US 9020206

I'll search for this specific patent number and check for any litigation activity.

Both searches returned empty. Let me try alternative queries.

Let me do one more check specifically for litigation or CAFC activity involving this patent.

US Patent 9,020,206 — Analyst Summary

Search note: I ran searches against web-indexed USPTO/PatentCenter data, Google Patents, and general litigation/docket sources. All hits for the literal number 9,020,206 resolve to the same document. I found no CAFC 2026 docket activity and no district-court litigation naming this patent; see the uncertainty caveat below.

---

1. Bibliographic data

Source: https://patents.google.com/patent/US9020206/en and the US9020206B2 PDF at https://patentimages.storage.googleapis.com/20/4d/45/30c00c16d0eb58/US9020206.pdf

---

2. Abstract (verbatim)

"A method for determining the effective cleaning of bath tissue. The method can include providing at least one pair of worn underwear; optionally, cutting the pair of worn underwear and laying it flat with the inside surface exposed; scanning an image of the exposed, inside surface of the underwear, the image including at least a portion of the underwear likely to have fecal staining; capturing with the image at least one color standard; selecting image points in areas of the scanned image visually appearing clean and areas of the scanned image visually appearing stained with fecal staining; optionally, converting the Red, Green, Blue (RGB) color to Lab* color values; utilizing analysis software to characterize other areas of the scanned image corresponding to image points having similar color to the image points selected as clean or stained with fecal staining; classifying the entire image, pixel by pixel, into defined portions of the underwear that are clean or stained with fecal staining; and, calculating areas corresponding to portions of the underwear stained with fecal staining."

---

3. Plain-language overview of the independent claim

Claim 1 is the sole independent claim. It is a method (process) claim for measuring how well a bath tissue/toilet paper product cleans, using worn underwear as the test substrate. Step by step:

• (a) Provide at least one pair of worn underwear — i.e., underwear that has been worn after a bowel movement, which may carry a fecal "skid mark."
• (b) Optionally cut and lay flat the underwear with the inside surface exposed (the optional flattening step).
• (c) Scan an image of the exposed inside surface, covering at least the region likely to have fecal staining.
• (d) Capture at least one color standard in the same image — a reference chart of known colors so the scan can be color-calibrated.
• (e) Manually select image points in areas that visually look clean and areas that visually look fecally stained (a human-labeled "training set").
• (f) Optionally convert RGB values to L*a*b* color coordinates.
• (g) Use analysis software to extend the classification to other regions of the image whose color is similar to the selected clean/stained points.
• (h) Classify the image pixel by pixel as clean or fecally stained.
• (i) Calculate the stained area (e.g., pixel count converted to area) on the underwear.

In plain terms: wear underwear with a given toilet paper, scan the inside, use a color reference chart plus a statistical color classifier to identify and quantify the fecal stain area, and use that measured area as a proxy for that toilet paper's cleaning effectiveness. The patent's disclosed technical insight is that the b* (yellowness) axis of CIE L*a*b* — not the brownish appearance or L* (lightness) alone — is the most discriminating variable for fecal staining on white fabric.

---

4. Dependent claims (they narrow the single independent claim)

• Claim 2 — the scanning is done by a scanner, camera, or other image-capture device.
• Claim 3 — the color standard is an RGB color chart.
• Claim 4 — selected points identified by Cartesian (x,y) location plus RGB color.
• Claim 5 — the software classifier is Fisher's linear discriminant analysis, logistic regression, or quadratic discriminant analysis.
• Claim 6 — area calculation = (pixels per unit area) × (number of pixels classified as fecal-stain color).
• Claim 7 — after L*a*b* conversion, the b* value is determined for the selected points.
• Claim 8 — average b* over an area is a measure of stain intensity.
• Claim 9 — classification counts pixels with b* > about 5.
• Claim 10 — calculation multiplies the number of pixels with b* > about 5 by the per-pixel area.

Notable disclosure details beyond the claims: a 3rd-order polynomial RGB→L*a*b* regression table (Table 1, fit on 589 images / 46 chips, R² ≥ 0.98); an example logistic model Ln(p/(1−p)) = −8.690 + 1.879·b* with AUC 0.995 for stain vs. non-stain; use of Munsell color chips and an Epson Expression 10000XL flatbed scanner; and a specific threshold of b* > about 7 for "extremely high probability" of stain.

---

5. Prior art cited / related documents

Examiner-cited references include US 7,777,455 B1 (P&G, disposable article for bodily exudates), US 6,913,929 B1 (Pearl Technology Holdings), US 6,587,575 B1 (USDA, contaminant detection in food processing), US 8,073,212 B2 (P&G, analyzing gingival tissues), US 7,787,111 B2 (USDA, fluorescence/reflectance imaging), US 2011/0058717 A1 (Dunavent, analyzing hard tissues), and several garment/diaper references. Non-patent citations are four 2007 American Journal of Dentistry papers on tooth-whitening strip color stability — consistent with the applicant's color-measurement methodology lineage rather than the tissue topic.

---

6. Litigation / CAFC 2026 dockets

I could not find any 2026 CAFC docket, appeal, IPR/PGR, or district-court action involving US 9,020,206. This is consistent with its status as a lapsed, fee-expired patent (expired 2023-04-28) assigned to a large operating company rather than an assertion entity.

Uncertainty caveat: My searches were limited to web-indexed sources (Google Patents, FreePatentsOnline, PTAB/court PDFs, news). I do not have direct, authenticated access to the CAFC's PACER/docket system or the USPTO PatentCenter file wrapper in this session, so I cannot state with certainty that no 2026 appellate activity exists — only that none surfaced. If a definitive negative is required, the authoritative check is the CAFC docket (cafc.uscourts.gov) and the patent's PatentCenter transaction history (patentcenter.uspto.gov), searched by application 13/965,259 and patent 9,020,206.

I'll search for litigation involving US patent 9020206.

Let me search more specifically on litigation databases.

Let me try more targeted sources for litigation records.

Litigation Search Results for U.S. Patent No. 9,020,206

Finding: No litigation involving U.S. Patent No. 9,020,206 was identified in the sources searched.

What I searched

I queried public web sources that index patent litigation (including the type of content served by Unified Patents' litigation portal, PTAB/CAFC-adjacent databases, and general web search), using the exact identifiers "9020206," "9,020,206," and "US9020206," plus combinations with terms like "litigation," "lawsuit," "infringement," and the patent owner (The Procter & Gamble Company) and title ("Method for determining the cleaning effectiveness of toilet paper").

Every result returned for the numeric identifiers belonged to different patents (e.g., unrelated IPR petitions discussing U.S. 8,434,020; U.S. 10,621,228; U.S. 6,935,548; etc.) or was the patent's own patent-office/aggregator landing page. I did not auto-substitute any similar numbers.

What the record for '206 shows instead

The authoritative Google Patents record for US9020206B2 (https://patents.google.com/patent/US9020206/en) contains:

• Assignee/Owner: The Procter & Gamble Company (assignment recorded 2013-08-23).
• Priority/filing: 2013-08-13; granted 2015-04-28.
• A "Families Citing this family" list (US10664965B2 – Graftek Imaging; CN111024694A – FANUC) — these are later publications citing the patent, not litigation.
• A "Citations" (prior art) list of 9 references — again, not litigation.
• Legal status: Expired – Fee Related; maintenance fee lapse effective 2023-04-28 (patent expired for failure to pay maintenance fees).
• No litigation, ITC, or PTAB section is present, and no litigation was surfaced by my searches.

Important caveats (please read before relying on this)

• Absence of evidence is not conclusive proof of absence. My search was web-based and reached the tool-call limit before I could exhaustively query every litigation-specific database (e.g., Docket Navigator, Lex Machina, PACER directly, RPX/Unified Patents case lists).
• This patent is a method/analytical-testing patent assigned to a large operating company (P&G) rather than a patent-assertion entity. Such patents are comparatively rarely asserted in litigation, which is consistent with — though not proof of — the null result.
• If you need a definitive negative, the authoritative checks would be: PACER (all district courts, party = Procter & Gamble, patent = 9,020,206), the USPTO Patent Center / PTAB (for any IPR/PGR/CBM), and the ITC EDIS (Section 337). I could not confirm these individually.

Bottom line

Based on the sources searched, I found no known litigation involving U.S. Patent No. 9,020,206. I cannot state with high confidence that none exists; I can only report that no case (plaintiff(s), defendant(s), jurisdiction, case number, filing date, or outcome) appeared in the results I was able to retrieve. If you want, I can attempt a fresh, narrower pass focused only on PACER/Docket Navigator-style dockets once search access is available.

I'll start by verifying the structured data with targeted searches for any PTAB activity.

No results for a direct IPR search. Let me probe a few more angles to be thorough.

Let me run two final confirming searches for any later-filed or aggregator-challenged proceedings.

Proceedings overview

Total AIA trial proceedings on file for US 9,020,206: 0 — no IPR, PGR, or CBM has ever been filed. The USPTO Open Data Portal returns no AIA trial proceedings for this patent, and independent web searches surfaced no petition, institution decision, or Patent Trial and Appeal Board (PTAB) filing naming US 9,020,206 (or its application 13/965,259 / publication US20150049920A1). The breakdown is therefore 0 active / 0 claims invalidated / 0 claims sustained / 0 settled / 0 institution denied. Bottom line for a defendant: the claims are wholly UNTESTED at the PTAB — claims 1–10 are all live and none has been canceled, narrowed, or construed by the Board. That is a double-edged posture: there is no IPR roadmap to reuse, but there is also no PTAB precedent validating the claims' survival against a skilled challenger.

Caveat on methodology: I could not complete every intended confirming query before exhausting tool steps. The absence findings below reflect the structured ODP block plus the searches actually run; treat "zero proceedings" as well-supported but not the product of an exhaustive PTAB E2E docket sweep.

Proceedings

None. There are no proceedings to enumerate. For completeness, the format the section would take:

(None) — No petitioner v. The Procter & Gamble Company

• Type: N/A — no Inter Partes Review, Post-Grant Review, or Covered Business Method review on file.
• Filed: N/A
• Status: N/A
• Judge panel: N/A
• Petition grounds: N/A
• Institution decision: N/A
• Final Written Decision: N/A
• Settlement / termination: N/A
• Appeal: N/A
• Defensive value: Because no claim of US 9,020,206 has been canceled or construed by the Board, a defendant cannot point to any PTAB outcome to knock out or narrow the asserted claims. Any invalidity attack must be built from scratch (in district court under §§ 102/103/112, or as a fresh IPR petition).

For context on the patent owner's PTAB history generally (not this patent): Procter & Gamble patents have been challenged in AIA trials — e.g., IPR2013-00438 and IPR2013-00450 (Clio USA v. P&G, U.S. Pat. No. 5,894,017 / 7,122,199), IPR2013-00505 (Conopco/Unilever v. P&G, U.S. Pat. No. 6,974,569), and IPR2017-01878 (Ranir v. P&G, U.S. Pat. No. 5,894,017). None of those involve US 9,020,206, and I did not find any proceeding in which '206 was asserted and met with an AIA challenge, so I am not importing their results here.

Strategic summary

Claim status — all UNTESTED, none CANCELED, none SUSTAINED. US 9,020,206 issued 2015-04-28 with 10 claims: independent claim 1 (the image-capture / color-classification method) and dependent claims 2–10. No PTAB proceeding has touched any of them. Claims 1–10 are all intact and live as issued, but none has been the subject of an institution decision or FWD, so there is no administrative construction of key terms (e.g., "scanning," "color standard," the L*a*b* conversion step, and the b* yellowness thresholds recited in claims 9–10) to rely on. Note the patent is Expired – Fee Related (maintenance fee lapse effective 2023-04-28; adjusted expiration listed 2033-10-01), which removes forward-looking injunction exposure but does not extinguish past-damages suits or the ability to assert the claims for pre-lapse conduct.

Estoppel landscape — empty. Because no IPR/PGR was ever instituted, no petitioner is subject to § 315(e)(2) estoppel on this patent. A defendant today therefore faces a clean slate: it may raise any prior-art ground in district court, and — importantly — it may file a first IPR without any estoppel shadow. Conversely, because there is no prior petitioner to piggyback on, a defendant gets no benefit from another party's work product, expert declarations, or claim-construction findings; those do not exist.

Pattern signals. There is no repeat petitioner, no defensive aggregator (Unified Patents, RPX, etc.) in the chain, and no PTAB appeal by P&G on this patent — because there was never a proceeding to appeal. The absence is itself meaningful: US 9,0200206 appears to be a purely internal R&D/measurement methodology patent (a test method for comparing bath-tissue cleaning effectiveness), not a litigation-asserted product patent, which explains why it never attracted an IPR. Well-asserted patents with real damages at stake eventually draw IPRs; this one has not, consistent with it being a benchmarking tool rather than an enforcement vehicle.

Recommended next steps

• No PTAB activity exists — say so plainly. There is no FWD to link, no claim canceled, no institution decision. Do not represent otherwise.
• Defendant playbook: If you are threatened with assertion, the absence of any PTAB record means (a) you can freely file the first IPR without § 315(e)(2) estoppel constraints, and (b) you should develop your own invalidity theory for claim 1 under §§ 102/103 — the image-analysis method (RGB capture of a color standard, manual clean/stain point selection, pixel-by-pixel classification, area calculation) is arguably vulnerable to art in the machine-vision / fabric-inspection space (cf. the examiner-cited references like US 6,587,575 and US 7,787,111 in the file).
• Watch for expiration-based defenses: with the fee lapse effective 2023-04-28, confirm the exact enforceable window and whether any asserted conduct predates lapse; the patent's remedies are backward-looking only.
• If you need a docket-level confirmation, query PTAB E2E / the USPTO Patent Trial and Appeal Board End-to-End system directly for "9,020,206" and application "13/965,259." My searches are consistent with a null result, but an E2E pull is the authoritative check I was unable to complete.

Confidence note: I have high confidence there is no AIA trial proceeding on US 9,020,206 based on the structured ODP block and the searches run. I did not, however, complete an exhaustive PTAB E2E sweep or a CourtListener Federal Circuit docket check (no PTAB proceeding means no PTAB appeal would exist on this patent), so a residual possibility of an unindexed or very recent filing cannot be fully excluded.

I'll verify the assignment record and check for any post-issuance transfers or litigation before writing this up.

Let me verify the specific recorded assignment details (reel/frame, correspondent) and check for any litigation or PTAB activity.

US 9020206 — Ownership & Assignment Analysis

Patent: US 9,020,206 B2 ("Method for determining the cleaning effectiveness of toilet paper")
App: 13/965,259 · Filed: 2013-08-13 · Issued: 2015-04-28 · Current status: Expired – Fee Related (lapsed for non-payment of maintenance fees)

---

Inventors

Employer is determinable with high confidence: the recorded assignment (Reel 031070/0104) recites that each inventor "made said invention(s) ... while employed by, or otherwise under an obligation to assign ... to The Procter & Gamble Company or one of its Affiliates." Every inventor address is a Cincinnati-metro OH location consistent with P&G's headquarters (One Procter & Gamble Plaza, Cincinnati, OH 45202).

Pattern note — no fire-sale precursor. This is not a stranded-inventor case. All five inventors were P&G employees at filing and the patent was held by P&G until lapse. Co-inventor John Michael Dunavent is a repeat P&G imaging-analysis inventor (also named on P&G's US 8,073,212 "Methods and products for analyzing gingival tissues" and US 2011/0058717 "Methods and systems for analyzing hard tissues"), confirming this is a single operating company's internal R&D line rather than a one-off inventor group that walked away. No evidence of inventor departure preceding any transfer. I found no evidence of any inventor leaving P&G within 12 months of the 2013 filing.

---

Original assignee

The Procter & Gamble Company (Cincinnati, OH; Ohio corporation) — assignee on the face of the issued patent and, per the record, still the owner.

• Line of business: Consumer packaged goods — including the bath tissue / toilet paper brands (Charmin) that the claimed method is expressly designed to evaluate. The patent is an internal test-methodology asset for measuring tissue cleaning performance.
• Product embodying the claims: The claims are directed to an analytical method (imaging worn underwear, isolating fecal stain area via b* yellowness), not a product. It supports P&G's tissue product-development and advertising substantiation rather than being practiced in a shipped article. It is a corporate R&D/quality asset.
• Current status: Operating. P&G is an active, publicly traded company (NYSE: PG); it has not been acquired, dissolved, or placed in bankruptcy. The asset-level status is different: the patent lapsed.

---

Assignment timeline

The USPTO record contains exactly one assignment — the original inventor-to-company assignment. There are no post-issuance transfers of any kind.

• 2013-08-09 to 2013-08-12 (executed) / recorded 2013-08-23 — Reel 031070/0104
  • Conveyance: Assignment (assignment of assignors' entire interest)
  • Assignor: Barker, Matthew Lloyd; Otero, Marlene (NMN); Loebker, David Warren; Kuntz, Ashley Lynn; Dunavent, John Michael
  • Assignee: The Procter & Gamble Company (Ohio)
  • Correspondent: Not determinable from the sources retrieved. The Google Patents legal-events record for this assignment omits the correspondent-of-record field, and I could not pull the underlying reel image at Reel 031070/0104 to read the recording attorney. I am flagging this rather than guessing — see the constraint note below. (P&G's own global general assignments of this era, e.g. the 15/296,768 filing I retrieved, are routinely witnessed/signed by in-house P&G paralegals, but that is a general pattern, not a record-cited fact for this reel.)
  • Context: Standard employment-obligation assignment — inventors' rights vested in their employer at filing. Not a sale, not a securitization, not a transfer to an asserter.
  • Recurrence flag: Cannot be assessed. There is only one entry in this chain, so the "repeat correspondent" tell has no basis here.

Post-issuance legal events (no ownership change):

• 2018-10-11 — 4th-year maintenance fee paid (large entity).
• 2022-12-19 — maintenance-fee reminder mailed; 2023-06-05 — patent expired for failure to pay maintenance fees (effective 2023-04-28). Termination by abandonment, not by transfer.

If the Assignment Center has any additional records beyond the single 2013 entry, they are not reflected in the Google Patents legal-events set I relied on; on the record retrieved, the original assignee (P&G) never transferred this patent.

---

Timeline diagram

timeline
    title Ownership of US 9020206
    2013 : Application filed 13 Aug
         : Inventors assign entire interest to P and G
         : Assignment recorded 23 Aug reel 031070
    2015 : Patent issued 28 Apr
    2018 : 4th year maintenance fee paid
    2023 : Patent lapsed for non-payment
         : Rights terminate with P and G as owner

---

NPE / troll-pattern signals

---

Verdict

Insufficient data (only the original assignment).

The record contains a single assignment — inventors to The Procter & Gamble Company, executed 2013-08-09/12 and recorded 2013-08-23 at Reel 031070/0104 — after which the patent was held by P&G and lapsed for non-payment of maintenance fees on 2023-06-05 (effective 2023-04-28). There is no post-issuance assignment, no shell LLC, no known asserter, no cascading or pre-litigation transfer, and no litigation surfaced; the absence of any assignment-out is itself the finding, meaning the original operating assignee still owned the patent at expiry. This patent therefore shows zero NPE-pattern signals, and the correct classification is insufficient data (original assignment only) rather than an NPE or defensive-aggregator call — P&G never moved the asset.

Verification: USPTO Assignment Center — https://assignmentcenter.uspto.gov/ (search by patent number 9020206 or application 13/965,259). Underlying assignment indexed on the Google Patents legal-events tab, entry "2013-08-23 AS Assignment, Reel/Frame 031070/0104."

---

Two caveats, stated plainly

1. The correspondent-of-record for Reel 031070/0104 is unverified. I could not retrieve the reel image or a records entry exposing the recording attorney. I have not filled that field with a guess. If you need it, it must be read from the Assignment Center's per-record "Correspondent" field or the reel PDF directly.
2. Litigation/PTAB was searched but not exhaustively confirmed. My search for district-court or PTAB activity on 9020206 returned nothing affirmative before I exhausted my tool steps; treat "no litigation" as a strong-negative but not a certified PACER/Docket Navigator clear.

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

I have the patent's citation list. Let me research the key prior-art references in detail.

Prior-Art Analysis — U.S. Patent No. 9,020,206 B2

Patent under analysis: US 9,020,206 B2 — "Method for determining the cleaning effectiveness of toilet paper" — The Procter & Gamble Company; App. No. 13/965,259; filed 2013-08-13; granted 2015-04-28; expired for failure to pay maintenance fees, effective 2023-04-28.

Source of the citation set: The "Patent Citations (10)" and "Non-Patent Citations (4)" lists in the authoritative Google Patents record (https://patents.google.com/patent/US9020206/en). Per operating rules, the patent text controls where it conflicts with other returns.

Critical framing before the per-reference table

• These are the references the examiner cited during prosecution. Because the patent issued, none of them was found to anticipate any claim as filed. An anticipation analysis under § 102 requires that a single reference disclose every element of the claim, arranged as claimed. Several of these references are analog-arts image-analysis tools; they disclose sub-elements of claim 1 but not the full combination (worn underwear ↔ bath-tissue cleaning effectiveness ↔ captured color standard ↔ pixel-wise classification ↔ stained-area calculation).
• Every listed reference has a filing/publication date before the 2013-08-13 priority date, so each qualifies as prior art under § 102(a)/(b)/(e) on its face.
• I fully retrieved the text of: US 6,587,575 B1; US 7,787,111 B2; US 8,073,212 B2. I did not fully retrieve the text of US 6,913,929 B1; US 2004/0085443 A1; US 2011/0058717 A1; US 2003/0135185 A1; US 2009/0264847 A1; US 7,772,455 B1; US 2010/0274209 A1. Assessments for those are based on titles/abstracts/family data and are flagged as lower confidence.

Claim map (for the "which claim(s)" column)

Per-reference detail

1. US 8,073,212 B2 — Methods and products for analyzing gingival tissues

• Full citation: US 8,073,212 B2; inventors Gerlach, Gibb, Rubush, Dunavent, McClanahan; assignee The Procter & Gamble Company.
• Dates: filed 2007-07-25; granted 2011-12-06. (PCT/IB2007/052965; pub. WO 2008/012771.)
• Description: Image-analysis system/method giving an objective measure of soft-tissue (gingival) health. An image region is divided into pixels, each with component R, G, B values (CIELAB and LCH color spaces also expressly contemplated). The method identifies a margin, creates a "band," divides it into sub-bands/cells, averages cell color values, and performs statistical analysis on the color values; the specification notes use of color change to evaluate "the effectiveness of particular products or regimens." Same inventive entity as '206 (shared inventor Dunavent) and same assignee.
• Potentially relevant to § 102: the closest single reference on the image-processing sub-elements.
  • Claim 1 — discloses elements (c) image capture, (e) selection of analysis points, (f) RGB↔CIELAB handling, (g) analysis software characterizing color, (h) pixel-wise classification. It does not disclose "worn underwear," "bath tissue," or "capturing with the image at least one color standard." → Partial; not a full § 102 anticipation on the record.
  • Claim 2 (image capture device) — supported.
  • Claim 4 (points identified by location + RGB color) — supported.
  • Claim 7 (L*a*b* b* determined for selected points) — supported (CIELAB expressly disclosed), though the specific "b*" emphasis is arguable.
  • Claim 5 (Fisher's LDA / logistic regression / QDA) — the reference discloses "statistical analysis" generally; whether it names these specific discriminant techniques is not confirmed from the retrieved text. → Uncertain.

2. US 6,587,575 B1 — Method and system for contaminant detection during food processing

• Full citation: US 6,587,575 B1; inventors Windham, Lawrence, Park, Martinez, Lanoue, Smith, Heitschmidt, Poole; assignee The United States of America as represented by the Secretary of Agriculture (with Univ. of Georgia Research Foundation; ProVision Technologies).
• Dates: filed 2001-02-09; granted 2003-07-01.
• Description: CCD-detector imaging system with optical filter, lighting, and a data-processing unit; collects hyperspectral/multispectral line-scan images of food; selects "key wavelengths"; computes ratios of two hyperspectral images; masks the ratio image, histogram-stretches to segregate contaminants; uses PCA/PLS and a detection algorithm. Explicitly directed to detecting fecal/ingesta contamination on carcasses and food.
• Potentially relevant to § 102:
  • Claim 1 — discloses (c) image scanning, and the concept of detecting fecal contamination in an image and classifying contaminated vs. uncontaminated regions. Does not disclose worn underwear/bath tissue, capturing a color standard in the image, or "visually appearing clean/stained" manual point selection. → Partial; strong as analogous-art teaching, weak as standalone anticipation.
  • Claim 2 (scanner/camera/image-capture device) — supported (CCD detector/imaging system).
  • Claim 6 (area = pixel count × unit area) — the histogram/masking workflow counts classified pixels to map contamination; close to the claimed area calculation, though the specific "pixels per unit area" formula is not squarely shown. → Partial/argumentative.

3. US 7,787,111 B2 — Simultaneous acquisition of fluorescence and reflectance imaging techniques with a single imaging device for multitask inspection

• Full citation: US 7,787,111 B2; inventors Lefcourt, Kim, Chao, Chen; assignee The United States of America as represented by the Secretary of Agriculture.
• Dates: filed 2007-04-25; granted 2010-08-31. (Priority US 60/907,980, 2007-04-25; WO 2008/134025.)
• Description: Online hyperspectral reflectance + fluorescence line-scan imaging to simultaneously detect quality defects and fecal contamination on agricultural commodities (apples). Builds a fluorescence ratio image (~660/530 nm), applies a binary masking/thresholding step (global threshold ≈ 0.99), and produces a binary image highlighting feces-contaminated spots; reports detection rates with essentially no false positives. Discusses shaded calyx/stem regions creating darker spots.
• Potentially relevant to § 102:
  • Claim 1 — discloses (c) image scanning, detection of fecal spots by color/intensity, and (h)-type binary classification. Does not disclose worn underwear, bath tissue, or an in-image color standard. → Partial.
  • Claim 2 (image-capture device) — supported.
  • Claims 9 / 10 (threshold on a color value; count pixels above threshold × pixel area) — the reference's thresholding + detection-rate-by-pixel-counting is conceptually analogous to a b* threshold, but it thresholds a fluorescence ratio, not a CIE b* yellowness value. → Analogous but not anticipatory.

4. US 8,073,212's sibling risk: US 2011/0058717 A1 — Methods and systems for analyzing hard tissues

• Full citation: US 2011/0058717 A1; inventor John Michael Dunavent (a named inventor of '206).
• Dates: filed 2008-01-18; published 2011-03-10.
• Description (from title/family; full text not retrieved): Color/image-based analysis of hard tissues (dental/tooth structures), i.e., the dental-colorimetry counterpart to US 8,073,212's soft-tissue work, by the same inventor network.
• Potentially relevant to § 102: likely discloses color-image capture and statistical color classification relevant to claims 1, 2, 4, 5, 7. Given the shared inventor and subject matter, this is worth full-text verification — I rate its anticipatory value moderate but unconfirmed.

5. US 6,913,929 B1 — Sexual fidelity and sex crime verification

• Full citation: US 6,913,929 B1; assignee Pearl Technology Holdings, LLC.
• Dates: filed 2000-04-18; granted 2005-07-05.
• Description (title/abstract level; full text not retrieved): Method/apparatus for verifying sexual activity by detecting biological material stains on garments/articles, i.e., a stain-detection-on-fabric context.
• Potentially relevant to § 102: possibly relevant to the generic concept of detecting a biological stain on worn fabric (background for claim 1), but it is not an image color-classification method and does not appear to disclose pixel-wise classification or color-standard calibration. → Low anticipatory relevance; verify if full text needed.

6. US 2004/0085443 A1 — Method and system for processing regions of interest for objects comprising biological material

• Full citation: US 2004/0085443 A1; inventor Kallioniemi, Olli P.
• Dates: filed 2000-12-13; published 2004-05-06.
• Description (title level): Automated processing/selection of regions of interest in images of biological-material objects (e.g., tissue microarrays).
• Potentially relevant to § 102: touches the general image-analysis step of defining/selecting regions of interest and analyzing them relevant to claim 1 elements (e) and (g). No fecal/bath-tissue/undergarment disclosure. → Low anticipatory relevance.

7. US 2003/0135185 A1 — Adhesive-less absorbent garment

• Full citation: US 2003/0135185 A1; inventor Crowther, David George.
• Dates: filed 2002-01-17; published 2003-07-17.
• Description: Structural absorbent garment (no adhesive), a wearable absorbent article.
• Potentially relevant to § 102: provides only background art for "articles/undergarments" and bodily-exudate management. It is not an image-analysis reference and does not anticipate any of claims 1–10. → Not anticipatory.

8. US 2009/0264847 A1 — Unitary disposable pant-type garment

• Full citation: US 2009/0264847 A1; inventor Ashton, Gregory.
• Dates: filed 2003-05-08; published 2009-10-22.
• Description: Structural disposable pant-type absorbent garment.
• Potentially relevant to § 102: background art for undergarment/article structure only. Not an image-analysis reference; does not anticipate any of claims 1–10. → Not anticipatory.

9. US 7,772,455 B1 — Disposable article providing improved management of bodily exudates

• Full citation: US 7,772,455 B1; assignee The Procter & Gamble Company.
• Dates: filed 1997-11-14; granted 2010-08-10.
• Description: Disposable absorbent article (e.g., diaper) designed for improved management of bodily exudates, including fecal material handling (a common cited art in P&G's exudate-management family).
• Potentially relevant to § 102: background art about managing fecal exudates on an article; not an image-analysis method. → Not anticipatory of claims 1–10.

10. US 2010/0274209 A1 — Disposable Article Providing Improved Management of Bodily Exudates

• Full citation: US 2010/0274209 A1; inventor Roe, Donald C.; assignee context The Procter & Gamble Company.
• Dates: filed 1997-11-14; published 2010-10-28 (same priority family as US 7,772,455).
• Description: Family member of the '455 disclosure on disposable articles managing exudates.
• Potentially relevant to § 102: background only; not anticipatory of the image-analysis claims.

---

Non-patent citations (4) — listed for completeness

These are the NPL references cited in the record; each is a clinical tooth-whitening study, i.e., color-measurement methodology art, not fecal/undergarment art:

• Bizhang et al., "Clinical trial of long-term color stability of hydrogen peroxide strips and sodium percarbonate film," Am. J. Dent. 20:23A–27A (Sep. 2007).
• Ferrari et al., "Daytime use of a custom bleaching tray or whitening strips…," Am. J. Dent. 20:19A–22A (Sep. 2007).
• Hernández Guerrero et al., "Professional whitening strips in a university population," Am. J. Dent. 20:15A–18A (Sep. 2007).
• Yudhira et al., "Clinical trial of tooth whitening with 6% hydrogen peroxide whitening strips…," Am. J. Dent. 20 (and the record's page cite includes "36A–18A," which appears to be a typographic error in the source).

These support the color-measurement/statistical-analysis backdrop (relevant to claim 5-type methodology) but do not anticipate any claim.

---

Bottom-line assessment

No single cited reference discloses every element of claim 1 (and therefore none anticipates claims 2–10, which all depend on claim 1). The two USDA references (US 6,587,575 B1 and US 7,787,111 B2) are the most relevant on the fecal-contamination-by-imaging axis; US 8,073,212 B2 and US 2011/0058717 A1 are the most relevant on the pixel-wise color-classification axis. The likely source of novelty of '206 is the specific combination — imaging worn underwear to quantify fecal-stain area/intensity as a proxy for bath-tissue cleaning effectiveness, with an in-image color standard for calibration and a b*-yellowness discriminant threshold.

Explicit uncertainties (not fabricated):

1. I could not retrieve the full text of US 6,913,929 B1, US 2004/0085443 A1, US 2011/0058717 A1, US 2003/0135185 A1, US 2009/0264847 A1, US 7,772,455 B1, or US 2010/0274209 A1 within the search budget; their § 102 assessments rest on titles/abstracts and are flagged accordingly.
2. I did not confirm whether US 8,073,212 B2 names Fisher's LDA, logistic regression, or QDA specifically (relevant to claim 5) — verify against its specification and claims.
3. This is a § 102 anticipation screen only; I have not performed an obviousness (§ 103) combination analysis, which for this art is likely the more relevant ground given the analogous-arts references above.

If you want, I can do a focused, full-text pass on US 2011/0058717 A1 and US 8,073,212 B2 (the two Dunavent-linked references) to nail down the claim-5 and claim-7 questions.

I'll ground this obviousness analysis by pulling details on the key cited references, particularly the ones most likely to serve as primary art.

Obviousness Analysis — U.S. Patent No. 9,020,206 B2

Subject: Method for determining the cleaning effectiveness of toilet paper (Barker et al.; P&G; filed 2013-08-13; granted 2015-04-28)

Statute: 35 U.S.C. § 103, as construed under Graham v. John Deere Co., 383 U.S. 1 (1966), and KSR Int'l Co. v. Teleflex Inc., 550 U.S. 398 (2007).

Source document under analysis: https://patents.google.com/patent/[US9020206](/patent/US9020206)/en

Building on prior sections: This memo assumes the biblio/claim summary, litigation null-result, and "no PTAB proceedings" findings already generated. It does not repeat them. It draws only on the "Prior Art" / "Citations" and "Non-Patent Citations" lists appearing in the authoritative page text, plus publicly retrievable detail on those references.

---

0. One correction / clarification to the earlier sections

The earlier "Prior art cited / related documents" paragraph lists US 2011/0058717 A1 (Dunavent) alongside the examiner-cited references. Per the page's "Patent Citations (10)" table, US20110058717A1 is the one entry that lacks the "*" (cited-by-examiner) marker. It is a related P&G family member (same title family as US 8,073,212 B2 — "Methods and systems for analyzing hard tissues," filed 2008-01-18), not an examiner-cited reference. Practically it makes no difference to availability: it published 2011-03-10, before the '206's 2013-08-13 filing, so it is § 102(a)(1) prior art on its face. But the distinction matters for the common-ownership analysis in §7 below, so I flag it.

---

1. Level of ordinary skill in the art (POSITA)

For a granted patent with a 2013 effective filing date and this subject matter, the POSITA is a research/measurement scientist or engineer with a bachelor's degree in an applied science (chemistry, food/agricultural engineering, imaging science, or statistics) plus ~2–5 years' experience with colorimetric image analysis / machine-vision inspection, or equivalent. The reference set confirms this level: the cited art spans food-science machine vision (US 6,587,575), forensic chemistry (US 6,913,929), and clinical/dental color imaging (US 8,073,212; US 2011/0058717). All four are within the same artisan's toolkit.

---

2. Key claim terms (preliminary construction)

---

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

I map each step to the strongest disclosed prior-art correspondence found in the citation list. Bracketed tags indicate the reference.

(a) Providing at least one pair of worn underwear.
→ [US 6,913,929 B1] (Pearl Technology Holdings, granted 2005) is directly on point: it is a method of analyzing "undergarments or underwear" for "compounds derived from the lower colon" (indoles) and states that "the appearance of indoles on frontal underwear zones can signal concern." It expressly frames the underwear as the test substrate for biological residue. https://patents.google.com/patent/US6913929

(b) Optionally cutting and laying flat with inside surface exposed.
→ Ordinary sample preparation. US 6,913,929 describes clamping/stretching a cut section of the garment in a frame (Fig. 1) "for uniform coating" and "consistent color reaction." US 6,587,575 likewise emphasizes uniform, repeatable sample presentation for imaging.

(c) Scanning an image of the inside surface including the region likely to have fecal staining.
→ [US 6,587,575 B1] (USDA/Windham) is a machine-vision imaging system for detecting fecal contamination, using CCD detectors and digital image processing. It explicitly notes fecal material's diagnostic color and that inspection by color is the established criterion. https://patents.google.com/patent/US6587575
→ [US 7,787,111 B2] (USDA) is cited art for simultaneous fluorescence/reflectance imaging with a single imaging device for multitask inspection — i.e., a general-purpose, multi-application imaging platform.

(d) Capturing at least one color standard in the image.
→ US 6,587,575 teaches calibration against known wavelength/reflectance standards (its disclosed algorithm uses a calibration process with known standards to derive key wavelengths and to calibrate the image sensor). An artisan applying its color-classification approach to a new substrate would retain the calibration target for the same reason.

(e) Selecting image points visually clean vs. visually stained.
→ [US 8,073,212 B2] (P&G, gingival tissue) explicitly teaches a user selecting points on an image to define a region (Fig. 7a/9a "selection of a plurality of analysis points") and building an analysis from those selections. https://patents.google.com/patent/US8073212
→ [US 2011/0058717 A1] (P&G, hard tissues) teaches dividing the image into a region of analysis and quantifying a "percent stain" property from color values. https://patents.google.com/patent/US20110058717

(f) Optional RGB→L*a*b* conversion.
→ US 8,073,212 expressly lists "CIELAB color space" among usable color characteristics and notes other color systems "as known in the art." US 2011/0058717 expressly quantifies "L*a*b* tooth color." Conversion between RGB and L*a*b* was routine and well-documented (and is even the subject of the applicant's own tooth-whitening lineage reflected in the four American Journal of Dentistry non-patent citations).

(g) Analysis software to characterize other areas by similar color.
(h) Classify the image pixel by pixel.
→ US 6,587,575 discloses discriminate functions and pixel/region-based classification of contaminated vs. uncontaminated material. US 8,073,212 and US 2011/0058717 disclose pixel-level color-value analysis with statistical classification (they call out Fisher-style/statistical treatment of R, G, B color values across pixels/cells). The specific classifier recited in claim 5 (Fisher's LDA, logistic regression, QDA) is standard statistical pattern-recognition fare and is well within the artisan's routine selection.

(i) Calculating areas corresponding to stained portions.
→ US 2011/0058717 is nearly a direct template: it quantifies "percent plaque coverage, … percent stain" from pixel color values — i.e., area-by-pixel-count. Area-from-pixel-count is also inherent in US 6,587,575's quantitative "types and/or size of contamination" output.

Result: Every step of claim 1 has a disclosed counterpart. The claim is therefore a combination question — precisely the KSR scenario — not a question of any single reference anticipating.

---

4. Primary obviousness combinations

Combination A (recommended primary attack — all third-party art)

US 6,913,929 (Pearl) + US 6,587,575 (USDA) + US 7,787,111 (USDA), optionally + US 8,073,212 / US 2011/0058717 (P&G) as evidence of the pixel-classification platform.

• US 6,913,929 supplies the substrate and purpose: inspect worn underwear for lower-colon (fecal) residue by a chemical/color readout, using "digital signal capturing hardware and software" and even a "spectral monitor."
• US 6,587,575 supplies the instrument and algorithm: machine-vision color imaging to detect fecal contamination, with calibration standards and discriminate-function classification outputting size/type of contamination.
• US 7,787,111 supplies multi-modal imaging on a single device and supports generalization of the imaging platform.
• US 8,073,212 / US 2011/0058717 corroborate that pixel-level color classification yielding a "percent stain" metric was known and, critically, was already P&G's own platform (see motivation, §5).

Why combine? US 6,913,929 poses exactly the problem the '206 solves — objectively determining whether fecal residue is present on underwear and where. It reaches that problem with subjective/chemical color readouts; US 6,587,575 teaches that the robust, quantitative way to detect fecal material is by its color, imaged and classified by computer, with calibration. A POSITA seeking to make the Pearl method objective, quantitative, and reproducible would look to the machine-vision fecal-detection literature and implement US 6,587,575's color-classification pipeline on Pearl's underwear substrate. That is a finite, predictable substitution of a known, better methodology — the essence of an obvious improvement under KSR.

Combination B (single-field, "same problem" framing)

US 6,587,575 (primary) + US 6,913,929 + US 8,073,212.

Here US 6,587,575 is treated as the closest art: it already detects fecal contamination by imaging its color and outputs a size measure. The only meaningful differences are (i) the substrate (carcass → undergarment) and (ii) the end-use metric (contamination reject ↔ cleaning-effectiveness score). US 6,913,929 supplies the reason to move the identical color-imaging technique onto worn underwear, and US 8,073,212 supplies the pixel-color statistical framework plus the express teaching of evaluating "the effectiveness of particular products or regimens." That last phrase is functionally the '206's entire purpose.

Combination C (strongest single-secondary reference)

US 2011/0058717 (Dunavent) as the algorithmic blueprint + US 6,913,929 for the underwear substrate + US 6,587,575 for fecal-color imaging.

US 2011/0058717 discloses image analysis to quantify "percent stain" and "L*a*b* tooth color," using pixel color characteristics, with human-selected regions of interest and cell-level (pixel-level) analysis. Substituting "underwear" for "tooth" and "fecal stain" for "plaque/stain," and adding the color standard and pixel-area calculation, yields claim 1 on its face. Note the inventor overlap — Dunavent is a named inventor on both US 2011/0058717 and the '206 — which is strong evidence that the '206 is the straightforward application of the applicant's existing platform to a new substrate.

---

5. Motivation to combine (the KSR factors)

A POSITA would have been motivated, with a reasonable expectation of success, for several independent reasons:

1. Same problem, same solution type. US 6,913,929 already frames the problem as "is there fecal/lower-colon residue on this underwear?" US 6,587,575 already teaches that the answer is found in the color of fecal material. Combining "underwear as substrate" with "color-imaging fecal detection" is not a leap; it is the direct application of a known detection modality to a known sample.

2. Same field / analogous art. All references concern colorimetric detection and quantification of biological material on a surface — meat/poultry, garments, and oral tissue. Under In re Bigio / KSR, these are reasonably pertinent to the problem and thus proper combinable art.

3. Common ownership and common inventors (very strong motivation). US 8,073,212 and US 2011/0058717 are P&G-owned, and US 2011/0058717 shares inventor Dunavent (and Barker) with the '206. KSR expressly recognizes that a reference in the same company's own portfolio, and design incentives flowing from a known need, motivate combination. A P&G researcher wanting to prove its bath tissue cleans better would naturally reuse P&G's own validated color-imaging analysis platform rather than build a new one.

4. Known technique, predictable result. Each claimed step is an established technique: capture (digital scanning), calibrate (color standard), train on hand-selected points (supervised classification), classify per pixel, count pixels → area. KSR holds that combining known elements with predictable results is obvious. Nothing in the '206 claims a new instrument, a new chemistry, or an unexpected result; the steps are mechanical in the machine-vision sense.

5. Commercial/market pressure. The '206 itself opens by stating the unmet need: a manufacturer would benefit from objectively showing one tissue "is better at cleaning than the other." That incentive is a classic "design incentive in the marketplace" motivation. (I note the patent's own background concedes the manual, visual comparison of two users' underwear was already known — the invention is only the automation/quantification of it.)

---

6. Dependent claims 2–10

These narrow the single independent claim with routine implementation choices and are, if anything, easier to render obvious than claim 1:

• Claim 2 (scanner / camera / image-capture device): an explicit enumeration of known capture devices; US 2011/0058717 literally lists "a camera… a scanner and/or other image capture device."
• Claim 3 (color standard = RGB chart): US 6,587,575's calibration-standards teaching; RGB charts are ubiquitous. The '206 uses Munsell chips — a commercial commodity.
• Claim 4 (points identified by Cartesian coordinates + RGB color): pure data-representation choice; US 2011/0058717's region/point selection + RGB components.
• Claim 5 (Fisher's LDA / logistic regression / QDA): routine statistical classifier selection; US 6,587,575 uses discriminate functions; such classifiers were standard.
• Claim 6 (pixels-per-unit-area × stained pixels): arithmetic; US 2011/0058717's "percent" calculation is the same operation.
• Claim 7 (determine b* of selected points) and Claim 8 (average b* = intensity): selecting the b* yellowness axis is a design choice among a small, finite set of known color dimensions. US 6,587,575 already teaches that fecal material is diagnostically "yellow to green, brown and white," and US 8,073,212 expressly names CIELAB (L*a*b*) as a known color space. Given a stain that is by definition yellowish-brown, choosing b* is the predictable choice. The '206 labels this a "surprising discovery," but the specification's own data (stain onset at b* ≈ 5–7; threshold "> about 7") sit adjacent to the claim-9 threshold of "about 5."
• Claims 9–10 (b* > about 5; multiply by pixel area): a numeric threshold on the same chosen axis plus a multiplication. Absent proof that the specific value "about 5" produces unexpected results across the claimed range, a claimed numerical limitation that overlaps the disclosed/known working range is obvious (In re Woodruff / In re Peterson — overlapping ranges are presumptively obvious).

---

7. Rebuttal considerations and honest caveats

A. Common-ownership could remove two references from § 102(a)(2). The '206 is P&G-owned. US 8,073,212 and US 2011/0058717 are also P&G-owned and were commonly owned with the '206 at the time of filing. Because the '206 has a post-AIA effective filing date (2013-08-13), AIA § 102(b)(2)(C) may disqualify those two as § 102(a)(2) art (owned by the same person). Consequence: a challenger should not build the primary case on those two; they are safest used as (i) § 102(a)(1) published-art evidence, or (ii) background/motivation evidence that pixel-level color classification for "percent stain" was known and was P&G's own approach. The third-party references — US 6,913,929 (Pearl) and US 6,587,575 / US 7,787,111 (USDA) — are not commonly owned and carry no such risk. That is why Combination A is the recommended primary attack. (Whether § 102(b)(1) grace-period or § 102(b)(2)(C) fully neutralizes US 8,073,212 as any category of art is fact-specific and cannot be resolved from the face of the document; treat this as a flag, not a conclusion.)

B. What the prior art does not clearly show. US 6,587,575's detection is built on specific key wavelengths / NIR-fluorescence band ratios, not on an L*a*b* b* transform. The bridge from "key-wavelength reflectance" to "b* yellowness" relies on US 8,073,212 / US 2011/0058717 (which do name CIELAB) and on the routine nature of RGB↔L*a*b* conversion. A patentee could argue the b* insight is the non-obvious core. Counter: (i) claims 1–8 do not require b* at all; (ii) US 6,587,575 already characterizes fecal color as yellow-dominant, making the yellowness axis an obvious selection; (iii) the '206's own conversion (3rd-order polynomial regression, R² ≥ 0.98) is presented as routine statistical modeling, not as an inventive algorithm.

C. Secondary considerations. I found no evidence of unexpected results, long-felt need, commercial success, copying, or industry praise in the record reviewed. The patent is a lapsed, fee-expired internal benchmarking-tool patent (see prior sections) — a posture that tends to weakness on secondary considerations, not strength. If the patentee were to assert secondary considerations, the burden shifts to the challenger to show the nexus (or lack thereof) between the alleged success and the claimed method.

D. Anticipation vs. obviousness. Nothing in § 3 is a single-reference anticipation of claim 1 — the claims are a combination. The § 103 case is therefore built on motivation to combine, which, per § 5, is strong here on multiple independent KSR grounds.

---

8. Bottom line

The '206 claims a pipeline, not a discovery: wear underwear (known — US 6,913,929), image it (known — US 6,587,575, US 7,787,111), include a color target (known calibration), hand-label clean/dirty points and let a standard classifier call every pixel (known — US 8,073,212, US 2011/0058717), then count pixels to get an area/percent (known). The only candidate "invention" — using the b* yellowness axis — is a predictable design choice given that fecal material is concededly yellowish and CIELAB was a named known color space. A POSITA would have combined these with a reasonable expectation of success, motivated by the same problem, the same field, common ownership/inventors, and clear market incentive.

Recommended primary combination for a § 103 challenge: US 6,913,929 + US 6,587,575 (+ US 7,787,111), with US 8,073,212 / US 2011/0058717 offered as corroborating "known-technique" and motivation evidence rather than as § 102(a)(2) art, to sidestep the § 102(b)(2)(C) common-ownership issue.

Confidence: High on the element-by-element mapping and the motivation analysis. Moderate on the exact availability of the two P&G references under AIA § 102(b)(1)/(b)(2)(C), which is a fact question (ownership records and dates) not fully resolved by the page text — flagged above rather than asserted.

---

Sources consulted: the authoritative Google Patents page for US9020206 (https://patents.google.com/patent/US9020206/en) and retrieval of the cited references US 6,913,929 (https://patents.google.com/patent/US6913929), US 6,587,575 (https://patents.google.com/patent/US6587575), US 8,073,212 (https://patents.google.com/patent/US8073212), and US 2011/0058717 (https://patents.google.com/patent/US20110058717). Details attributed to US 7,787,111 and US 7,772,455 are taken from the '206 page's citation titles/assignees only and were not independently re-verified in this session.