Neural Pith Location Estimation from Four-Face Timber Board Images
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for accurately determining the pith location in timber boards are limited by assumptions about growth ring concentricity and require manual parameter adjustments, leading to inefficiencies and inaccuracies, especially in industrial applications.
Innovation Solution
A computer-implemented method using optical scanning and artificial neural networks, specifically conditional generative adversarial networks (cGANs) and multilayer perceptrons (MLPs), to estimate pith location in timber boards by analyzing longitudinal surfaces, without requiring pre-processing and with improved accuracy and speed.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional methods using assumptions about growth ring concentricity are used, then measurement process is simple, but measurement precision deteriorates
Solution Approach 1:
The patent replaces conventional mechanical/optical scanning methods with artificial intelligence-based image analysis. The AI model automatically detects growth rings and calculates pith location from standard digital images, eliminating the need for complex pre-processing and manual parameter adjustments while achieving higher precision than traditional methods.
Solution Approach 2:
The patent changes the approach from assuming fixed geometric parameters (concentric growth rings) to using AI-based parameter detection. The system learns optimal parameters from training data and adapts to actual wood structure variations, improving measurement accuracy without requiring complex manual calibration.
2Measurement precision
If manual parameter adjustments are made for accurate pith location, then measurement precision improves, but productivity deteriorates
Solution Approach 1:
The patent implements self-service through automated AI-based analysis that performs parameter detection and pith location calculation without manual intervention. The system automatically processes standard digital images, eliminating the need for operators to manually adjust parameters while maintaining high accuracy, thus improving both precision and productivity.
Solution Approach 2:
The patent applies preliminary action by pre-training the AI model with extensive training data before deployment. This preliminary training enables the model to automatically recognize growth ring patterns and calculate pith locations accurately without requiring manual parameter adjustments during actual processing, achieving high speed and accuracy simultaneously.
3Measurement precision
If pre-processing steps are applied to improve measurement accuracy, then measurement precision improves, but device complexity increases
Solution Approach 1:
The patent replaces complex pre-processing steps with AI-based direct analysis of standard digital images. The AI model inherently handles image variations and extracts growth ring information without requiring manual pre-processing operations, simplifying the system while maintaining or improving measurement precision.
4Measurement precision
If conventional scanning methods are used, then ease of operation is maintained, but measurement precision deteriorates
Solution Approach 1:
The patent implements self-service through automated AI analysis that eliminates the need for manual parameter adjustments and pre-processing steps. The system automatically processes standard digital images and outputs accurate pith locations, maintaining ease of operation while dramatically improving measurement precision compared to conventional methods.
Data Source
AI summary
A computer-implemented method for estimating a pith location with regard to a timber board, including:receiving a pixelated actual digital image of each lateral face of at least a longitudinal part of the timber board, extending along a longitudinal axis of the timber board;identifying an input portion in said longitudinal part of the timber board, where the input portion is a portion of the timber board which extends along the longitudinal axis;extracting from each pixelated actual digital image of the longitudinal part of the timber board, an input image representing said input portion, so obtaining four input images representing an appearance of the input portion at each lateral face of the timber board;inputting said four input images into the input layer of an artificial neural network and making the artificial neural network operate; andreading, at an output layer of the artificial neural network, output data defining a location of a pith of a log from which the timber board has been obtained, in a plane perpendicular to the longitudinal axis of the timber board at the input portion.