the DE 3347049 A1/C2 family record).

I'll ground this analysis in the actual cited prior art. Let me pull details on the four examiner-cited references and the family-cited art.

Now let me pull the family-cited art, especially the DE reference described as "simulating multicolor printing on a color monitor" — which looks highly relevant.

Obviousness Analysis — US 5,675,717 under 35 U.S.C. § 103 (pre‑AIA)

Note on scope of this section: This builds on the earlier bibliographic, claim‑scope and litigation sections rather than repeating them. Here I work only from the prior‑art record supplied on the patent page — the four examiner‑cited references and the two family‑cited references — and treat the "Cited By" documents as what they are (post‑priority art that cannot be § 102/§ 103 art against this patent). No prosecution file, office action, or examiner's statement of reasons for allowance was available to me, so everything below is a reconstructed, hypothetical § 103 analysis, not a reporting of what the examiner actually did. I flag every inference as such.

---

1. The prior-art universe actually available (critical date: 1994‑01‑31)

Pre‑AIA § 102/103 governs (filing 1995‑01‑25; earliest priority JP 6‑009563, 1994‑01‑31). The controlling date for a § 102(b) printed publication is therefore on or before 1994‑01‑31; § 102(a)/§ 102(e) art needs a date before the invention, and anything published less than a year before the US filing is at best § 102(a)/(e) art.

Immediate observation: the four examiner‑cited references are all color‑reproduction art (calibration, white balance, scan‑parameter extraction, finishing‑information‑based hue control). Only one of them (US 5,146,323) involves a printer with a monitor path at all, and only DE 3347049 C2/EP 0 147 813/US 4,639,770 squarely discloses monitor simulation of the printed result before printing — i.e., the "preview" concept itself.

---

2. POSITA and the state of the art as of 1994‑01‑31

A POSITA would be a software/hardware engineer with ~2–3 years' experience in color image reproduction and printer controllers, familiar with: RGB↔CMY(K) color conversion via look‑up tables and matrices (US 5,081,529; US 5,210,600); binary‑coding of contone data (error diffusion/dither — discussed in the '717 spec itself as conventional); page‑description‑language printer controllers and bidirectional printer ports (the '717 claim 1 and all four embodiments use a "bidirectional Centronix interface"); and the use of a host‑resident preview in desktop publishing that simulated printer color processing. The '717 background section itself concedes this: color‑correction parameters were "determined in advance by a printer driver or an application program," and "the color printer, in some cases, generates the color correction parameters." That concession is important — it means printer‑side color processing is admitted prior art; the fight is over where the preview data is generated and which way it travels.

---

3. Element-by-element mapping of independent claim 1

Claim 1: (a) input means for inputting color‑image information from a host computer via a bidirectional communication line; (b) image forming means forming an image onto a recording material; (c) generation means for generating color‑preview‑image information for reproducing the formed image on a color monitor; (d) transmitting means for transmitting the color‑preview‑image information from the image forming apparatus to the host computer via the bidirectional communication line.

Net: claim 1 is not anticipated by, and is not rendered obvious on the face of, any single reference in the record. The two‑way, printer‑originated preview channel is the crux, and no cited reference discloses it, alone or in combination.

---

4. The § 103 grounds a challenger would actually plead

I rank these from strongest to weakest, with the motivation rationales a KSR‑style petition would use.

Ground A (strongest for the dependent claims; the practical attack)

US 5,081,529 in view of US 5,146,323, further in view of US 5,276,511.

• What each supplies: US 5,081,529 supplies printer‑resident color/tone‑scale calibration tables (tone‑scale LUT 36, color matrix 48 in the printer), the notion of generating a monitor image that reproduces the printed result, and host→printer transmission of adjustment/conversion data. US 5,146,323 supplies a printer that already has a monitor output path in addition to its print path, plus per‑channel gain-based color adjustment performed inside the printer before printing, and expressly contemplates the operator viewing the image on a monitor television while adjusting. US 5,276,511 supplies displaying image data on a CRT display unit and setting image‑processing conditions so that image data in a selected hue region (skin color/memory color) is treated as the optimum hue.
• Mapping to dependents: claim 2/10 (color processing per image type) and claim 4/12 ("preferred color matching," skin color for natural images vs. color‑difference‑minimizing matching for CG) map onto US 5,276,511's "fine skin"/memory‑color hue‑space selection; claim 5/13 (determination whether preview is to be performed) maps onto US 5,081,529's operator‑controlled selection between the application‑software path (switches 21a/21b) and the calibration utility; claim 6/14 (host transmits color‑conversion information) maps onto US 5,081,529's operator‑derived correction terms sent to the printer; claim 7/15 (storage means for color‑conversion information) maps onto US 5,081,529's tone‑scale LUT 36/color matrix 48 resident in printer 14; claim 8/16 (image data corresponding to the number of gradation steps of the image forming means) maps onto US 5,081,529's tone‑scale look‑up table (gradation mapping) and US 5,210,600's binarization of gradation toner signals.
• Motivation to combine: same field (electronic color reproduction), same problem (make the monitor image correspond to the image the printer will produce), and complementary disclosures — US 5,146,323 establishes that a printing device can feed a monitor, and US 5,081,529 establishes what must be computed to make that monitor feed color‑faithful to the print. Under KSR, "a known technique used to improve one device" applied to a similar device in the same field with predictable results is obvious. Rational design incentive as of 1994: off‑load host CPU from preview simulation (a concern stated in the '717 background).
• Result: claims 2–8 as dependent claims are vulnerable; claim 4/12 in particular is closely tracked by US 5,276,511.

Ground B (the only real attack on independent claim 1)

DE 3347049 C2 (≡ US 4,639,770) in view of US 5,146,323.

• DE 3347049/EP 0 147 813/US 4,639,770 supplies everything about how to generate the preview: call up the print's color‑separation values, modify them according to the gradation curve that the later printing process will actually use, map them through measured Y,M,C→r,g,b allocations, combine with K′, and drive the monitor so the monitor image is color‑equivalent to the anticipated print — i.e., a faithful "color‑preview‑image information" generator, and one that is explicitly concerned with the print‑simulation problem.
• US 5,146,323 supplies a printer that emits image signals to a monitor as well as to its print engine, and performs its color adjustment in‑printer.
• Alleged motivation: put the simulation engine of DE 3347049 into the printer that already has the monitor output of US 5,146,323, and send that data to the host that owns the display and the job — an arguable "finite number of identified, predictable solutions" for decentralizing preview, especially given the admitted practice of printer‑side color processing.
• Why this is still weak: DE 3347049's simulation engine sits in a dedicated proofing/scanning console that has no host computer and no bidirectional printer link at all; US 5,146,323's monitor signal is the print‑bound image itself, not a separately generated simulation of the print, and it is fed to a television, not back to a host. Combining them requires an added step — an outward‑facing preview channel on the printer's bidirectional port — that no reference in the record teaches or suggests. A petition would be arguing "it would have been obvious to transmit," which is close to relying on the inventor's own insight that this data should be sent back to the host. Absent art showing a peripheral generating a rendition/preview and returning it to a host (a client‑server or PDL‑printer‑renders‑preview teaching), this ground would likely draw a "missing element not supplied by the art" finding.

Ground C (attack on claim 9's resolution limitation — the strongest single‑limitation attack)

Any of US 5,081,529 / DE 3347049 in view of the ordinary knowledge that monitor resolution is lower than printer resolution, with US 5,276,511 and US 5,210,600 as secondary.

• Claim 9 requires generating preview information, reducing its resolution, and transmitting the reduced‑resolution data to the host for display. The specification justifies the step purely as a transfer‑size/speed economy ("reduce the amount of preview image data transmitted ... and thereby shorten the transfer time"), and the specification's own words — "In general, the resolution of the monitor is lower than that of the printer" — supply the predictability: down‑sampling print‑resolution data to monitor resolution is the natural, expected implementation of a preview‑display path, not a new result.
• Supporting art: US 5,210,600 performs a pre‑scan at sampling points and computes parameters before the main scan, then drives a CRT‑based flow — i.e., generating intermediate, reduced/sampled image data for evaluation before the final output. US 5,276,511 displays the pre‑scan image on the CRT display unit. DE 3347049's color viewing apparatus includes an image repetition memory and address controller for reading out the stored print‑separation values to drive a raster monitor, which differs in address rate from the print raster. I have not verified that any of these expressly describes down‑converting already‑generated preview data from 360 dpi to 90 dpi, so I would not represent that they disclose the limitation expressly.
• Assessment: the resolution‑reduction step is a strong candidate for a "mere design choice / predictable variation with predictable result" (§ 103; KSR) argument once the underlying preview‑generation concept is supplied. But because claim 9 is an independent claim that also requires the printer‑side generation and the transmission to the host, Ground C does not independently dispose of claim 9 unless the § 103 defect in the base concept is cured.

Ground D (claim 8/16 only — gradation steps)

US 5,081,529 (tone‑scale LUT; gradation mapping) or US 5,210,600 (RGB→YMC(K) then binarized gradation toner signals). Claim 8/16 requires the generation means to "obtain image data corresponding to a number of gradation steps of the image forming means." Both references concern exactly the number of reproduction levels of the output device. This is the most vulnerable limitation in the patent.

---

5. Why a § 103 rejection of independent claims 1 and 9 would likely fail on this record

1. Directionality of the channel is unmatched. Every record reference that touches monitor/preview data moves information host/console → printer (US 5,081,529 correction terms; the '717's admitted printer‑driver parameters) or printer → television (US 5,146,323). Nothing returns printer‑generated preview image data to a host computer over a bidirectional line. Claim 1's "transmitting means for transmitting the color‑preview‑image information from the image forming apparatus to the host computer" is the whole inventive contribution; it is also the one element absent from the art.
2. US 5,081,529 arguably points the other way. There the monitor corrections are made "substantially the inverse" of the printer corrections, and the operator's goal is to drive the monitor to look like the (bad) print — the reverse of the '717 architecture, in which the printer computes and supplies the monitor‑facing data. That is not classic "teaching away" (it does not disparage the '717 approach), but it is the opposite arrangement of function, which weakens any "the art suggests this specific allocation" argument.
3. The record supplies no motivation for the specific repartitioning. The KSR motivations a challenger would invoke (decongest the host CPU, avoid simulating the printer's PDL interpreter and binary coding in the host, reuse the printer controller that already does color processing and halftoning) are all plausible — but the record's references are silent on host CPU load and on PDL‑level preview. The '717 background section supplies the problem statement, and a reference cannot be assumed to know the inventor's problem merely because the specification states it.
4. No supporting secondary evidence either way. As the earlier litigation section found, I located no assertion, licensing, or challenge to this patent, and it expired 2015‑01‑25. So there is no objective‑indicia record (no nexus evidence of commercial success, no copying, no industry praise) to weigh in either direction. I state that as a high‑confidence negative finding about publicly indexed litigation, not as proof of absence.
5. Procedural inference (flagged as inference only). The four cited references are all pre‑1994 color‑reproduction documents, and the patent issued without the record showing any post‑allowance reexamination, reissue, or IPR. Consistent with that, the allowance most likely turned on the printer‑originated, printer→host preview channel. I do not have the file wrapper, so I cannot confirm this; it remains an inference from the citation set. (A certificate of correction was recorded 1998‑04‑14; the record does not show its content, so I cannot say whether it touched claim scope.)

---

6. Bottom line

The decisive single point: the patent's novelty as drafted lives in architecture — the image forming apparatus itself produces the color‑preview image information and returns it to the host over the bidirectional interface — not in any color‑processing mathematics. On the prior‑art record actually on this page, the mathematics is thoroughly old (Helmuth Hell's print simulation, Kodak's printer/monitor calibration, Fuji's memory‑color finishing) and the one element nobody in the record supplies is the printer‑originated, printer‑to‑host preview channel (with claim 9's resolution reduction being an obvious economy once that channel exists). A § 103 rejection of the dependents is readily constructible; a § 103 rejection of claims 1 and 9 would require a reference — not present here — in which a printer or peripheral renders a simulation of its own output and transmits that simulation back to a requesting host.

---

Flags and caveats for the reviewer

• I could not retrieve JPS 5438921 B2. The patent page gives it no title or assignee, and my single search attempt hit the tool's step limit. I have therefore excluded it from every combination above rather than speculate. If its content is relevant (it is only disclosed on the page as "cited by the examiner during prosecution" family art), the analysis should be revisited.
• Date inconsistency carried forward. The two earlier sections are dated April 26, 2026, yet contain evidence dated September 25–28, 2026, and this task states the current date as April 26, 2026, while the system context says 2026‑10‑01. None of that affects a § 103 analysis of a patent that expired 2015‑01‑25, but the "as of" horizon should be fixed before the memo is relied upon.
• DE 3347049 C2 is identified on the patent page only by number and title. I treated US 4,639,770 A and EP 0 147 813 A2/B1 as its family equivalents because the DE number appears as the priority in those records; if the reviewer needs a different member, the disclosure is materially the same (print‑simulation color viewing apparatus).
• No file wrapper, no office action, no examiner's statement. Everything in § 3–§ 6 is a reconstructed analysis. I have not verified that the examiner applied US 5,081,529 or US 5,146,323 against claims 2–8; those references are listed as citations, and citation is not application.

Sources: US 5,675,717 (full text, claims, citation lists); US 5,081,529 A and EP 0 515 665 B1; US 5,146,323 A; US 5,210,600 A; US 5,276,511 A; US 4,639,770 A / DE 3347049 C2 and EP 0 147 813 A2.