US 9891040

I'll search for the specific patent number and check for any litigation dockets.

Let me check specifically for any litigation or docket activity naming patent 9891040.

Analysis of US Patent 9,891,040

Search scope note: I searched for the exact number "9891040" / "9,891,040" and related docket terms. I found no CAFC 2026 docket, IPR, or district-court litigation naming this exact patent. Search hits that surfaced other "…040"-type numbers (e.g., U.S. 9,510,040 in LG Electronics v. Multimedia Technologies, IPR2024-00351, and various Lenovo/InterDigital matters involving U.S. 8,737,933, 9,173,054, 10,250,877, etc.) are different patents and are not results for 9,891,040. I therefore report no known litigation activity for this patent and flag that as a limitation of the search rather than a finding that none exists.

The following is drawn primarily from the authoritative full patent text supplied.

---

Bibliographic Data

---

Abstract

A method for measuring distance in which an image collection unit captures a first picture of an object at a measuring position, a first displaying area of the object is determined, then a second picture is captured at a reference position (different from the measuring position) and a second displaying area is determined. A first actual distance between the measuring and reference positions is obtained, and a second actual distance from the object to the measuring position is computed from the two displaying areas and the first actual distance. A corresponding electronic device is also disclosed.

---

Plain-Language Overview of the Independent Claims

The patent has three independent claims. Claim 1 (method), Claim 6 (electronic device), and Claim 11 (non-transitory computer readable medium). Note that in the granted claims the object is recited as a "fixed object."

Claim 1 — Method (independent)

A processor is used to:

1. Capture a first picture of a fixed object from a measuring position.
2. Determine a first displaying area of the object in that picture (in practice, via edge detection on pixels — see claim 2).
3. Capture a second picture of the same fixed object from a reference position that differs from the measuring position.
4. Determine a second displaying area in the second picture.
5. Obtain the first actual distance between the measuring and reference positions (e.g., via GIS or sensors).
6. Compute the second actual distance from the object to the measuring position from the two areas and the first distance.

Key limitations:

• The measuring position, reference position, and fixed object must be collinear (in a straight line).
• When the two focal lengths are equal, the distance is computed by:

$$L_2 = (-1)^n , L_1 , \frac{C_2}{C_1 - C_2}$$

where L₂ = second actual distance, L₁ = first actual distance, C₁ = first displaying area, C₂ = second displaying area; n is even when the measuring position lies between the reference position and the object, and n is odd when the reference position lies between the measuring position and the object. (The signed exponent is a compact way of encoding the sign/direction of the result for the two geometry cases.)

Claim 6 — Electronic Device (independent)

An electronic device comprising a hardware processor adapted to perform the same sequence of steps as claim 1 (capture first picture at measuring position → first displaying area → capture second picture at reference position → second displaying area → obtain first actual distance → compute second actual distance). It carries the same collinearity limitation and the same equal-focal-length equation with the even/odd-n convention. It is essentially the apparatus counterpart of claim 1.

Claim 11 — Non-Transitory Computer Readable Medium (independent)

A non-transitory computer-readable medium storing a computer program with code means adapted to perform all the steps of claim 1 when run on a computer. It is a standard Beauregard-style medium claim that inherits all of claim 1's limitations (including collinearity and the same equation with the even/odd-n convention).

---

Dependent Claims (for context)

• Claims 2 / 7 — Determining the displaying areas via M / N characteristic pixel points at the object's edges, then computing area from those points (edge detection, e.g., RGB-based).
• Claims 3 / 8 — User-guidance: if the first displaying area is ≥ a first predefined threshold, prompt the user to move backwards; if ≤ a second predefined threshold, prompt the user to move forwards.
• Claims 4 / 9 — Handling different focal lengths by adjusting the two displaying areas to a common focal length before computing distance.
• Claims 5 / 10 — The differing-focal-length equation:

$$L_2 = (-1)^n , L_1 , \frac{f_2 C_2}{f_1 C_2 - f_2 C_1}$$

with the same even/odd-n convention (f₁, f₂ are the two focal lengths).

---

Technical Takeaway

The invention is a single-camera, area-based long-range distance estimator. Rather than stereoscopic/depth-image approaches (which the specification criticizes as needing multiple cameras, calibration, and large camera baselines for long distances), it leverages the known small displacement between two capture positions: an object's apparent (displayed) area shrinks as observed distance grows, so comparing the two areas against the known baseline distance yields the object's distance. The disclosure claims <5% error and requires no camera calibration.

Uncertainty / Caveats

• No litigation found. I could not confirm any CAFC 2026 docket, district-court case, or IPR naming US 9,891,040. This is a negative search result, not proof of absence; a PACER/Docket Navigator query on the exact number would be needed for a definitive answer.
• Legal status ("Active," adjusted expiration 2036-01-21) is per the Google Patents annotation and is not a legal conclusion; maintenance-fee events show the 4th-year (2021) and 8th-year (2025) fees paid.
• There is an apparent typographical error in the granted specification: the equation labeled "(4)" (different focal lengths, reference position between measuring position and object) is mislabelled as "(3)" in the description text. The claims themselves (claim 5) recite the correct differing-focal-length equation with f₂C₂ / (f₁C₂ − f₂C₁). Treat claim language as controlling.
• The granted claims use "fixed object" whereas the pre-grant publication (US 2015/0300814 A1) uses "object"; the independent claims also add the collinearity limitation and the explicit even/odd-n convention that appear more fully in the granted claims.

I'll search for litigation involving US patent 9891040 across patent litigation databases.

Let me search more targeted litigation databases.

Let me search more specifically on patent litigation databases.

Litigation Search Results for US Patent 9,891,040

No litigation was found involving US Patent 9,891,040.

I searched multiple patent-litigation-oriented sources (Unified Patents portal/litigation pages, RPX/insight.rpxcorp.com, Docket Alarm, Justia/PACER-adjacent court filings, and general web queries keyed to the specific number "9891040"). None of the results returned any case in which US 9,891,040 was asserted, challenged, or otherwise placed at issue.

What the searches did return (and why they are not responsive)

The queries surfaced several Lenovo-related litigations, but none of them involve US 9,891,040:

I want to be explicit: several of these hits contained the digit string "9891040" or "9,891,040" purely as coincidental text (e.g., a monetary figure such as "$9,891,040" in an unrelated IPR exhibit, or an unrelated application number). None of these are citations to US Patent No. 9,891,040, and I have not treated them as such.

Caveats on completeness

• Database access limitation: My searches were constrained to web-accessible sources. Unified Patents' litigation case-list pages, RPX Insight docket records, and PACER's full docket text are not fully indexable via open web search, and I was unable to retrieve a case-list entry specifically for this patent number before reaching my search limit. A negative result from these searches is not conclusive proof that no litigation exists.
• Licensing vs. litigation: I found no evidence of assertion, but absence of indexed litigation does not rule out licensing demands, cease-and-desist letters, or ITC proceedings that are not surfaced in these databases.

Identification of the patent (for confirmation)

Per the authoritative patent record provided, the patent at issue is:

• US 9,891,040 B2 ("Method and electronic device for measuring a distance")
• Application No. US 14/560,633; filed 2014-12-04; granted 2018-02-13
• Priority: CN 201410153338.0, filed 2014-04-16
• Assignee: Lenovo (Beijing) Limited
• Status: Active, adjusted expiration 2036-01-21
• Family member: CN105091847B

Bottom line

To the best of my ability given the available search results, there is no known litigation involving US Patent 9,891,040. The patent appears to be held by Lenovo (Beijing) Limited and remains active, but I found no record of it being asserted in any district court, the ITC, or the PTAB. I recommend confirming this conclusion directly against a fee-based docket service (e.g., PACER, Docket Navigator, or Lex Machina) before relying on it, since a definitive negative requires those sources.

I'll verify the structured data against live sources before drafting, since the "no proceedings" block is the default but web search can surface unindexed filings.

Search results confirm the structured data — no PTAB filings surfaced. Let me run two more targeted checks before I finalize.

Proceedings overview

Total AIA trial proceedings on US 9,891,040: 0. The structured ODP "PTAB proceedings on file" block returns no IPR, PGR, or CBM for this patent, and independent web checks (PTAB E2E / PTABCTS petition documents, Unified Patents portal, general docket searches) surfaced no petition, no institution decision, no FWD, and no PTAB appeal naming US 9,891,040 or its application number 14/560,633. There is accordingly no breakdown by status to report — the categories (active / claims invalidated / claims sustained / settled / institution denied) are all empty. Bottom-line defensive posture: the patent is UNTESTED before the PTAB. All 11 original claims remain as issued, so a defendant gets no free cancellation argument — but also no adverse estoppel, and no precedent FWD construing the claims against the patent owner. This is a "clean slate" patent, not a "hardened" one.

A caveat on evidence quality: the absence of PTAB activity is sourced from the ODP block (authoritative, most-recent ingest) plus negative web searches. I cannot rule out a very recently filed petition that has not been ingested, but nothing in the public record as of 2026-09-29 suggests one exists.

---

No proceeding sections follow, because there are no proceedings to report. Per the task constraints, I am not fabricating docket numbers to fill the template.

---

Strategic summary

Claim status: all 11 claims of US 9,891,040 are UNTESTED — none canceled, none sustained, none adjudicated. The claim set breaks down as: claims 1–5 (method), claims 6–10 (electronic device including a hardware processor), claim 11 (non-transitory computer readable medium). Claim 1 is the sole independent method claim and the anchor of the set — it requires collecting a first picture at a measuring position, determining a first displaying area, collecting a second picture at a reference position, determining a second displaying area, obtaining the first actual distance between positions, and computing L₂ via L₂ = (−1)ⁿ L₁ C₂ / (C₁ − C₂) with n even when the measuring position is between the reference position and the fixed object, and n odd in the reverse geometry. Claims 4–5 and 9–10 add the variable-focal-length variant using f₁/f₂. Claim 11 depends from claim 1 in substance. Because no tribunal has construed any term, "fixed object," "displaying area," and the even/odd-n branch logic are all unconstrained — that cuts both ways.

Estoppel landscape: nothing is foreclosed yet, and nothing is available from anyone else's work. § 315(e)(2) estoppel only attaches to a petitioner that actually reached a final written decision, and no petitioner has. So a defendant today faces no estoppel as a respondent, and equally gains no benefit from a prior petitioner's forfeited grounds. Every invalidity theory — § 102, § 103, and § 112 — remains on the table in district court and would be available for a fresh IPR petition. The one real constraint is timing: this patent's adjusted expiration is 2036-01-21 (filed 2014-12-04, priority 2014-04-16), so it has roughly a decade of life left and is worth attacking on the merits. Note the patent dates to a 2014 priority date — a 2014-04-16 CN priority (CN 201410153338.0, granted as CN105091847B). Prior art must predate that, but the field (monocular/single-camera range estimation from apparent object size) has dense pre-2014 art, and the seven references already cited on the face of the patent — US 2003/0091228 (Honda), US 2005/0261849 (Topcon), CN 103075998 (monocular space target distance/angle measurement, 华中科技大学), CN 103499334 (Xiaomi distance measurement), CN 101876532, CN 102095469, and US 2014/0267775 (Lablans) — are a ready starting kit.

Pattern signals: none present. No serial petitioner has targeted this patent; there is no Unified Patents or defensive-aggregator activity; the patent owner (Lenovo (Beijing) Limited) has neither appealed anything to the Federal Circuit nor been forced to defend the claims. The maintenance-fee record shows a live, maintained asset — 4th-year fee paid 2021-07-27, 8th-year fee paid 2025-07-29 — i.e., the owner is paying to keep it alive, which is mildly consistent with it being held for assertion, but there is no evidence in the record of an active assertion campaign. The absence of any litigation or IPR is itself the notable fact: patents that are widely asserted tend to accumulate IPRs, and this one has not.

Recommended next steps

• If you are a defendant and have received a demand citing this patent: do not expect a prior FWD to help you, and do not assume the patent is weak. There is no PTAB record to cite, and — critically — there is also no PTAB record against you. Your invalidity case must be built from scratch.
• Consider filing an IPR yourself. With zero prior petitions, no § 315(b) bar and no § 325(d) "same or substantially the same art" problem exists (nothing has been presented to the Office beyond the seven prosecution citations), and no § 314(a)/Fintiv discretionary-denial track record has been established for this patent. The claim-1 equation and the "fixed object" / "displaying area" limitations are the natural § 112 and § 103 attack surfaces — particularly whether the display-area-to-distance proportionality math is enabled and whether "fixed object" has antecedent-basis and definiteness support across the specification.
• Note the trial-stage clock if you file. PTAB must issue a Final Written Decision within 12 months of institution under § 316(a)(11). Institution decision is due within 6 months of the petition's filing under § 314(b). After an FWD, appeal to the Federal Circuit would run under §§ 141–144 — docket it via CourtListener (https://www.courtlistener.com) and the Federal Circuit's own docket once it exists.
• Monitor the Chinese counterpart. CN105091847B (granted 2017-12-29) is the family member; CNIPA invalidation proceedings are not AIA trials and would not appear in the ODP block, but a Chinese invalidation outcome can be persuasive evidence about the prior art. I could not confirm any CNIPA challenge from the available sources — treat that as unknown, not as absent.
• Verify independently before relying on this memo. Confirm the empty-state on PTAB E2E (https://ptacts.uspto.gov) by patent number and by application 14/560,633, and re-check after the next ODP ingest in case a petition has been filed and not yet indexed.

I'll start by verifying the assignment record and looking for any post-issuance transfers, litigation, or NPE activity tied to US 9891040.

Let me check for litigation, any post-issuance transfers, and the current status of the original assignee.

Let me do a final check for any litigation or transfer activity involving this specific patent.

Assignment Record Reconstruction — US 9,891,040 B2

Patent: US 9,891,040 B2 — "Method and electronic device for measuring a distance"
Application: 14/560,633 · Filed: 2014-12-04 · Issued: 2018-02-13
Priority: CN 201410153338.0, filed 2014-04-16 (CN 105091847 B)
Current assignee of record: LENOVO (BEIJING) LIMITED

Data-source note: The Google Patents legal-events table for US 9,891,040 (fetch reproduced in the source text) lists only one recorded assignment — the inventor-to-company transfer — followed by the grant and two maintenance-fee payments. No post-issuance assignment, security agreement, merger, or change-of-name record appears. I was unable to load the USPTO Assignment Center detail view within the tool budget; the reel/frame, dates, assignor, and assignee below are taken from the assignment abstract embedded in the patent record. The "correspondent of record" field is not exposed in any source I could retrieve, and I will not invent one.

---

Inventors

Pattern note: None of the fire-sale precursors apply on the inventor side. This is a standard single-inventor corporate work-for-hire fact pattern: one inventor, rights assigned to the employer before the US filing date (executed 2014-11-25, US application filed 2014-12-04). There is no evidence of inventor departure, co-inventor split, or independent inventor retention of rights that would precede a portfolio liquidation. The CN priority application and the US filing share the same sole inventor and the same assignee, so no ownership fragmentation occurred at prosecution.

---

Original assignee

LENOVO (BEIJING) LIMITED (联想（北京）有限公司), Haidian District, Beijing, China.

• Relationship to issue patent: Named original assignee on the issued US 9,891,040 B2 and on the granted CN 105091847 B family member.
• Product embodying the claims: The patent claims a camera-based method for computing object distance from two pictures taken at two positions plus a known baseline (claims 1, 6, 11). Lenovo ships camera-equipped smartphones, tablets and notebooks, and the specification expressly frames the electronic device as "a smart mobile phone, a tablet computer and a digital camera." However, I found no public evidence that any specific Lenovo product ships a feature practicing the claims, so product-embodiment is plausible but unconfirmed.
• Primary line of business: Personal computers, smartphones (incl. the Motorola Mobility line), tablets, servers and data-center hardware. Lenovo (Beijing) Limited is the PRC operating/R&D subsidiary of Lenovo Group Limited (HKEX: 0992), the world's largest PC vendor by unit share.
• Current status: Operating. Lenovo Group is an active, publicly traded manufacturer. Maintenance fees on this patent were paid for the 4th year (recorded 2021-07-27) and the 8th year (recorded 2025-07-29), with the patent shown Active and adjusted expiration 2036-01-21. This is a still-live, still-maintained corporate asset — the opposite of an abandoned asset available for scrap.

---

Assignment timeline

Chronological list of every recorded assignment:

• 2014-11-25 (executed) / recorded 2014-12-04 — Reel 034378/0719
  • Conveyance: Assignment (ASSIGNMENT OF ASSIGNORS INTEREST; SEE DOCUMENT FOR DETAILS)
  • Assignor: XU, LI (individual inventor)
  • Assignee: LENOVO (BEIJING) LIMITED
  • Correspondent: Not exposed in the retrieved record — could not be confirmed. No assertion made.
  • Context: Standard inventor-to-employer transfer of rights, executed 9 days before the US national filing and ~5 months before grant of the corresponding CN priority case. This is a routine corporate intake, not an acquisition, fire-sale, reorg or securitization.

No further assignments are recorded. The chain terminates at the original assignee. This is itself the finding: because there is no post-issuance assignment, Lenovo (Beijing) Limited still owns US 9,891,040 B2.

---

Timeline diagram

timeline
    title Ownership of US 9891040
    2014 : CN priority filed by Lenovo Beijing
         : US application filed
         : Inventor Xu Li assigns to Lenovo Beijing
    2018 : US 9891040 issued
    2021 : Maintenance fee paid
    2025 : Maintenance fee paid

---

NPE / troll-pattern signals

---

Verdict

Insufficient data — no records beyond the original assignment.

The complete ownership chain for US 9,891,040 B2 is a single link: inventor Li Xu → Lenovo (Beijing) Limited, executed 2014-11-25 and recorded 2014-12-04 at Reel 034378/0719. There are no subsequent recorded assignments, no shell-entity transfer, no known asserter, no cascading transfers, and no bankruptcy event; the patent is still Active with the 8th-year maintenance fee paid on 2025-07-29 (adjusted expiration 2036-01-21), which is consistent with a manufacturing company retaining its own asset rather than a divestiture or assertion vehicle. Because the assignee is a product-shipping operating company rather than a licensing entity, this is not an NPE chain — but with only the original assignment on record, the correct characterization is "insufficient data," not an affirmative anti-NPE finding.

Verification: USPTO Assignment Center — https://assignmentcenter.uspto.gov/ (search by patent number 9891040), mirrored at https://assignment.uspto.gov/patent/index.html.

Caveat: I could not open the Assignment Center's underlying page image or correspondent field for Reel 034378/0719. If the reel/abstract is later transcribed with a correspondent name, that field should be re-checked — but with a single assignment and no downstream transfers, correspondence analysis cannot generate an NPE signal regardless.

I'll search for the patent and its cited references to build the prior-art analysis.

I have good detail on several references. Let me gather the remaining cited references.

Prior-Art Analysis — US Patent No. 9,891,040

Scope note: I searched for the exact identifier 9891040 and worked from the authoritative full text you supplied. The patent text lists seven (7) "Patent Citations" (references cited on the face of the patent). I analyze each below. Two of the Chinese references (CN101876532A, CN102095469A) are characterized from the authoritative table plus what I could retrieve; I flag where detail is limited. Everything below is a preliminary technical analysis, not a legal conclusion — "potentially anticipates" means "warrants a § 102 look," not "does anticipate."

Governing date (AIA): US 9,891,040 has an effective filing date of 2014-04-16 (CN priority 201410153338.0) and was filed 2014-12-04. Because it was filed after March 16, 2013, the AIA version of 35 U.S.C. § 102 applies. Prior art therefore includes: (a) patents/printed publications publicly available before 2014-04-16; and (b) U.S. patents, U.S. application publications, and WIPO publications effectively filed before 2014-04-16.

Independent claims at issue: Claim 1 (method), Claim 6 (electronic device), Claim 11 (non-transitory CRM). All three require (i) two pictures of a fixed object from a measuring position and a different reference position; (ii) a first displaying area and second displaying area; (iii) a first actual distance between the two positions; (iv) computing the second actual distance from the areas + first distance; (v) collinearity of the two positions and the object; and (vi) the equal-focal-length equation L₂ = (−1)ⁿ L₁ · C₂/(C₁ − C₂) with the even/odd-n convention. Claims 4/9/5/10 add different-focal-length handling with L₂ = (−1)ⁿ L₁ · f₂C₂/(f₁C₂ − f₂C₁).

---

The Seven Cited References (with § 102 potential)

---

1. US 2003/0091228 A1 — "Image recognition apparatus" (Honda Giken Kogyo Kabushiki Kaisha)

• Citation / dates: US 2003/0091228 A1; priority 2001-11-09; published 2003-05-15.
• Description: A vehicle-mounted image-recognition apparatus for detecting pedestrians/obstacles. It uses two infrared cameras (2R, 2L) with parallel optical axes and computes parallax by correlation between the left/right images to determine whether objects overlap and to issue collision warnings. It is a multi-camera stereovision system.
• § 102 potential: No anticipation. It teaches two-camera parallax, the very paradigm the '040 specification criticizes and distinguishes ("two (or more) cameras are needed… however, in the present embodiment, only one camera is used"). It does not disclose (i) a single image-collection unit capturing two pictures at two positions, (ii) a "displaying area" comparison, or (iii) the claimed equation. Relevant only as background/§ 103 context for "distance from images."

2. US 2005/0261849 A1 — "Image calibration method, image calibration processing device, and image calibration processing terminal" (Topcon Corporation)

• Citation / dates: US 2005/0261849 A1 (from PCT/JP03/11899; granted as US 7,715,644 B2); priority 2002-09-19; published 2005-11-24. Inventors Kochi & Noma.
• Description: A client/server camera-calibration method that corrects lens distortion of a general camera so that precise 2D measurement / 3D measurement via the stereo method can be performed. The processing device receives a calibration-object image, computes calibration results (principal point, distortion data, focal length), and returns a corrected subject image. Notably, it receives focal length and camera model / imaging-device size as inputs and contemplates 3D measurement using the corrected image.
• § 102 potential: No anticipation. It is calibration-centric and relies on the stereo method (multiple images/parallax) for 3D measurement. It does not disclose two-position area-of-object ranging. It is, however, at most § 103-relevant because it shows the state of the art regarding use of focal-length parameters in image-based measurement (which relates to dependent claims 4/5/9/10). It does not touch claim 1.

3. CN 101876532 A — "Camera Field Calibration Method in Measuring System" (Dalian University of Technology)

• Citation / dates: CN 101876532 A; priority 2010-05-25; published 2010-11-03; assignee 大连理工大学 (Dalian University of Technology).
• Description: A camera field-calibration method for a measurement system — i.e., calibrating a camera's parameters in the field so that subsequent image-based measurement is accurate. (Characterization limited; the reference is calibration-directed.)
• § 102 potential: No anticipation. Calibration art; no teaching of single-camera two-position displaying-area ranging, the collinearity limitation, or the claimed equations. Background only.

4. CN 102095469 A — "Device and method for measuring liquid level in storage tank by utilizing camera" (Shenyang Aerospace University)

• Citation / dates: CN 102095469 A; priority 2010-12-04; published 2011-06-15; assignee 沈阳航空航天大学.
• Description: A camera-based liquid-level measurement device/method for a storage tank.
• § 102 potential: No anticipation. The problem solved (liquid level) and the measured quantity differ from distance-to-a-fixed-object via area comparison between two positions. No overlap with the claim-1 limitations. Background only.

5. CN 103075998 A — "Monocular space target distance-measuring and angle-measuring method" (Huazhong University of Science & Technology)

• Citation / dates: CN 103075998 A; filed 2012-12-31 (application CN201210594277.2); published 2013-05-01; assignee 华中科技大学; inventors 张天序 et al.
• Description: A single-camera (monocular) method for measuring the distance and attitude angle of a space target. Steps: (1) receive one frame of the target; (2) preprocess (Otsu threshold segmentation, morphological operations, labeling) to extract the target from background; (3) feature-extraction and pose recognition against a template feature library; (4) compute the target's 3D centroid distance Δp and attitude angle using the target's known 3D structure/size and geometric imaging relations (involving camera focal length focslen and pixel size pixlen). Notably suited to far targets (>100 m) with no markers, wide range, high precision.
• § 102 potential: Closest single-camera ranging art among the citations, but no anticipation. It computes distance from one frame using a known target geometric dimension — not from comparing two displaying areas obtained at two collinear positions with a measured baseline distance, and it does not disclose the claimed equation or the even/odd-n convention. Strong § 103 candidate when combined with a two-position capture scheme, but it does not disclose every limitation of claim 1/6/11 as arranged.

6. CN 103499334 A — "Distance measurement method, apparatus and electronic instrument" (Xiaomi Technology Co., Ltd.)

• Citation / dates: CN 103499334 A (申请号 CN201310400156.4); filed 2013-09-05; published 2014-01-08; assignee 小米科技有限责任公司; later granted CN 103499334 B (2016-08-17). Inventors 王博, 许斐, 相里飞.
• Description: Obtains a first image of a reference object at a detection position, obtains the size information of the reference object in the image, and computes the distance between the detection position and the reference object from the reference object's standard (known) size versus its image size. Used for horizontal distance or vertical height; not location-restricted. (I also note this reference was later used as X-category art against subsequent filings — see the WO 2022/042275 A1 search report — underscoring its relevance in this art.)
• § 102 potential: No anticipation of the '040 independent claims. It uses a single capture position and relies on the object's known/standard size — not two positions and not a comparison of two displaying areas. This is a meaningful difference: claim 1 expressly requires a first and second displaying area plus the measured baseline L₁. At most § 103 relevant.

7. US 2014/0267775 A1 — "Camera in a Headframe for Object Tracking" (Peter Lablans)

• Citation / dates: US 2014/0267775 A1; priority 2013-03-15; published 2014-09-18; inventor Peter Lablans.
• Description: A head-mounted camera (in a headframe) used for object tracking (e.g., eye/head-based tracking applications).
• § 102 potential: No anticipation. Object tracking with a wearable camera; no two-position displaying-area ranging, no baseline distance, no claimed equation. Date-eligible under § 102(a)(2) (effective filing 2013-03-15 predates 2014-04-16) but substantively remote. Background only.

---

Bottom Line on the Cited Art

• No single one of the seven cited references appears to anticipate any claim of US 9,891,040 — least of all independent claims 1, 6, 11, each of which requires the combination of (two capture positions + two displaying areas + a measured inter-position baseline + the specified equation + collinearity).
• The two references with the greatest substantive proximity are CN 103075998 A (monocular, far-range distance measurement using target geometry) and CN 103499334 A (distance from object/image size). Both are better characterized as § 103 (obviousness) references, singly or in combination, than as § 102 anticipatory art — because each uses only one capture position / one image rather than the claimed two-position area comparison with a measured baseline.
• The remaining five (Honda stereo recognition; Topcon and Dalian calibration; Shenyang liquid-level; Lablans headframe tracking) are background/§ 103 context at best and do not disclose the distinguishing limitations.

---

Additional Observation (flagged separately — NOT among the 7 citations)

While retrieving the citations, I incidentally encountered US 2009/0003654 A1 ("Single-aperture passive rangefinder"), which describes forming a first image at a first position and a second image at a second position with a single imaging system and determining range based on the separation between the two positions and the point-of-interest location. That reference is conceptually much closer to the two-position, single-camera concept of claim 1 than any of the seven cited references. I flag it because it appears not to have been cited on the face of US 9,891,040 and may warrant a closer § 102/§ 103 review — but I did not obtain its full text within this search, so I cannot confirm whether it discloses the displaying-area comparison or the claimed equations, and I do not treat it as a confirmed anticipation.

---

Caveats and Confidence

• Source priority: The seven citations, their dates, assignees, and titles are taken from the authoritative patent text you supplied (Google Patents citation table) and cross-checked with web results where available (Honda, Topcon, Huazhong, Xiaomi confirmed).
• Detail limits: I could not fully retrieve the technical descriptions of CN 101876532 A and CN 102095469 A before exhausting my search steps (two queries returned a step-limit message). Those two characterizations are accordingly lower-confidence; a full-text review is recommended.
• Anticipation is claim-chart-specific: A definitive anticipation call requires mapping each claim limitation (including the equation and the collinearity/even-odd-n convention) to express disclosure in a single reference. On the material available, none of the seven cited references does so for the independent claims.
• Not legal advice / not a legal conclusion. This is a preliminary technical mapping, not an invalidity opinion.

I'll ground this § 103 analysis in the actual prior-art references cited on the patent record. Let me research the key references to state their disclosures accurately rather than relying on memory.

I have strong grounding on the key references. Let me verify the disclosure specifics of two more cited references and check for any reference that expressly teaches two-position size-change ranging.

Obviousness Analysis — US 9,891,040 B2 under 35 U.S.C. § 103

Built on the prior sections. This analysis uses the claim constructions, bibliographic data, and technical framing already established above (single-camera, area-based, two-position long-range estimator; the "fixed object" and collinearity limitations; the even/odd-n convention). It does not repeat them. The prior-art universe analyzed here is the "Prior Art" section of the Google Patents record, i.e. the Patent Citations (7) list — US 2003/0091228 A1; US 2005/0261849 A1; CN 101876532 A; CN 102095469 A; CN 103075998 A; CN 103499334 A; US 2014/0267775 A1 — plus the stereoscopic depth-image method admitted as prior art in the '040 background.

Threshold flag on the art set: The "Families Citing this family (8)" block (CN106170064A, CN109460690B, CN107799174A, CN108225278A, CN108717724A, CN110555873A, CN111339808B, CN113358090A) is not prior art — every one has a priority date of 2016 or later, after the '040 priority date of 2014‑04‑16. I have excluded them from the combination analysis.

---

1. Legal framework applied

Because the '040 has a CN priority of 2014‑04‑16 and was filed 2014‑12‑04, it is an AIA patent, so § 103 is governed by AIA § 103 / § 102(a)(1)–(a)(2). The governing test is Graham v. John Deere (scope/content of prior art; differences; PHOSITA level; secondary considerations), as refined by KSR Int'l v. Teleflex, 550 U.S. 398 (2007): a claim is obvious where the elements are found in the prior art and there is a reason to combine — including "known technique," "combination of familiar elements according to known methods," "a finite number of identified, predictable solutions," design incentives/market forces, and the PHOSITA's creativity. The level of ordinary skill here is a person with a bachelor's in EE/CS/optics and ~2–3 years of camera-based measurement or mobile imaging experience.

Critically for this patent, the "invention" is a closed-form application of the elementary perspective relationship that apparent image size is inversely proportional to object distance. A claim that recites a mathematical relationship between measured quantities, combined with conventional image-capture and area-measurement steps, is highly vulnerable to § 103 where the physical relationship is known or predictable.

---

2. Claim 1 decomposed (limitation chart)

The only elements that could be argued to be non-conventional are C+D used jointly with F/H (two-position area comparison with the closed-form solution). Elements A, B, C, D (basic), E, G are individually conventional; H is a rearrangement of the standard inverse-distance/apparent-size law. The obviousness question therefore reduces to: was it obvious, given the art, to take image-size (area) measurements from two known-separated positions and solve the resulting two-equation system for object distance?

---

3. Grounded disclosures of the cited references

3.1 CN 103499334 A — Xiaomi (filed 2013‑09‑05; published 2014‑01‑08) — the closest art

This is the single most material reference and is squarely on an electronic device. Its abstract (confirmed in two independent databases and in the title record) states the method:

"在检测位置获取参考物的第一图像；获取所述第一图像内参考物的尺寸信息；根据参考物的标准尺寸信息和所述第一图像内参考物的尺寸信息，获取检测位置与参考物之间的距离."
("obtain a first image of a reference object at a detection position; obtain size information of the reference object within the first image; based on the reference object's standard size information and the size information in the first image, obtain the distance between the detection position and the reference object.")

It is classified G01C3/00 — the same primary class as the '040. Its disclosed purpose is measuring horizontal distance or standing height, "不受限制" (unrestricted location), "方便简单" (convenient and simple). It is a single-camera, image-size-based, distance-on-a-phone teaching. Its independent claim 1 is a distance-measuring method; claim 2 derives height data from the computed distance.
Sources: agri.nais.net.cn patent detail; tianyancha record; tmhub CPC listing (URLs in the search results above).

Materiality corroboration: In the ISR/search report of WO 2022/042275 A1, CN 103499334 A is cited as "X" category (novelty-destroying/single-reference) against claims 1–15 of a distance-measurement application (description ¶¶[0084]–[0170], figs. 1–19). That establishes the reference as a recognized, highly material teaching of camera-based distance measurement — though those claims (digital-zoom-to-distance mapping) differ from claim 1 here.

Difference from claim 1: CN 103499334 A needs the object's known/standard size and uses one image; claim 1 needs neither the object's true size nor any reference dimension — it derives distance purely from the ratio of two areas and the known baseline between two capture positions.

3.2 CN 103075998 A — Huazhong Univ. of Science & Technology (published 2013‑05‑01) — monocular ranging

"一种单目空间目标测距测角方法" — a single-camera (monocular) method to measure distance and attitude angle of a target by geometric imaging, using only the target's 3D structure and size information, with no cooperative markers on the target, and expressly characterized in its abstract as having "wide distance-measuring range and high precision." Its background expressly criticizes the "corresponding-point calibration" monocular method because it "is only applicable to the case where the camera position is fixed" and requires re-calibration whenever a camera parameter changes, and notes the prior marker-based method "does not apply" at longer ranges (100 m+). Classified G01C3/00.
Source: Google Patents CN103075998A/en; tianyancha claim text; Eureka/Patsnap abstract.

This reference supplies (i) the single-camera, size/structure-based ranging premise and (ii) an explicit motivation to avoid calibration and free camera movement — the '040 stresses both themes.

3.3 US 2003/0091228 A1 — Honda (published 2003‑05‑15) — object detection/distance from images

An image-recognition apparatus that detects a target object in images, computes an image-space size of the object (the Justia assignee listing for the Honda family describes: "An image processing unit determines the size of a circumscribed quadrangle of the target object…"), and uses image shift (parallax) between camera images to compute the distance to the target object. The described hardware is a vehicle computer with two infrared cameras and outputs warnings/HUD.
Sources: Google Patents US20030091228; Justia Honda assignee listing; EP 2278574 A1 (Honda) specification quoting the same bounding-box size approach.

Relevance: Honda grounds two propositions a PHOSITA would take for granted — (a) image-space size of a detected object is a measurable proxy for its distance, and (b) a known spatial separation between image-acquisition points (here camera baseline) plus an image-space measurement yields object distance. Honda also concerns tracking a target across successive images, which foreshadows the "same object in two pictures" element.

3.4 US 2005/0261849 A1 — Topcon (published 2005‑11‑24) — image calibration / focal length handling

Discloses correcting camera/lens distortion for image-based dimensional measurement and states the device "receiv[es] at least a focal length in the photographing and a model of the camera, or a focal length in the photographing and a size of an imaging device … together with the calibration object image …", and that "the results of the calibration process include at least one of a principal point position, distortion data, and a focal length of a lens used in the photographing." It also notes that "when an image photographed two-dimensionally is used to measure dimensions of an object, it is necessary to correct the photographed image so as to remove image distortions."
Source: Google Patents US20050261849A1/en; USPTO.report text; patents-review.com record.

Relevance: directly supports the different-focal-length dependent claims (4/5, 9/10) — Topcon teaches using the focal length at which an image was photographed as a normalization parameter, which is exactly the adjustment the '040 requires.

3.5 CN 101876532 A — Dalian Univ. of Technology (published 2010‑11‑03)

"Camera Field Calibration Method in Measuring System" (measurement-camera calibration). By its title and classification it teaches calibrating a camera used in a measurement system — i.e., establishing the mapping between image-space quantities and real-world quantities. (Full text not retrieved; description limited to the record's title/classification — flagged as an unverified characterization.)

3.6 CN 102095469 A — Shenyang Aerospace Univ. (published 2011‑06‑15)

"Device and method for measuring liquid level in a storage tank by utilizing a camera" — a camera-based measurement in which an image-derived quantity yields a physical dimension/level. (Full text not retrieved; description limited to title/classification.) General evidentiary value: camera-image quantities used to compute real-world measurements.

3.7 US 2014/0267775 A1 — Lablans (filed 2013‑03‑15; published 2014‑09‑18) — headframe object tracking

"Camera in a Headframe for Object Tracking." Publication postdates the '040 priority, but it was effectively filed 2013‑03‑15, so it can qualify as § 102(a)(2) art (US application publication, different inventor, effectively filed before the '040 effective filing date). Subject matter is object tracking with a worn camera; I could not confirm it discloses two-position area-based ranging. Treated as low-relevance secondary art and flagged as unverified.

3.8 Admitted prior art — stereoscopic depth-image ranging

The '040 background expressly admits that two cameras at different "capturing angles" produce a depth image used to find the object distance, and that long-range use requires an impractically large inter-camera distance (meters, for 1000 m range) that a phone cannot provide. This admission is prior art under § 102(a)(1)/§ 103 and frames the problem the '040 addresses.

---

4. Primary obviousness theory

Combination 1 — CN 103499334 A (Xiaomi) + CN 103075998 A (HUST monocular) [optionally + US 2003/0091228 A1 (Honda)]

What the combination supplies, element by element:

The gap the combination must bridge is narrow: substituting "distance derived from one image + known object size" (Xiaomi) with "distance derived from two images + known baseline" — a substitution that (as shown in §5) eliminates the reference-object-size requirement and the calibration constraint, which is exactly the problem the art was trying to solve.

Combination 2 — CN 103499334 A + the admitted stereoscopic art (+ Honda)

The admitted stereo method already teaches capturing two images of the same object from two known-separated positions and computing distance from the image difference. It differs from claim 1 only in the image feature exploited (pixel parallax in stereo vs. displaying-area difference here). Since Xiaomi teaches that displayed size is itself a distance proxy, a PHOSITA seeking to perform single-camera ranging would find it an obvious substitution to replace the baseline-proportional parallax cue with the baseline-and-size-change cue — both are standard triangulation parameters and the same baseline L₁ is the free variable. This is a textbook KSR "known technique / predictable substitution" scenario.

Combination 3 — CN 103499334 A + Honda + Topcon (for focal-length claims)

Adds Topcon whenever the two pictures are taken at different focal lengths: Topcon teaches recording/using focal length at photographing to correct image-based dimension measurement (US 2005/0261849 A1). Applying a focal-length normalization to convert two areas to a common focal length (claim 4/9) is then a routine camera-calibration step.

---

5. Motivation to combine (the KSR rationales)

A PHOSITA would have been motivated to combine the above with a reasonable expectation of success for several independent, mutually reinforcing reasons:

1. Same field, same problem, same classification. CN 103499334 A, CN 103075998 A, CN 102095469 A and the '040 all sit in G01C3/00 (measuring distances in line of sight). CN 103499334 A is expressly about a distance-measuring method on an electronic device — the identical context. Combining references in the same field aimed at the same problem is the paradigm case for obviousness.
2. The admitted stereo art names the very problem and the very variables. The '040's background states that the two-camera method fails for long distances because the required camera separation grows impractically large for a portable device. Given that explicit teaching, a PHOSITA would look for a way to reuse a large, user-controlled separation instead of a fixed inter-camera baseline — leading directly to "the user walks between two positions L₁ apart," which is claim element E/G. The '040 itself confirms the operation is trivial: "it only needs to move … backwards to 0.5‑1 meter" for a 1000 m object.
3. The physics is elementary and the equation is a predictable algebraic result. Apparent image size scales inversely with object distance (simple perspective projection). With two known distances (L₂ and L₂±L₁) and two measured areas, the single unknown L₂ is over-determined and the closed form L₂ = (−1)ⁿ L₁·C₂/(C₁ − C₂) follows by one elimination step. Under KSR, where the result is "predictable," a mere rearrangement of a known mathematical relationship incorporated into a process is obvious. Honda's own family derivation (Ghv = Ghd0·Z0/(Z0+ΔZ); Z1 = Z0 + ΔZ) is the same relationship expressed for linear size, confirming the formula was within routine skill.
4. Design/market drivers for single-camera smartphone ranging. CN 103075998 A supplies the express motivation to use one camera (simple structure, simple calibration, avoids stereo-matching and small-FOV pitfalls); Xiaomi supplies the motivation to put distance measurement in a consumer electronic device. Cost, size, and portability of phones are classic design incentives (KSR), and the '040 admits it "saves the number of the cameras and simplifies the hardware."
5. Removal of known obstacles yields a predictable advantage. Xiaomi's one-image method requires knowing the object's standard size; CN 103075998 A's method requires the target's known 3D structure and (per its background) is hostile to moving/unknown targets. Using two size measurements at a known baseline cancels the unknown object size — an obvious and desirable simplification a PHOSITA would pursue. This is precisely the "finite number of identified, predictable solutions" situation.
6. The user-guidance feature is a routine design choice. Prompting the user to step forward/backward until the object's apparent area is within a preferred band (claim 3/8) is ordinary UI/ergonomics once the two-position method is chosen; it merely ensures the object lies between, or beyond, the two positions so the geometry is well-conditioned.

---

6. Dependent-claim obviousness

• Claims 2 / 7 (M/N edge pixel points → area): Edge/contour detection to delimit an object's image region is routine image processing (KSR "known technique"), and Honda already computes a circumscribed quadrangle of the detected object — an area delimitation. The '040's specific RGB edge-detection example is a conventional implementation detail.
• Claims 3 / 8 (forward/backward user prompts at thresholds): Routine user-guidance/UX; the thresholds ("≥80% of display," "≤20%") are printed-matter-like numeric design choices with no asserted criticality beyond conditioning the geometry.
• Claims 4 / 9 (adjust areas to a common focal length): Directly supported by Topcon US 2005/0261849 A1, which teaches capturing and using the focal length at photographing when measuring object dimensions from images, and by the universal fact that image size scales with focal length. Normalizing two images to a common focal length is a standard step.
• Claims 5 / 10 (the differing-focal-length equation): Same analysis as claim 1 element H — a predictable algebraic combination of the inverse-size law with the relative focal-length ratio (C₁/C₂ ≈ (f₁/f₂)·(L₂/(L₂±L₁))). Obvious once the equal-focal-length form is obvious.

If claim 1 is obvious, dependent claims 2–5 (and apparatus/media counterparts 6–11) fall with it, absent a showing of unexpected results — which the '040 does not provide beyond an asserted "error rate < 5%" that is a mere accuracy statement, not a difference in kind.

---

7. Counterarguments and weaknesses in the obviousness case

To be candid, a strong rebuttal exists and should be weighed:

1. No single cited reference expressly discloses the two-position, two-area method. None of the seven cited references, on the text I could verify, teaches taking two pictures of the same object from two user-selected positions and comparing the two displaying areas. The § 103 case therefore depends on combining a size-based single-image method (Xiaomi) with the general scale-change/parallax principle (Honda/stereo). A patent owner would argue this is impermissible hindsight reconstruction from the '040's own disclosure.
2. The closest "size-change" teaching (Honda's ΔZ change-rate ranging) is not in the cited list for '040. I identified it in the Honda family (e.g., EP 2278574 / CN 102985957), not in the '040's citation record. If an examiner must rely on it, its provability as prior art must be independently established. Within the cited set, US 2003/0091228 A1 itself is parallax-based, not magnitude-of-size-change-based — so the express "compare two sizes over a known baseline" teaching is weaker than it first appears. Flagged as the principal vulnerability.
3. Motivation is inferential. The "remove dependence on known object size" motivation is reconstructed from the references' stated problems; the references do not state it. Under KSR this is usually sufficient (the PHOSITA is presumed to know the art), but it is not the strongest form of motivation.
4. The "fixed object" and collinearity limitations are not clearly taught as such in the cited art; they would be argued to be arbitrary narrowing. (They are nonetheless conventional measurement constraints.)
5. Secondary considerations are absent from the intrinsic record (no unexpected results, licensing, or praise) — the "<5% error" assertion cuts against non-obviousness by framing the result as an expected accuracy level rather than a surprising one.

---

8. Bottom line

On the art in the '040's own "Prior Art" section, claim 1 is, more likely than not, obvious under § 103. The combination of CN 103499334 A (Xiaomi) — single-camera, image-size-based distance measurement on an electronic device — with CN 103075998 A (HUST) — monocular size/geometry-based ranging expressly motivated to avoid calibration and fixed-camera constraints — and, for the object-size/distance and known-baseline elements, US 2003/0091228 A1 (Honda), renders the two-position, two-area method an expected, predictable application of the elementary inverse-size/perspective law. The claimed closed-form equation is a routine algebraic rearrangement of that law (cf. Honda's ΔZ-based derivation), and the dependent features map onto Topcon US 2005/0261849 A1 (focal-length handling) and ordinary image-processing/UX practice.

The strongest weakness is that no cited reference expressly discloses the two-image, two-displaying-area comparison; that express teaching appears in the broader Honda family (EP 2278574 / CN 102985957), which is not in the '040 citation record and would need to be independently qualified as prior art. I have flagged this rather than assert a stronger position than the evidence supports.

Confidence: Moderate-high that claim 1 is obvious as a combination; low confidence that any single one of the seven cited references alone anticipates or renders claim 1 obvious. CN 101876532 A, CN 102095469 A, and US 2014/0267775 A1 were characterized only from their record titles/abstracts (full text not retrieved) and should be verified before being relied upon in a formal office action.