Tire Profile Interface Recognition for Precise Layer Measurement
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing tire production technologies lack efficient and adaptable methods for quality measurement of tire layers to ensure performance, often requiring expensive and inflexible hardware solutions.
Innovation Solution
A computer-implemented process using neural networks and image analysis to automatically recognize and measure tire product profiles, incorporating a system with a digital profile projector and imaging installation for capturing and training images, enabling flexible and accurate detection of layer interfaces and variations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional hydraulic backscatter sensors or laser scanning systems are used to measure tire product profiles, then measurement precision can be achieved, but device complexity and cost increase significantly
Solution Approach 1:
The patent replaces complex mechanical measurement systems (hydraulic backscatter sensors, laser scanning apparatus) with a neural network-based image analysis system. The neural network processes images of tire product profiles to automatically detect layer interfaces and boundaries, substituting physical measurement hardware with computational intelligence while maintaining measurement precision.
Solution Approach 2:
The patent uses image copies of tire product profiles as input for the neural network. Instead of directly measuring physical dimensions with complex sensors, the system captures images and uses the neural network to analyze and extract measurement information from these visual representations, simplifying the measurement approach.
2Reliability
If specialized measurement hardware is deployed for each tire production environment, then measurement reliability improves, but adaptability to different production environments deteriorates
Solution Approach 1:
The patent creates a universal neural network model that can be deployed across different tire production environments. The model is trained on diverse image data from various production settings, enabling it to adapt to different tire types, layer configurations, and imaging conditions while maintaining consistent measurement reliability without requiring environment-specific hardware customization.
Solution Approach 2:
The neural network dynamically adjusts its processing parameters based on the input image characteristics and production environment. By changing parameters such as threshold values, processing algorithms, and reference profiles, the system maintains reliable measurements across varying production conditions without requiring physical hardware changes.
3Productivity
If manual analysis of tire product profiles is performed, then adaptability to different tire types is maintained, but productivity and measurement precision decrease
Solution Approach 1:
The patent implements a self-service measurement system where the neural network automatically performs image analysis, interface detection, and measurement extraction without requiring manual intervention. The system self-calibrates using reference profiles and automatically adapts to different tire types, eliminating the need for manual analysis while achieving both high productivity and precision.
Solution Approach 2:
The system incorporates feedback mechanisms where measurement results are continuously refined by comparing against reference profiles and ground truth data. The neural network learns from measurement outcomes and adjusts its processing to improve both speed and accuracy, achieving automated high-precision measurements that exceed manual analysis capabilities.
Data Source
AI summary
A computer-implemented process for training a model for automatically recognizing positions in tire product profiles includes calculating an error term with respect to a label assigned in a reference of the interfaces searched in captured images of the tire product profiles that will serve as inputs to the neural network to be trained. A system (100) automatically recognizes interface variations captured in images of samples according to the disclosed process.