High-Precision Detection of Defects of Tire Texture Through X-ray Imaging Based on Local Inverse Difference Moment...

Area of Science:

• Materials Science
• Non-Destructive Testing
• Image Processing

Background:

• Tire quality control is crucial for vehicle safety, but inspecting inner tire structures is challenging.
• Existing methods struggle with detecting internal defects accurately, especially in complex tire textures.
• Automatic defect detection systems are needed for efficient and reliable tire inspection.

Purpose of the Study:

• To propose a high-precision automatic defect detection algorithm for tire texture images using X-ray imaging.
• To develop an effective representation for tire X-ray images using Local Inverse Difference Moment (LIDM) features.
• To enhance the robustness and accuracy of defect detection in tires.

Main Methods:

• Utilized Local Inverse Difference Moment (LIDM) features for effective tire X-ray texture image representation.
• Constructed a Defect Feature Map (DFM) using Hausdorff distance between feature distributions.
• Enhanced the DFM with background suppression for improved robustness.
• Developed a pixel-level defect detection algorithm operating on the enhanced DFM.

Main Results:

• The proposed method achieved high-precision, pixel-level defect detection in tire X-ray images.
• The algorithm demonstrated robustness to background noise and effectiveness in handling various defect shapes.
• Experimental validation confirmed the method's good performance in defect feature mapping and detection.
• Comparative analyses showed superior performance over existing algorithms based on quantitative metrics.

Conclusions:

• The proposed LIDM-based approach offers an effective and robust solution for automatic tire defect detection.
• This method significantly improves the accuracy and reliability of non-destructive tire inspection.
• The algorithm contributes to enhanced tire quality control, ultimately improving vehicle safety.