Tire Tread Thermal Image Segmentation for Pixel Attribution
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods struggle to accurately attribute pixel values in thermal IR images of tire tread patterns to specific components like ribs, grooves, or background, hindering understanding of thermal evolution during driving conditions.
Innovation Solution
A computer-implemented method using a convolutional neural network (CNN) model with atrous convolution and Gabor filters to segment tire tread patterns in thermal footprint images, enabling precise identification of components like ribs and grooves.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Temperature
If thermal IR imaging is used to capture tire footprint, then temperature measurement capability is improved, but ability to attribute pixel values to specific tread components deteriorates
Solution Approach 1:
The patent applies image segmentation to divide the thermal footprint image into distinct regions corresponding to different tread components (ribs, grooves, shoulders). By segmenting the image based on thermal patterns and geometric features, the system attributes pixel values to specific components, resolving the information loss problem while preserving temperature measurement capability
Solution Approach 2:
The patent introduces an intermediary processing system that includes edge detection algorithms and pattern recognition modules. This intermediary layer processes the raw thermal image to identify component boundaries and characteristics, enabling pixel attribution without directly modifying the thermal measurement process itself
2Measurement precision
If high-resolution thermal imaging is used, then temperature measurement precision is improved, but image processing complexity increases
Solution Approach 1:
The patent segments the high-resolution thermal image into manageable regions corresponding to tread components. This segmentation reduces the computational burden by processing smaller, focused regions rather than the entire high-resolution image at once, while maintaining measurement precision through careful boundary detection
Solution Approach 2:
The patent extracts key features and characteristics from the high-resolution thermal image, such as edge patterns, thermal gradients, and component boundaries. By extracting only the essential information needed for component attribution, the system reduces processing complexity while preserving the precision benefits of high-resolution imaging
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
The method provides accurate segmentation of tire tread components, allowing attribution of pixel values and insight into thermal evolution, facilitating improved tread design for enhanced traction and handling.
Implementation Method 1
High-resolution thermal infrared, IR, imaging is a non-contact technology that converts the radiation pattern of an object within the range of [3 μm-14μm] into a visible image in the range of [400 nm-700 nm], which is proportional to the emitted infrared energy
Data Source
AI summary
The invention provides a computer-implemented method and a system for automatically segmenting thermal images of a tire's footprint in order to identify components of the tire's tread pattern. The invention relies on a convolutional neural network model and on digital image pre-filtering. It is capable of accurately segmenting thermal footprint images of tires in either straight rolling or cornering conditions, which allows for the accurate extraction of temperature data for each identified tread pattern component.