DCT

3:18-cv-11542

NCR Corp v. Teraoka Seiko Co Ltd

Log In

I. Executive Summary and Procedural Information

  • Parties & Counsel:
  • Case Identification: 3:18-cv-11542, D.N.J., 07/11/2018
  • Venue Allegations: Venue is alleged to be proper as Defendant Digi America, Inc. is incorporated in New Jersey and has a regular and established place of business in the district. Venue over foreign Defendant Teraoka Seiko Co. Ltd. is asserted based on its business activities and shipment of accused products into the district.
  • Core Dispute: Plaintiff alleges that Defendant’s produce weighing scales and electronic shelf label systems infringe six patents related to automated produce recognition and electronic display management.
  • Technical Context: The technologies at issue relate to retail automation, specifically systems for identifying items without barcodes (like produce) and for dynamically updating pricing information on shelves, both of which are central to modern retail efficiency.
  • Key Procedural History: The complaint does not mention any prior litigation, inter partes review proceedings, or licensing history related to the patents-in-suit.

Case Timeline

Date Event
2000-09-20 Priority Date for U.S. Patent No. 6,412,694
2002-01-11 Priority Date for U.S. Patent Nos. 6,877,133; 7,340,412; 7,007,219; 6,976,206
2002-07-02 U.S. Patent No. 6,412,694 Issues
2005-04-05 U.S. Patent No. 6,877,133 Issues
2005-12-13 U.S. Patent No. 6,976,206 Issues
2006-02-28 U.S. Patent No. 7,007,219 Issues
2008-03-04 U.S. Patent No. 7,340,412 Issues
2010-07-01 Priority Date for U.S. Patent No. 8,136,727
2012-03-20 U.S. Patent No. 8,136,727 Issues
2018-07-11 Complaint Filed

II. Technology and Patent(s)-in-Suit Analysis

U.S. Patent No. 6,412,694 - Produce Recognition System and Method Including Weighted Rankings

Issued July 2, 2002.

The Invention Explained

  • Problem Addressed: The patent addresses the inefficiency and high error rate of manual produce identification at checkout counters, and notes that automated methods relying solely on a single data source like spectral analysis can be unreliable and may not consistently rank the correct item highest (’694 Patent, col. 1:33-56).
  • The Patented Solution: The invention proposes a two-stage recognition method. First, the system generates an initial list of possible produce items with rankings based on collected data (e.g., spectral analysis) (’694 Patent, col. 2:15-17). Second, it applies a "Difficulty Index" or "weights" to this list. These weights are derived from historical data, specifically how often a candidate item was correctly selected by an operator and the screen position where it was displayed in past events. This re-orders the list to produce a final, more accurate ranking that is more likely to place the correct item in the top position (’694 Patent, col. 7:51-62; Abstract).
  • Technical Importance: This approach aimed to improve the usability and accuracy of automated checkout systems by learning from past user interactions, making the system "smarter" over time.

Key Claims at a Glance

  • The complaint asserts independent method claim 1 (Compl. ¶22).
  • The essential elements of claim 1 are:
    • Collecting produce data from a produce item.
    • Comparing the data to reference data to get a list of candidates with initial (first) rankings.
    • Determining weights for candidates based on how many times they were previously correct identifications and were displayed in specific screen locations.
    • Applying these weights to the candidates to get second rankings.
    • Displaying the candidates in order of their second rankings.
    • Recording the operator's selection.
  • The complaint does not explicitly reserve the right to assert dependent claims for this patent.

U.S. Patent No. 8,136,727 - Produce Weighing Scale with a Camera and Methods of Operating a Produce Weighing Scale Having a Camera

Issued March 20, 2012.

The Invention Explained

  • Problem Addressed: The patent background describes drawbacks of prior art retail systems, including the need for a store clerk to weigh produce or the customer being unaware of the item's price until reaching a checkout station, which complicates self-service shopping (’727 Patent, col. 1:12-45).
  • The Patented Solution: The patent describes a self-service produce weighing scale intended for use in the produce aisle, separate from the main checkout area. The invention is defined by a specific physical structure: a base with a weighing platform, a first support arm extending vertically, and a second support arm extending horizontally over the platform. An "actuatable picture camera" is mounted at the end of the second arm to capture an image of the produce, and internal circuitry communicates with a self-service checkout station elsewhere in the store (’727 Patent, col. 13:8-43; Fig. 2).
  • Technical Importance: This design facilitates a "scan-as-you-go" shopping model, allowing customers to weigh, price, and label produce items themselves before proceeding to a final payment point.

Key Claims at a Glance

  • The complaint asserts independent apparatus claim 5 (Compl. ¶39).
  • The essential elements of claim 5 are:
    • A produce weighing scale located in a produce section, remote from the store exit.
    • A base and a weighing platform.
    • A user interface for customer interaction.
    • A first support arm extending "substantially vertically" from the base.
    • A second support arm extending "substantially transverse" to the first arm and "substantially horizontally" over the platform.
    • An actuatable camera at the distal end of the second arm, positioned over the platform.
    • Electronic control circuitry with a communications circuit (for networking with a checkout station) and a processor (for actuating the camera).
  • The complaint does not explicitly reserve the right to assert dependent claims for this patent.

U.S. Patent No. 6,877,133 - Methods and Apparatus for Error Detection and Correction in an Electronic Shelf Label System

Issued April 5, 2005.

  • Technology Synopsis: The patent addresses data errors in Electronic Shelf Label (ESL) registers. The invention is a method where a host computer sends a "bedcheck message" containing a "sumcheck" (a form of checksum) to an ESL. The ESL compares this with a sumcheck of its own data; if they match, it sends a positive acknowledgement, and if not, a negative one, triggering a data correction from the host (’133 Patent, Abstract).
  • Asserted Claims: At least claim 1 is asserted (Compl. ¶53).
  • Accused Features: The Accused ESL System is alleged to perform this sumcheck-and-acknowledgement method to automatically detect and correct errors (Compl. ¶55).

U.S. Patent No. 7,340,412 - Methods and Apparatus for Performing Delta Updates of an Electronic Shelf Label

Issued March 4, 2008.

  • Technology Synopsis: The patent targets inefficient ESL updates where the entire data set is re-sent even for minor changes. The invention provides a "delta update" method where the host system compares the current data on an ESL with the planned new data, creates a third data image containing only the changes, and sends this smaller "delta" package to the ESL, conserving bandwidth and battery life (’412 Patent, Abstract).
  • Asserted Claims: At least claim 1 is asserted (Compl. ¶72).
  • Accused Features: The Accused ESL System is alleged to perform these delta updates, for example, by changing only the price on an ESL (Compl. ¶74). The complaint references FIG. 8, which illustrates the broadcast of product content such as price to an ESL (Compl. ¶77).

U.S. Patent No. 6,976,206 - Methods and Apparatus for Intelligent Data Bedcheck of an Electronic Shelf Label

Issued December 13, 2005.

  • Technology Synopsis: The patent addresses the inefficiency of performing a "full data bedcheck" on all of an ESL's registers. It proposes an "intelligent bedcheck" that verifies only the subset of registers currently being used to control the information displayed on the ESL, making the verification process faster and more efficient (’206 Patent, Abstract).
  • Asserted Claims: At least claim 1 is asserted (Compl. ¶86).
  • Accused Features: The Accused ESL System is alleged to perform this intelligent bedcheck to ensure "100% integrity" on price changes by verifying only the relevant subset of data (Compl. ¶¶88, 92).

U.S. Patent No. 7,007,219 - Methods and Apparatus for Error Detection and Correction of an Electronic Shelf Label System Communication Error

Issued February 28, 2006.

  • Technology Synopsis: This patent addresses communication errors, particularly "false positives" where a host believes a message was successfully received when it was not. The invention is a multi-step communication protocol: after sending an initial message and receiving a positive acknowledgement, the host sends a second, verification message to confirm the ESL's contents, ensuring the initial acknowledgement was valid before logging the transaction as successful (’219 Patent, Abstract).
  • Asserted Claims: At least claim 1 is asserted (Compl. ¶100).
  • Accused Features: The Accused ESL System is alleged to use a multi-step protocol involving initial messages and acknowledgements to ensure "100% integrity" on price changes (Compl. ¶102). The complaint includes a diagram, FIG. 8, showing a three-step data broadcast process that includes an acknowledgement from the ESL back to the RF station (Compl. ¶¶103-104).

III. The Accused Instrumentality

Product Identification

  • The complaint names two instrumentalities: the SM-5600BS Camera Scale ("Accused Scale") and the T@POP Electronic Shelf Label System ("Accused ESL System") (Compl. ¶¶22, 53).

Functionality and Market Context

  • Accused Scale: The complaint describes the SM-5600BS as a produce scale that "uses a built-in camera to identify fruits and vegetables" (Compl. ¶25). It allegedly employs "advanced artificial intelligence" to "learn and improve accuracy with increased usage," performing "intelligent colour and shape recognition" to narrow down item choices for the user (Compl. ¶¶26, 28). The product is marketed as a self-service "solution for [a] produce area" that allows a user to "[g]et a label for a quick checkout" (Compl. ¶41).
  • Accused ESL System: The T@POP system is described as a solution for displaying and updating prices on electronic tags in a retail environment (Compl. ¶53). The complaint highlights its use of "2-way digital communication" to ensure "100% integrity on price change" and alleges it performs functions like error correction, delta updates, and intelligent bedchecks to maintain data accuracy while managing bandwidth and battery life (Compl. ¶¶55, 58-59, 74, 88).

IV. Analysis of Infringement Allegations

'694 Patent Infringement Allegations

Claim Element (from Independent Claim 1) Alleged Infringing Functionality Complaint Citation Patent Citation
collecting produce data from the produce item; The Accused Scale uses a "built-in camera" to collect data from items via "intelligent colour and shape recognition" (Compl. ¶26, FIG. 2). ¶26 col. 6:15-17
comparing the produce data to reference produce data to obtain a list of candidate identifications with first rankings based upon probability of match; The Accused Scale compares collected data to reference data to obtain a list of candidates, narrowing down the selection to the "closest match" for user presentation (Compl. ¶27). ¶27 col. 6:17-20
determining weights for the candidate identifications based upon numbers of times the candidate identifications were correct identifications and were displayed in predetermined operator selection screen locations during past recognition events; On information and belief, the Accused Scale's "advanced artificial intelligence" system learns from increased usage, thereby determining updated weights based on past correct identifications and their screen locations (Compl. ¶28). ¶28 col. 6:20-25
applying the weights to the candidate identifications to obtain second rankings; On information and belief, the "advanced artificial intelligence learning system" applies the determined weights to obtain second rankings to provide the "closest match" (Compl. ¶29). ¶29 col. 6:25-26
displaying the candidate identifications in order of the second rankings; and The Accused Scale displays the candidate identifications resulting from the AI learning process to provide the operator the "closest match" for selection (Compl. ¶30, FIG. 3). ¶30 col. 6:27-28
recording an operator selection of one of the candidate identifications. The Accused Scale records the operator's selection of an "icon" from the displayed candidates (Compl. ¶31). ¶31 col. 6:29-30
  • Identified Points of Contention:
    • Technical Question: The central dispute will likely be whether the Defendants' "advanced artificial intelligence" performs the specific, multi-step weighting and re-ranking process required by claim 1. The complaint relies on high-level marketing claims of "learning" to infer these steps, raising the question of what evidence exists that the accused AI's algorithm matches the patent's detailed method for determining and applying weights based on past selections and screen locations.

'727 Patent Infringement Allegations

Claim Element (from Independent Claim 5) Alleged Infringing Functionality Complaint Citation Patent Citation
A produce weighing scale in a produce section which is remote from a store exit section of a retail grocery store... The Accused Scale is described in product literature as a "solution for [a] produce area" that allows a user to get a label for checkout, implying a location remote from the exit (Compl. ¶41). ¶41 col. 13:8-11
a base; a weighing platform disposed on the base; The Accused Scale includes a base and a weighing platform, as shown in a product diagram (Compl. ¶42, FIG. 4). ¶42 col. 13:12-13
a user interface arranged to enable a retail customer... to interact with the produce weighing scale; The Accused Scale includes a user interface where a customer can "touch the picture" to interact with the device (Compl. ¶43, FIG. 5). FIG. 5 is a product image with the User Interface labeled. ¶43 col. 13:14-17
a first support arm which extends substantially vertically from the base; a second support arm which extends substantially transverse relative to the first support arm and which extends substantially horizontally relative to the weighing platform; The Accused Scale includes a first vertical support arm and a second horizontal/transverse support arm, as identified in a provided product schematic (Compl. ¶¶44-45, FIG. 4). The schematic, FIG. 4, explicitly labels these two arms. ¶¶44-45 col. 13:18-22
an actuatable picture camera disposed at a distal end of the second support arm and positioned over the weighing platform... The Accused Scale "uses a built-in camera" located at the end of the second support arm over the weighing platform to identify produce (Compl. ¶46). ¶46 col. 13:23-28
electronic control circuitry including (i) a communications circuit which can electronically communicate over a store communications network with a... self-service checkout pay station... and (ii) an electronic processor... arranged to... actuate the picture camera... On information and belief, the Accused Scale includes an "Ethernet 10/100 Base T" communications circuit and an "Intel Core 2 Duo T7400" processor that actuates the camera and communicates over a network (Compl. ¶¶47-48). ¶¶47-48 col. 13:29-43
  • Identified Points of Contention:
    • Scope Questions: While the structural mapping appears strong, a potential point of contention may arise from claim terms requiring interpretation, such as whether the accused arms are "substantially" vertical and horizontal. A further question relates to the preamble, which is asserted as a limitation: is the Accused Scale used "in a produce section which is remote from a store exit section," and what evidence will be required to prove this intended use environment?

V. Key Claim Terms for Construction

  • Term from '694 Patent, claim 1: "determining weights for the candidate identifications based upon numbers of times the candidate identifications were correct identifications and were displayed in predetermined operator selection screen locations during past recognition events"

    • Context and Importance: This term is the central inventive concept of the '694 patent. The entire infringement allegation hinges on whether the defendant's "advanced artificial intelligence" performs this specific, explicitly defined function. The construction of this term will determine whether a general-purpose learning algorithm can be found to infringe this highly specific method step.
    • Intrinsic Evidence for Interpretation:
      • Evidence for a Broader Interpretation: A party might argue that any system that "learns" from past correct user selections and uses that information to adjust future rankings falls within the scope, focusing on the general concept of using historical data.
      • Evidence for a Narrower Interpretation: The specification provides a detailed formula for a "Difficulty Index D" which is calculated based on the number of times a correct choice appeared in specific display locations (N₁ through Nₙ) (’694 Patent, col. 7:35-50, eq. 2). A party could argue that this specific implementation defines and limits the scope of "determining weights," requiring an accused system to track not just correctness but also the screen position of past selections.

  • Term from '727 Patent, claim 5: "a first support arm which extends substantially vertically from the base; a second support arm which extends substantially transverse relative to the first support arm and which extends substantially horizontally relative to the weighing platform"

    • Context and Importance: This structural limitation defines the physical configuration of the patented scale. Practitioners may focus on this term because the infringement analysis for this apparatus claim depends on a direct comparison of physical structures. The defendant may challenge whether its product's arms meet the specific geometric and relational requirements of being "substantially vertically," "substantially transverse," and "substantially horizontally."
    • Intrinsic Evidence for Interpretation:
      • Evidence for a Broader Interpretation: A party could argue "substantially" allows for reasonable deviation from perfect 90-degree angles, encompassing any generally L-shaped arm configuration that positions a camera over a scale.
      • Evidence for a Narrower Interpretation: The patent figures (e.g., FIG. 2) consistently depict a distinct, two-part structure with clear vertical and horizontal components. A party could argue that these embodiments limit the term to configurations that are not, for example, a single curved arm or an arm angled significantly away from vertical or horizontal.

VI. Other Allegations

  • Indirect Infringement: The complaint alleges inducement of infringement for all six patents. The allegations are based on Defendants providing product literature and instruction manuals that allegedly instruct and encourage customers and end-users to operate the Accused Scale and Accused ESL System in a manner that performs the claimed methods (Compl. ¶¶ 34, 67, 81, 95, 111).
  • Willful Infringement: Willfulness is alleged for all patents. The complaint asserts that Defendants had knowledge of their infringing activities "at least from the time of receipt of the complaint" (Compl. ¶¶ 34, 67, 81, 95, 111) and prays for treble damages (Compl. p. 36).

VII. Analyst’s Conclusion: Key Questions for the Case

  1. A central issue will be one of functional equivalence: does the accused "advanced artificial intelligence" in the produce scale perform the specific, multi-factor weighting-and-reranking method recited in the '694 patent, or is there a technical mismatch between the patent's prescribed algorithm and the accused product's learning mechanism? This extends to the ESL patents, where the court will examine whether the accused system's error-checking and update protocols perform the precise, detailed steps required by the claims.

  2. A second key question will be one of structural and contextual scope: does the physical construction of the Accused Scale, particularly its two-arm support structure, map onto the specific geometric limitations of the '727 patent's claims? Furthermore, can the plaintiff prove that the scale is intended for the specific use environment recited in the claim preamble—"in a produce section which is remote from a store exit section"—thereby distinguishing it from scales integrated into checkout counters?