US 9835492

I'll verify the patent details against live sources, including any litigation/docket activity.

Let me pull the issued claim set to characterize the independent claims accurately.

Let me try to retrieve the verbatim issued claims and check assignee/litigation records directly.

US 9,835,492 B2 — Verification Summary

Search notes upfront (important caveats):

• No CAFC 2026 docket activity found. A targeted search for "9835492" + CAFC/2026 dockets returned no Federal Circuit or district-court litigation involving this patent. The only hit resembling the string was an unrelated WO 98/35492 A2 ("Method and apparatus for recognising video sequences") — a 1998 PCT publication, not the same identifier and not related. I did not find any evidence this patent has been litigated.
• Cannot confirm PTAB activity from the sources retrieved. Treat as "no litigation found" rather than "none exists."
• Verbatim issued claims not retrieved. The claim set itself was not returned by the live sources I could reach. The independent-claim overview below is built from the patent's own specification text (which is drafted in application-example/claim-style language) plus a Google Patents snippet of claim language. Claim count and exact wording should be confirmed against the USPTO PatentCenter / issued PDF.

Bibliographic Data

Abstract (as published)

A spectroscopic camera includes a wavelength variable interference filter having a pair of reflection films and an electrostatic actuator for changing the gap between them; an imaging unit with multiple imaging pixels; and a filter control unit controlling the actuator based on "light-received wavelength data" in which a control amount for receiving light of a target wavelength is recorded for each imaging pixel. An acquisition order of the target wavelength emitted from the filter is set for each imaging pixel, and the acquisition order is set in a descending order of the driving amount of the gap, with the value for emitting the target wavelength in that order recorded as the light-received wavelength data.

⚠️ Literal-text discrepancy worth flagging: the abstract says the order is set in a "descending order of the driving amount." The Summary/Description repeatedly says "in order of changing the driving amount… in a driving area in one direction (for example, in order of small driving amount)" and an embodiment says the data are listed "in an ascending order," with the gap made smaller than the initial gap. These are not identical statements. Per your instruction I am reading the identifiers/text literally and flagging rather than reconciling.

Independent Claims — Plain-Language Overview

Based on the application-example language in the specification (which tracks the claim structure) and the retrieved snippet:

1. Apparatus claim (core) — "spectroscopic image acquiring apparatus"
A spectroscopic camera/apparatus comprising:

• a wavelength variable interference filter with a pair of facing reflection films plus a gap changing unit (electrostatic actuator), the filter emitting light at wavelengths corresponding to multiple orders (i.e., 1st-order and 2nd-order transmission peaks);
• an imaging unit with many imaging pixels that receive the filter's emitted light; and
• a filter control unit that drives the gap changing unit based on "light-received wavelength data."

The data itself is the point: it contains, per imaging pixel, (a) an acquisition order for when the target wavelength is emitted from the filter, and (b) the control amount needed to emit that target wavelength in that order — and the acquisition order is arranged monotonically in one direction (i.e., the gap is swept steadily in a single direction rather than jumping back and forth).

In plain terms: you build a lookup table telling the camera "at voltage step #N, pixel A is collecting 600 nm via the 2nd-order peak while pixel B is collecting 600 nm via the 1st-order peak," then you sweep the drive voltage one way. Because each voltage step harvests data for several pixels at once, you take fewer measurements, move the membrane less, and avoid spring-ringing — while still recording the actual received wavelength per pixel.

2. Method claim — "spectroscopic image acquiring method"
The method counterpart: in an apparatus having the variable-gap interference filter (emitting multiple orders) and a multi-pixel imaging unit, the steps are (i) acquiring light-received wavelength data containing the control amount of the gap changing unit for receiving light of a predetermined target wavelength at each imaging pixel, and (ii) changing the gap size by controlling the gap changing unit based on that data — where the data again includes, per pixel, the acquisition order and the control amount for that order, with the order set as a monotonic sweep of the driving amount in one direction.

3. Second apparatus claim — "still another" formulation
An apparatus comprising a variable interference filter having a first reflection film and a second reflection film facing it plus a gap changing unit; an imaging unit receiving light from the filter; and a filter control unit driving the gap changing unit based on light-received wavelength data with the control amount for receiving a predetermined target wavelength at each of the plurality of imaging pixels — wherein the data includes, for each of the plurality of imaging pixels, the acquisition order when the filter transmits the target wavelength and the control amount for transmitting it in that order, and the gap is sequentially changed in one direction.

Dependent/Further Concepts Reflected in the Specification

• Per-pixel target wavelengths (2nd embodiment): the target wavelength need not be identical across pixels — e.g., 400–700 nm at 20 nm in pixel A versus 401–701 nm at 20 nm in pixel B.
• Pixel grouping by gap variation (3rd embodiment): pixels are grouped according to the variation in gap size along their optical path, with a target wavelength and control amount stored per group.
• Grouping unit linked to wavelength resolution: groups are formed by dividing the gap variation by a length unit less than the wavelength resolution (e.g., half the 20 nm resolution = 10 nm units), keeping the induced wavelength error below the resolution and thus negligible.
• Neighboring-pixel groups with averaged wavelength (4th embodiment): e.g., 2×2 adjacent pixels form one group, and the average of the actually received wavelengths is stored as the group's target wavelength, letting the group be replaced by a single image pixel.

Technical/Grounding Notes

• The described hardware: quartz-glass fixed and movable substrates bonded via a siloxane-based plasma-polymerization bonding film; Ag/Ag-alloy and/or TiO₂–SiO₂ dielectric multilayer reflection films; an annular electrostatic actuator (fixation electrode 561 / movable electrode 562); CCD/CMOS imaging unit; a band pass filter (400–700 nm passband in the embodiment) that shields out-of-band light so higher/lower order peaks don't contaminate readings.
• The worked example (Table 1) uses 20 control IDs (V1–V20) with the 2nd-order peak used for 400–600 nm and the 1st-order peak for 620–700 nm, illustrating how one voltage step covers different target wavelengths in pixels a/b/c simultaneously.
• Cited background art: JP-A-2009-141842 (interference-filter imaging apparatus).

Sources: https://patents.google.com/patent/US9835492/en ; https://patents.google.com/patent/US20160061661 (pre-grant publication/claims snippet); https://trade.patenthub.cn/cpc/patent-[166923](/patent/166923)-US2016061661A1-728c799414184f596934ad74d9ff2ff2.html (abstract/description mirror).

Bottom line: US 9,835,492 B2 is a Seiko Epson patent (inventor Teruyuki Nishimura) on reducing measurement time and drive excursion in a Fabry–Pérot-based spectroscopic camera by pre-storing a per-pixel acquisition order and sweeping the etalon gap monotonically in one direction. I found no CAFC 2026 docket or other litigation tied to this number, and I could not retrieve the verbatim issued claim set — the independent-claim descriptions above should be validated against the issued PDF before being relied on for anything claim-scope-specific.

I'll search for litigation involving US patent 9835492 across patent litigation databases and court sources.

Initial results are irrelevant (unrelated WO 9835492 publication). Let me search litigation-specific databases directly.

Let me search more targeted litigation databases and the patent's own litigation records.

Findings: Litigation Involving U.S. Patent No. 9,835,492

Short answer: I found no litigation (or IPR/PGR) specifically involving U.S. Patent No. 9,835,492.

Based on the searches I ran against patent-litigation resources (Unified Patents Portal, RPX/Insight, CourtListener/RECAP, PTAB/PTActs filings, Justia dockets, docketalarm), nothing identified U.S. Patent 9,835,492 as an asserted or challenged patent. There is therefore no case to report with plaintiff, defendant, jurisdiction, case number, filing date, or status.

Important caveat on search reliability

I was unable to complete a fully exhaustive set of targeted queries before hitting my search-step limit, and my web results were dominated by irrelevant hits. Notably, several hits referenced an "WO 9835492 A2" — that is a PCT international publication number (unrelated to the US patent) and must not be confused with U.S. Patent 9,835,492. Several Seiko Epson litigation results also surfaced, but those concerned entirely different patents (e.g., ink-cartridge patents such as the '422, '749, '116 patents, and the Honeywell '371 LCD patent dispute) and are not litigation over 9,835,492.

Patent context (for the record)

• Patent: U.S. 9,835,492 B2 — "Spectroscopic image acquiring apparatus and spectroscopic image acquiring method"
• Inventor: Teruyuki Nishimura
• Assignee: Seiko Epson Corporation
• Application: US 14/835,843; Priority date: 2014-08-29; Filed: 2015-08-26; Granted: 2017-12-05
• Technology: Fabry-Perot / wavelength variable interference filter with per-pixel light-received wavelength data and imaging order control (spectroscopic camera)

This is a niche component/algorithm patent in the spectroscopic-imaging space. It appears to be owned and held internally by Seiko Epson and was not, to the best of my search, put into assertion campaigns — consistent with the absence of litigation.

Recommended verification steps (since I cannot confirm a negative with certainty)

Because "no litigation found" is a negative result and not proof of none, I recommend confirming directly at:

1. Unified Patents Portal – litigation & PTAB case lists: https://portal.unifiedpatents.com/litigation (and /ptab/caselist)
2. CourtListener / RECAP (free PACER mirror): https://www.courtlistener.com — search docket field for "9,835,492"
3. USPTO PTAB (PTActs): https://ptacts.uspto.gov — search for the patent number in IPR/PGR/reexam filings
4. USITC EDIS: https://edis.usitc.gov — for any Section 337 investigations naming the patent
5. Google Patents "Litigation" tab for US9835492 and Lex Machina / Docket Navigator (paywalled but authoritative).

If you need, I can run a second, focused round of searches against CourtListener, PTActs, and the Unified Patents Portal to try to close the remaining gap.

Bottom line: No known litigation involving U.S. Patent 9,835,492 was found; I explicitly decline to fabricate a case list, and I flag the "WO 9835492" false-positive as a distinct (PCT) identifier.

PTAB / AIA Trial Landscape — US Patent 9,835,492

"Spectroscopic image acquiring apparatus and spectroscopic image acquiring method" — Seiko Epson Corp. (inventor: Teruyuki Nishimura), filed 2015-08-26, granted 2017-12-05, priority 2014-08-29, anticipated expiration 2035-08-26.

Proceedings overview

Zero AIA trial proceedings are on file against US 9,835,492 — the USPTO Open Data Portal returns no IPR, PGR, or CBM petition for this patent, and independent web searching surfaced no petition, institution decision, Final Written Decision, or Federal Circuit appeal involving it, so the breakdown is trivially 0 active / 0 claims invalidated / 0 claims sustained / 0 settled / 0 institution-denied — meaning all claims stand untested at the PTAB, and a defendant has no ready-made invalidity judgment to free-ride on; any AIA challenge would have to be built from scratch.

No proceedings to report

There is no proceeding number to list, and I will not invent one. To be explicit about what the absence means and what I could and could not verify:

• Canonical source: The structured "PTAB proceedings on file" block supplied with this task — sourced from the USPTO Open Data Portal — reports no AIA trial proceedings for US 9,835,492 as of the most recent ingest. That is the authoritative list, and it is empty.
• Corroborating search: I searched for the patent number in combination with PTAB/IPR/petition/litigation terms, for PTAB trials naming Seiko Epson on spectroscopic-filter subject matter, and for filings in the 2015–2025 window. Nothing matched. The only hits were unrelated documents that happen to share the numeric string "9835492" (e.g., a 1998 PCT publication, WO 9835492 A2) or Epson patent-family/search-report listings that merely cite neighbor patents in the same portfolio.
• Confidence: High that no IPR/PGR exists. The patent issued 2017-12-05; PGR (available only for post-AIA patents on § 112 and § 101 grounds, 9-month window) has long since lapsed, so CBM/PGR are off the table as a practical matter anyway and IPR is the only live AIA vehicle.
• One caveat worth flagging: AIA petitions can sit under a confidentiality or sealing order before institution, and third-party aggregators (Unified Patents, RPX) sometimes file without the patent owner's active involvement. I found no evidence of such a filing, but I cannot affirmatively rule out a petition filed after the ODP ingest date.

Related-but-distinct data point (not a proceeding against this patent)

Epson has been an NPE target, but not on this patent. In K.Mizra LLC v. Epson America Inc. et al, No. 8:25-cv-00027 (C.D. Cal., filed 2025-01-07), the asserted patents were US 10,018,938; US 7,570,400; and US 9,769,342 — US 9,835,492 was not among them, and the action was voluntarily dismissed on 2025-09-02. I report this only as context for the "has this patent been asserted?" question, and I flag that I am relying on a secondary aggregator summary rather than the docket itself; verify on PACER/CourtListener before relying on the dismissal's terms.

Strategic summary

Claim status: every claim is UNTESTED. Because no petition was ever filed or instituted, there is no claim of US 9,835,492 that has been canceled, none that has been sustained through an FWD, and none narrowed by certificate. Claims 1 and the other independent claims (the apparatus claim directed to the wavelength variable interference filter + imaging unit + filter control unit, and the corresponding method claim) retain the full scope they had on issuance. The specification discloses dependent/embodiment-level features — per-pixel target-wavelength assignment, pixel grouping by gap-size variation, dividing the imaging unit into groups by a length unit less than the wavelength resolution, and averaging received wavelengths over a close-packed 2×2 pixel group — that would be obvious candidates for a narrowing amendment if the patent were ever challenged, but none of that has happened.

Estoppel landscape: there is none, which cuts both ways. Because no petitioner has been through an FWD, § 315(e)(2) estoppel has never attached to anyone on this patent. That is good for a defendant in the narrow sense that no prior-art ground is precluded, but it is worse in the practical sense that there is no adjudicated invalidity finding to leverage and no PTAB claim construction (Phillips or otherwise) to bind a district court. Conversely, the patent owner is under no estoppel either and is free to argue the full breadth of the claims. The one hard statutory clock to keep in mind is § 315(b): if you are served with a complaint alleging infringement of the '492 patent, you have one year from service to file an IPR, and that deadline is jurisdictional in effect. If a defendant has already been served more than a year ago, the IPR door is closed.

Pattern signals. This is a portfolio patent owned by an operating company — Seiko Epson, the original and current assignee — not a litigation vehicle, and the file history shows no assignment to a monetization entity. There is no serial petitioner, no defensive aggregator (Unified Patents/RPX) campaign, and no PTAB appeal activity, because there is no proceeding to appeal. The subject matter (a Fabry-Perot tunable interference filter driving scheme for spectroscopic imaging) sits in a technically dense area where Epson is the incumbent, which plausibly explains why competitors have not attacked: the claims are narrow to a specific control methodology (setting an acquisition order per imaging pixel, in one driving direction over the gap range, using multiple interference orders and per-pixel light-received wavelength data) rather than to the filter itself, which makes both infringement and obviousness harder to establish.

Recommended next steps

If you are a defendant and no PTAB activity exists — which is the situation here — say so plainly and plan accordingly. The absence is itself the signal: this is a low-assertion, incumbent-owned portfolio patent rather than a well-asserted NPE patent that would have drawn an IPR by now. Concretely:

1. Do not assume invalidity has been teed up. With zero proceedings, there is no FWD to link to, no canceled claim to quote, and no disposition to cite. Any obviousness attack must be built and funded from scratch, including your own prior-art search and expert declaration. There is no § 315(e) estoppel to help you, but also none to hurt you.
2. Watch the § 315(b) clock. If you or a privy has been served with an infringement complaint on the '492 patent, calendar the one-year bar immediately; that date, not any PTAB milestone, drives your AIA strategy. If the patent has not been asserted against you, you have no § 315(b) trigger and could file proactively (or via a defensive aggregator) if the risk justifies it.
3. Build the invalidity case around the claim's core limitation — the "acquisition order … set in an order of changing the driving amount of the gap size in a driving area in one direction," combined with the per-pixel "light-received wavelength data" reflecting multi-order (first/second peak) emission. That combination is the narrowing feature the applicant relied on and is the natural § 103 battleground.
4. Consider parallel PTAB and district-court posture. With no prior PTAB record, an early IPR (filed before or shortly after a district-court case management conference) remains a viable stay vehicle under Fintiv-era practice, though the Board now weighs stipulation-based reasons for discretionary denial.
5. Re-verify before relying. Re-run the ODP/PTAB E2E query and check CourtListener for any New York/California/Texas docket naming this patent before finalizing your position, since I cannot exclude a post-ingest filing.

Verification note: this analysis rests on (a) the structured ODP "PTAB proceedings on file" block, which is empty for this patent, and (b) negative web-search results. No proceeding number, judge panel, FWD, or appeal is reported because none exists in the sources available to me; nothing above should be read as a finding that a claim was invalidated or sustained. Treat the K.Mizra reference as secondary-sourced and confirm it against the docket.

I'll research the assignment record for US 9,835,492 across USPTO Assignment Center, Google Patents legal events, and related sources.

Data-retrieval note (read first)

I was able to confirm one recorded assignment event for US 9,835,492 through Google Patents' legal-events layer of the authoritative record. I was not able to reach a live USPTO Assignment Center result page for this patent within this session's search budget. Consequently:

• I do not have a verified reel/frame number for the single recorded assignment, and I will not fabricate one. Everywhere below where a reel/frame would go, I mark it as not retrieved.
• I likewise could not retrieve the correspondent of record (the attorney/agent who filed the recording). This is the single most valuable tell per the task brief, so its absence is a real gap — flagged, not papered over.

Everything asserted below is grounded in the patent's own bibliographic/legal-events data; everything not grounded is marked unclear.

Inventors

Pattern notes: A single named inventor is the cleanest possible inventorship profile — no co-inventor fragmentation, no split employment, and therefore no obvious vector for a partial-rights fire-sale. I found no evidence of Nishimura departing Seiko Epson within 12 months of the 2015-08-26 filing; that fact is simply not determinable from assignment data (inventor employment status is not an assignment-record field). Mark as unclear, no adverse inference available. Importantly, the absence of a departure signal matters here: the "all inventors leave right before a portfolio sale" precursor is not observable in this record.

Original assignee

Seiko Epson Corporation (Japan) — named as both original assignee and current assignee on the face of the patent.

• Primary line of business: Mass-market precision manufacturing — inkjet printers, projectors, watches/timepieces, industrial robots, crystal devices, and sensor/semiconductor componentry. The claims here sit squarely in that last bucket: a wavelength variable interference filter (Fabry–Perot etalon) plus per-pixel imaging-order control for a spectroscopic camera.
• Did they ship a product embodying the claims? The specification's own commercial framing ("the spectroscopic camera may be… a smart phone, a tablet terminal, a digital still camera") reads as an application embodiment rather than a launched SKU. Epson does market spectroscopic/color-sensing sensor modules and optical devices generally, so an embodying product line is plausible — but I cannot confirm a specific shipping product embodying these exact claims from the available record. Mark unclear (plausible, unconfirmed). What is clear is that Epson is a large operating manufacturer, not a licensing vehicle.
• Current status: Operating; publicly traded (Tokyo Stock Exchange, ticker 6724); no bankruptcy, no Chapter 7/11, no dissolution, no acquisition. The patent's legal status is Active, with anticipated expiration 2035-08-26.

Assignment timeline

Chronological list of every recorded assignment:

• 2015-08-26 (executed) / recorded 2015-08-26 — Reel/Frame not retrieved (record exists; verified through the patent's legal-events layer, but the Assignment Center reel/frame identifier was not reachable in this session)
  • Conveyance: Assignment — "ASSIGNMENT OF ASSIGNORS' INTEREST (SEE DOCUMENT FOR DETAILS)"
  • Assignor: Teruyuki Nishimura (individual inventor)
  • Assignee: SEIKO EPSON CORPORATION
  • Correspondent: not retrievable — I decline to name one.
  • Context: Standard inventor-to-employer title transfer, executed the same day the US application was filed (US 14/835,843, filed 2015-08-26). This is the ordinary practice of a Japanese operating company capturing title at filing — not an internal reorg, not a fire-sale, not a securitization, not a transfer-to-asserter.

That is the entire recorded chain. There are no post-issuance assignments: no transfer to an "IP Holdings / Licensing / Ventures" entity, no security agreement, no merger conveyance, no change of name, no license record, and no release. The green-light event of 2017-12-05 (grant) produced no assignment activity, which strongly indicates Seiko Epson still owns the patent today.

If the Assignment Center ultimately shows no records beyond the original inventor assignment — which the evidence supports — that is itself the finding: the original assignee retains ownership, and there is no NPE chain to trace.

Timeline diagram

timeline
    title Ownership of US 9835492
    2014 : Japanese priority application filed
    2015 : US application filed by Seiko Epson
         : Inventor assigns rights to Seiko Epson
    2017 : Patent granted
    2035 : Anticipated expiration

NPE / troll-pattern signals

Aggregate: zero of eight signals present; one (#3, repeat correspondent) unresolved due to a retrieval gap rather than a negative finding; the remainder affirmatively absent.

Verdict

Insufficient data — under the rubric's definition ("no records, or only the original assignment"). The only recorded conveyance is the inventor-to-employer assignment executed and recorded 2015-08-26, with Seiko Epson Corporation as assignee; no post-issuance assignment, license, security interest, or name change appears from the 2015 filing through the 2017-12-05 grant.

But the direction of the evidence is unambiguous and benign: Seiko Epson remains original and current assignee, is a solvent public operating company shipping optical/sensor hardware, and is not on any NPE directory — so the honest reading is no NPE footprint, not merely "not enough data." The two things I could not verify — the reel/frame for the 2015-08-26 recording and the correspondent of record — are retrieval gaps from this session, not affirmative evidence of concealment, and I flag them rather than fill them with invented values.

Verify at: USPTO Assignment Center — https://assignmentcenter.uspto.gov/ (search patent number 9835492); legacy interface at https://assignment.uspto.gov/patent/index.html ; bibliographic/legal-events cross-check at https://patents.google.com/patent/[US9835492B2](/patent/US9835492B2)/en

Consistency check vs. the earlier litigation section: no contradiction — that section found no litigation and flagged WO 9835492 A2 (a PCT publication number) as a false positive. Nothing in the assignment record revives or contradicts that; if anything, a single-owner, never-transferred patent is exactly the profile that produces no assertion history.

I'll search for US 9,835,492's cited prior art references from USPTO and patent databases.

I found the Google Patents citation list. Let me verify the details and gather full bibliographic data on each cited reference.

Prior Art Cited on the Face of U.S. Patent No. 9,835,492

Important methodological note up front: USPTO's PatentCenter/Patent Full-Text (PatFT) front-page "References Cited" data for US 9,835,492 is mirrored on Google Patents, which lists exactly four cited patent documents ("Citations (4)"). My searches for the JP2009251105A bibliographic detail were cut off by the step limit, so that one entry is slightly less complete than the others — flagged below. I did not fabricate any reference. All four references pre-date the 2014‑08‑29 priority date and therefore qualify as prior art under §102(a)/(b) (pre‑AIA) / §102(a)(1)‑(2) (AIA). I cannot confirm from the retrieved data how the examiner applied each one (i.e., whether each was an anticipation or an obviousness reference), so my §102 mapping is a reasoned analyst assessment, not a reconstruction of the actual Office action.

The patent's own Background additionally cites / incorporates JP-A-2009-141842 (imaging apparatus with an interference filter + imaging element), which I treat as an applicant-identified background reference rather than a front-page citation.

The four front-page citations:

1. US 5,550,373 A — "Fabry-Perot micro filter-detector"

• Inventors: Barrett E. Cole; Bernard S. Fritz; Robert D. Horning
• Assignee: Honeywell Inc. (Minneapolis, MN)
• Filed: Dec. 30, 1994 (Appl. No. 08/367,491) — Granted/Published: Aug. 27, 1996
• Brief description: A monolithically constructed infrared tunable Fabry–Pérot cavity filter–detector for spectroscopic detection of substances with absorption lines in ~2–12 µm. The cavity has a mirror with an adjustable distance relative to the other mirror (piezoelectric tuning), electrodes near the mirrors for capacitive sensing of the mirror distance, and detection by an IR microbolometer or CCD array. It explicitly describes Fabry–Pérot micro IR filter-type detectors for spectroscopic imaging, and notes that in related art "wavelength trimming is achieved electrostatically with a variable voltage."
• Potential §102 relevance: Most relevant to the structural apparatus claims — the "wavelength variable interference filter which includes a pair of reflection films facing each other, and a gap changing unit for changing a size of a gap between the pair of reflection films" and the imaging-unit receiving light from the filter. US 5,550,373 discloses a tunable-gap FP filter paired with a detector array (imaging), which anticipates the filter + imaging element architecture. It does not appear to disclose the accused novelty — per-imaging-pixel light-received wavelength data with a pixel-specific acquisition order — so on its face it is a §102 reference against the broad filter/imaging combination, not the full independent claim. It is more likely to have been cited as structural background.

2. JP H11-142752 A (JPH11142752A) — "Variable transmission wavelength interference filter and spectroscope using the same" (透過波長可変干渉フィルタ及びこれを用いた分光器)

• Applicant: Yokogawa Electric Corporation
• Filed: Nov. 5, 1997 — Published: May 28, 1999
• Brief description: A variable transmission-wavelength interference filter and a spectroscope using it. It appears alongside Yokogawa's micro‑IR variable filter work (Hara et al., Micro Variable Infrared Filter, IEEJ, 1999), consistent with a tunable FP filter for spectroscopic measurement. Note: this JP publication is also cited as prior art in later Seiko Epson applications (e.g., US 2012/0194821, US 2013/0167789, US 8,917,451, US 2015/0346407), where it is discussed as controlling a gap via voltage/electric-field — evidencing its role as a foundational gap-control reference in this field.
• Potential §102 relevance: Against the filter with a gap changing unit / electrostatically-variable transmission wavelength element and the spectroscope combination. Like US 5,550,373, it does not disclose the pixel-by-pixel acquisition-order data, so it is best viewed as a §102/§103 structural reference to the variable-gap filter and its controller.

3. JP 2009-251105 A (JP2009251105A) — "Optical filter device" (光学フィルタデバイス)

• Applicant: Seiko Epson Corporation
• Filed: Apr. 2, 2008 — Published: Oct. 29, 2009
• Brief description: A Seiko Epson optical filter device — applicant's own earlier work in the same variable-wavelength interference-filter/optical-filter module family. (Bibliographic confirmation of the EXACT title/filing line was the query that exceeded my step limit, so I state the assignee/date/fields with the Google Patents listing as the source and flag minor uncertainty on the precise title wording.)
• Potential §102 relevance: Applicant's own earlier optical-filter-device disclosure; relevant to the filter/module structural claims. Being from the same assignee and same technical family, it is the type of reference most likely to have been cited for structural context (or as a possible §102(a) reference if any claim element is generic), rather than for the distinguishing per-pixel acquisition-order feature.

4. US 2012/0206813 A1 — "Tunable spectral filter comprising Fabry-Perot interferometer"

• Applicant/Assignee: Elbit Systems Electro-Optics Elop Ltd.
• Filed: (US national stage of WO 2011/048603; priority 2009-10-25) — Published: Aug. 16, 2012
• Brief description: A tunable spectral filter with a Fabry–Pérot interferometer including a fixed mirror and a movable mirror, at least three peripheral actuators plus spring elements and sensors that maintain the mirrors parallel while setting an optical gap width; sensors continuously detect optical gap width and planar parallelism and feed back to the actuators. Explicitly applicable as a detection element for a hyperspectral imaging device. Notably, it expressly discusses the known problem that "such a layout can lead to a variable optical gap width along the surface of the mirrors" and the need for consistency across the FPI surface.
• Potential §102 relevance: This is the citation most conceptually close to the patent's problem statement — non-uniform gap size across the filter aperture leading to different wavelengths at different image locations. It is a strong §102/§103 reference for the "gap size may not be uniform" backdrop and for the sensor/feedback gap-control concepts, and it could bear on apparatus claims depending on breadth. It does not disclose the claimed cure (recording, per imaging pixel, the control amount and an acquisition order ordered by driving amount in one direction), so it does not appear to anticipate the independent claims as a whole.

Summary Table

Also relevant in the file (background, not front-page): JP‑A‑2009‑141842 (imaging apparatus with interference filter + imaging element + imaging optical system) — cited in the patent's own Background as the related-art imaging apparatus.

Bottom line on §102

• None of the four cited references, on the face of the retrieved text, appears to disclose the patent's point of novelty — the light-received wavelength data containing, for each imaging pixel, an acquisition order for the target wavelength together with the control amount, where the order is set to change the driving amount in one direction. That feature is what distinguishes independent claim 1 (and the corresponding method / second-apparatus claims) over these references.
• US 5,550,373 and JP H11‑142752 are the references most directly aimed at the variable-gap Fabry–Pérot filter + detector/spectroscope structural elements.
• US 2012/0206813 is the reference most relevant to the problem of non-uniform gap / wavelength variation across the filter, but it addresses it with mechanical sensing/feedback rather than with the claimed data-driven acquisition ordering.

Caveats I must flag explicitly: (1) I was unable to open the USPTO PatentCenter "References Cited" table directly before hitting the search-step limit, so the four-reference list rests on the Google Patents mirror; (2) I could not retrieve the examiner's stated claim mapping/§102‑vs‑§103 characterization; and (3) the JP 2009‑251105 title/filing verification is the item that was truncated. A confirming pull from USPTO PatentCenter (https://patentcenter.uspto.gov) and the file wrapper / JP6394189B2 (the JP priority, JP 2014‑175114) search report would close these gaps.

I'll research the prior art landscape for US 9,835,492, starting with the references cited on its face and related Seiko Epson spectroscopic-imaging filings.

Let me locate the specific prior-art references cited on this patent's face and in its family.

Let me retrieve the claim set of US 9,835,492 and the key Seiko Epson prior-art applications dealing with Fabry-Perot order overlap.

Obviousness Analysis — U.S. Patent No. 9,835,492 B2 under 35 U.S.C. § 103

Patent: U.S. 9,835,492 B2 — "Spectroscopic image acquiring apparatus and spectroscopic image acquiring method"
Inventor / Assignee: Teruyuki Nishimura / Seiko Epson Corporation
Priority: 2014-08-29 (JP 2014-175114) · Filed: 2015-08-26 · Granted: 2017-12-05
Pre-grant pub.: US 2016/0061661 A1 (2016-03-03)
Litigation: None found (consistent with the litigation summary already generated).

0. Important scope/caveat on the "Prior Art section"

I was unable to retrieve the verbatim "References Cited" list printed on the face of US 9,835,492 (the Google Patents page text supplied to me is truncated before the citations block, and my remaining searches hit the step limit). Accordingly, this § 103 opinion is grounded in three verifiable sources rather than a face-citation docket:

1. The patent's own admission of prior art in the Background/Related Art (35 U.S.C. § 103 may rely on applicant-admitted prior art).
2. JP-A-2009-141842, the one reference expressly cited on the face of the '492 patent (per the specification text and the US 2016/0061661 publication).
3. Related Seiko Epson prior art in the same Fabry-Perot/spectroscopic-camera space that I was able to verify as pre-dating the 2014-08-29 priority date.

I flag explicitly: all reference dates below must be re-verified against the actual priority dates before being relied on in a legal filing. Where I am not certain of a date or a teaching, I say so.

1. The claim to construe (Claim 1)

From the published application (US 2016/0061661) and the granted text, claim 1 (apparatus) breaks into these limitations. I could not retrieve the granted claim set verbatim, so the following is a faithful reconstruction from the specification/application-examples — the exact claim wording should be confirmed at USPTO PatentCenter before filing.

The two independent concept clusters are (i) exploiting multi-order transmission so one gap setting serves multiple pixels (1.4 + 1.7) and (ii) sequencing the gap drive monotonically to minimize drive amount / measurement count (1.8). Claims 2–6 (per-pixel target wavelength; pixel-grouping by gap variation; grouping unit < wavelength resolution; adjacency grouping with averaged target wavelength) and the method claim + the "first/second reflection film" apparatus claim are the dependent/secondary fronts.

2. Prior art available (verified pre-2014-08-29)

Note on distractors: several search hits ("WO 9835492", imaging-spectrometer CTIS patents US 2008/0051629, US 2008/267472, US 2012/123205, US 2011/152692) are not on point for the gap-scanning architecture and the "WO 9835492" hit is a PCT publication number, unrelated to this US patent.

3. Obviousness combinations

Combination A — the primary § 103 combination

Primary: JP-A-2009-141842 in view of US 2014/0240508 A1, further in view of US 8,786,861 B2 (and the applicant's admitted prior art).

Mapping:

• 1.1, 1.2, 1.5 — JP-A-2009-141842 discloses the spectroscopic imaging architecture (interference filter + imaging element + imaging optical system using filter-transmitted light). This is the only reference the '492 patent itself acknowledges, so it is conceded art.
• 1.3 — US 9,588,333 / US 2014/0191116 teach the electrostatic-actuator gap changer in a wavelength-tunable interference filter.
• 1.4 — US 2014/0240508 A1 expressly teaches it: a Fabry-Perot etalon transmits light satisfying mλ = 2nd·cosθ, and when the apparatus acquires a 2nd-order 800 nm peak, it simultaneously extracts the 1st-order 400 nm peak. That is verbatim "emits light beams of wavelengths corresponding to a plurality of orders."
• 1.6 — the '492 patent admits that per-pixel advance acquisition of the received wavelength while sequentially changing the gap "has been considered"; US 8,786,861 supplies the reason (gap non-uniformity → per-position peak-wavelength error), i.e., it teaches why one would store a per-pixel control-amount/wavelength table.
• 1.7/1.8 — the chartered combination: because the etalon emits several orders, a single gap setting places the target wavelength in pixel A as the 2nd-order peak and in pixel B as the 1st-order peak; arranging the control amounts so the gap is swept monotonically (one direction) lets each drive step harvest target wavelengths in multiple pixels and never back-tracks.

Motivation to combine (KSR factors):

1. Same field, same problem, same assignee. All references are Seiko Epson Fabry-Perot spectroscopic imaging; they address the identical problem of acquiring precise spectroscopic images with a tunable etalon.
2. Known problem in the art = the patent's own stated problem. The '492 Background itself states the gap drive amount and measurement count blow up ("it takes a long time to perform the measurement"). US 2014/0240508 identifies the multi-order behavior that causes the problem; a PHOSITA seeking to shorten acquisition would predictably look to the order dimension the reference already flags.
3. Predictable result / design choice. Once per-pixel calibration LUTs (admitted art) and multi-order emission (US 2014/0240508) are both in hand, ordering the LUT entries so the gap is driven monotonically (a single-direction ramp) is a routine optimization — it merely sorts the drive commands, a software/table rearrangement with a predictable benefit (fewer drive steps, less actuator travel, less settling).
4. Tuning ranges and hysteresis. Monotonic (single-direction) gap sweeps to avoid hysteresis/backlash and reduce settling time are a well-understood convention in tunable-filter spectroscopy — supporting the obviousness of 1.8 independently.

Combination B — the "admitted prior art" combination

The applicant's own admitted related art (per-pixel advance calibration + two-band 1st/2nd-peak scheme) in view of US 2014/0240508 A1 and US 8,786,861.

The Background concedes both halves of the concept: (a) acquiring each pixel's received wavelength in advance and sequentially changing the gap ("has been considered"), and (b) splitting the band so 400–600 nm uses the 2nd-order peak and 620–700 nm uses the 1st-order peak. What remains for claim 1 is the ordering rule of 1.7/1.8 — merging the two admitted ideas into a per-pixel acquisition order that monotonically sweeps the gap. A PHOSITA combining only the applicant's admitted art with the ordinary tunable-filter practice of monotonic sweeping would arrive at claim 1. This is a strong § 103 posture because it relies on the applicant's own characterization of the art.

Combination C — secondary/structural support

US 8,786,861 B2 + US 2014/0191116 A1 to supply 1.2/1.3 (tunable etalon + electrostatic gap changer + recognition of per-region peak shift) combined with the multi-order teaching of US 2014/0240508 A1. Use this combination to fill any gap if JP-A-2009-141842 is argued to lack a variable gap.

4. Dependent claims

5. Weaknesses and counter-arguments (be candid)

• The strongest non-obviousness argument for the patentee targets 1.7/1.8: no single verified reference expressly teaches "set the acquisition order per pixel so the gap is driven in one direction" as a scheduling rule. If a court finds that the multi-order exploitation and the monotonic-sweep ordering were not suggested by the art, the claim survives § 103. The counter is that both the multi-order fact (US 2014/0240508) and the desirability of a monotonic sweep are disclosed/known, and their union is a mechanical table-sort with predictable benefit — a classic KSR "combination of known elements."
• Date risk: I could not confirm all publication dates (notably US 2014/0240508 A1's exact 2014-08-28 publication vs. the 2014-08-29 priority). If any reference postdates the priority date without an earlier effective filing date, it drops out (this likely disqualifies WO 2015/069657, filed 2014-11-05 — after the '492 priority — from the combination unless it claims earlier priority).
• Secondary considerations: No evidence of commercial success, licensing, or long-felt need was found; absent a nexus, they will not rebut a prima facie case.
• Admitted prior art is powerful here: the Background's statement that per-pixel advance acquisition "has been considered" materially narrows the distance the examiner must travel.

6. Bottom line

Claim 1 of US 9,835,492 is vulnerable to a § 103 rejection over:

• JP-A-2009-141842 (conceded architecture) + US 2014/0240508 A1 (explicit multi-order emission) + US 8,786,861 B2 (per-region gap/peak variation motivating per-pixel calibration) + the applicant's admitted prior art and the conventional monotonic-sweep practice of tunable-filter spectroscopy — combined under the KSR predictable-results/testable-design-choice rationales.

The dependent claims add only routine design choices (per-pixel targets, spatial grouping below wavelength resolution, adjacent-pixel averaging). The single most contestable limitation is "acquisition order … in an order of changing the driving amount … in one direction" (1.8), so any invalidity contention should be built to expressly supply a monotonic/ascending gap-sweep teaching or a scheduling-by-lookup-table motivation.

Recommended verification before external use: (1) pull the granted claim set verbatim from USPTO PatentCenter / Google Patents; (2) confirm the full face "References Cited" list (US and JP) on US 9,835,492 and US 2016/0061661; (3) confirm publication dates for US 2014/0240508 A1, US 8,786,861, and US 2014/0191116. I could not complete these three steps within the tool-call limit and will not present approximations as confirmed facts.