PTAB

PGR2025-00081

NicholSon Mfg Ltd v. Bid Group Technologies Ltd

Log In
Key Events
Petition
petition

1. Case Identification

2. Patent Overview

  • Title: System for Real-Time Characterization and Debarking of Logs Using AI
  • Brief Description: The ’947 patent describes a system and method for characterizing undebarked logs in real-time using a contactless scanning device and a trained deep learning model. The system computes and sends optimal operating parameters to a debarker based on the identified log characteristics to automate and improve the debarking process.

3. Grounds for Unpatentability

Ground 1: Obviousness over Gagnon and Kambla - Claims 1-4 and 7-10 are obvious over Gagnon in view of Kambla and the knowledge of a POSITA.

  • Prior Art Relied Upon: Gagnon (WO 2013/188948) and Kambla (Application # 2019/0003829).
  • Core Argument for this Ground:
    • Prior Art Mapping: Petitioner argued that Gagnon disclosed a debarking system that adjusts operating parameters based on intrinsic log characteristics such as wood species and diameter. Kambla taught using a non-contact camera and a trained neural network classifier to identify these same characteristics from images of undebarked logs. Petitioner asserted that the combination of Gagnon’s control system with Kambla’s automated characterization method meets the key limitations of independent claim 1. Specifically, Kambla provided the "contactless characterization" and the "trained deep learning model," while Gagnon provided the system for "computing operating parameters" and "sending" them to the debarker.
    • Motivation to Combine: A Person of Ordinary Skill in the Art (POSITA) would combine Kambla’s automated, camera-based log characterization with Gagnon’s debarking control system to improve the accuracy and efficiency of Gagnon’s system. Automating the identification of tree species and diameter—parameters Gagnon explicitly identifies as important—using Kambla’s teachings represented a predictable improvement over the manual or less-advanced input methods suggested by Gagnon.
    • Expectation of Success: A POSITA would have had a high expectation of success, as the combination involved integrating a known image analysis technique (Kambla) to provide input data to a known industrial control system (Gagnon).

Ground 2: Obviousness over Gagnon, Ido, and Petruzella - Claims 1-18 are obvious over Gagnon in view of Ido and Petruzella.

  • Prior Art Relied Upon: Gagnon (WO 2013/188948), Ido (a 2018 conference paper on CNN-based tree species identification), and Petruzella (a 2017 textbook on Programmable Logic Controllers).
  • Core Argument for this Ground:
    • Prior Art Mapping: This ground presented an alternative combination to arrive at the claimed invention. Gagnon provided the base debarking system controlled by a Programmable Logic Controller (PLC). Ido taught the specific use of a Convolutional Neural Network (CNN), a type of deep learning model, to identify tree species from bark images captured by a camera. Petruzella provided foundational teachings on the conventional structure and programming of PLCs, explaining how they receive inputs and use them to retrieve or calculate output parameters from memory tables or arrays. The combination of these references allegedly taught all elements of the challenged claims.
    • Motivation to Combine: A POSITA would combine Ido’s CNN-based species identification with Gagnon’s system to automate a critical input step. Further, a POSITA would look to a standard textbook like Petruzella to implement the logic within Gagnon’s PLC. This would involve using the species data from Ido's model to index a lookup table of pre-set debarker parameters, which Petruzella described as a standard PLC function, thereby improving the system's speed and reliability.
    • Expectation of Success: The proposed integration was a predictable application of known technologies. A POSITA would expect that using a known image classification model (Ido) to provide input to a standard industrial controller (Gagnon's PLC, as detailed in Petruzella) would function as intended.

Ground 3: Unpatentable Subject Matter under §101 - Claims 1-18 are directed to an abstract idea.

  • Prior Art Relied Upon: Not applicable.

  • Core Argument for this Ground:

    • Alice Step 1 (Abstract Idea): Petitioner argued the claims are directed to the abstract idea of collecting data (log characteristics), analyzing that data using mathematical techniques (a generic AI model), and generating an output based on the analysis (debarker parameters). Petitioner contended this is merely an automation of a fundamental economic practice and mental process previously subject to human judgment, performed using conventional computer components. The patent fails to disclose a specific improvement to computer functionality or the debarking process itself, instead focusing on the abstract result.
    • Alice Step 2 (Inventive Concept): Petitioner asserted the claims lack an inventive concept sufficient to transform the abstract idea into a patent-eligible application. The claims recite only generic and conventional elements, such as a "scanning device," a "trained deep learning model," and a general-purpose computer/controller, without any specific, unconventional implementation. The claim language simply instructs a POSITA to "apply" the abstract idea of using AI to the known field of log debarking, which is insufficient under Alice. The use of an indexed table was also characterized as a longstanding, conventional data storage method.

  • Additional Grounds: Petitioner asserted an additional obviousness challenge based on the combination of Gagnon, Kambla, and Petruzella. Petitioner also challenged claims 7 and 8 as indefinite under §112(b) for failing to provide objective boundaries for the term "predetermined value."

4. Relief Requested

  • Petitioner requests the institution of a post-grant review and the cancellation of claims 1-18 of Patent 12,163,947 as unpatentable.