Tire Surface Anomaly Detection via Gradient Orientation Filtering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Automated detection of blowholes on tire surfaces is challenging due to their geometric characteristics, making it difficult for even experienced operators to locate them during visual inspections, and existing image processing methods struggle to distinguish blowholes from other surface features like markings and deformations.
Innovation Solution
A method involving 3D image processing that captures elevation gradients, filters orientations using a digital filter to highlight blowhole-like structures, and combines this with elevation filtering to identify potential blowholes, while correcting for false detections by reducing gray levels at marking points and averaging elevations within neighborhoods to enhance detection accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If automated image processing methods are used to detect blowholes, then productivity is improved, but measurement precision deteriorates due to difficulty in distinguishing blowholes from other surface features
Solution Approach 1:
The inspection process is divided into multiple independent processing stages: gradient calculation, orientation filtering, elevation filtering, and combined analysis. Each stage processes specific features separately before integration, allowing the system to maintain high processing speed while achieving precise blowhole detection through systematic decomposition of the complex detection task.
Solution Approach 2:
The method transitions from analyzing only elevation data to incorporating orientation information of elevation gradients as an additional dimension. By calculating gradient orientations and applying orientation filtering, the system creates a new analytical dimension that distinguishes blowholes from other surface features, thereby improving measurement precision without sacrificing productivity.
2Device complexity
If simple elevation thresholding is used, then device complexity is reduced, but measurement precision deteriorates due to false detections from markings and deformations
Solution Approach 1:
Before applying elevation thresholding, the method performs preliminary processing steps including gradient calculation and orientation filtering. These preliminary actions prepare the data by emphasizing blowhole characteristics and suppressing marking-related signals, so that the subsequent simple thresholding operates on pre-conditioned data, maintaining precision without requiring complex final processing.
Solution Approach 2:
The gradient orientation filter acts as an intermediary between raw elevation data and final thresholding. It transforms the elevation information into orientation-based representations that highlight blowhole structures while suppressing markings, serving as a mediator that enables simple thresholding to achieve high precision detection.
3Measurement precision
If detailed visual inspection by operators is performed, then measurement precision is improved, but productivity deteriorates due to time-consuming manual inspection
Solution Approach 1:
The method replaces manual visual inspection with an automated optical-mechanical system that captures 3D surface data and processes it through algorithmic analysis. The automated system performs gradient calculation, orientation filtering, and elevation analysis that mimics and enhances human inspection capabilities, achieving both high precision and high throughput by substituting human operators with automated image processing.
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
Effectively highlights areas likely to contain blowholes, reducing false detections and improving the accuracy of blowhole identification by leveraging the unique gradient orientation patterns and elevation characteristics specific to blowholes.
Implementation Method 1
The three-dimensional image of the surface can be obtained using known means, based on the principle of optical triangulation, and using sensors coupled to light sources such as laser sources.
Data Source
Figure 1~4
Figure 5~9
AI summary
The invention relates to a method for analysing the three-dimensional digital image of the surface of a tire to be inspected in which the image is captured of the three-dimensional elevations of said surface while assigning each point of the surface represented by a pixel of the image a grey level proportional to its elevation relative to said surface, characterized in that: - from that image of the elevations, an image is formed of the orientation of the elevation gradients of the surface in which each point is assigned a grey level value proportional to the angle formed with a direction given by the projection in the plane of the image of a non-zero norm vector substantially corresponding, at that point, to the gradient vector tangent to the surface and oriented in the direction of the steepest slope. - a filtered image of the orientations is determined by converting the image of the orientation of the elevation gradients using a digital filter able to select the zones including structures similar to the reference image of the orientation of the elevation gradients with the steepest slope of a blister.