Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Defect Detection in Tire X-Ray Images: Conventional Methods Meet Deep Structures

Andrei Cozma, Landon Harris, Hairong Qi, Ping Ji, Wenpeng Guo, Song Yuan

TL;DR

The study addresses automated defect detection in tire X-ray images, a challenging task due to tiny, anisotropic defects and imbalanced data. It proposes a hybrid pipeline that combines traditional texture and frequency features (LBP, GLCM, Fourier, Wavelet) with a Random Forest classifier and augments YOLOv8 with engineered features, complemented by a probability-map ensemble to produce defect heatmaps. Key findings show that the GLCM, Fourier, and Wavelet feature set often outperforms the YOLO baseline and that a well-tuned feature ensemble can match or exceed deep-model performance in this domain, demonstrating the value of blending classical image analysis with modern deep learning for industrial QA. The work advances practical tire quality assurance by offering a real-time, hybrid framework that leverages both established texture analysis techniques and contemporary detection methods.

Abstract

This paper introduces a robust approach for automated defect detection in tire X-ray images by harnessing traditional feature extraction methods such as Local Binary Pattern (LBP) and Gray Level Co-Occurrence Matrix (GLCM) features, as well as Fourier and Wavelet-based features, complemented by advanced machine learning techniques. Recognizing the challenges inherent in the complex patterns and textures of tire X-ray images, the study emphasizes the significance of feature engineering to enhance the performance of defect detection systems. By meticulously integrating combinations of these features with a Random Forest (RF) classifier and comparing them against advanced models like YOLOv8, the research not only benchmarks the performance of traditional features in defect detection but also explores the synergy between classical and modern approaches. The experimental results demonstrate that these traditional features, when fine-tuned and combined with machine learning models, can significantly improve the accuracy and reliability of tire defect detection, aiming to set a new standard in automated quality assurance in tire manufacturing.

Defect Detection in Tire X-Ray Images: Conventional Methods Meet Deep Structures

TL;DR

The study addresses automated defect detection in tire X-ray images, a challenging task due to tiny, anisotropic defects and imbalanced data. It proposes a hybrid pipeline that combines traditional texture and frequency features (LBP, GLCM, Fourier, Wavelet) with a Random Forest classifier and augments YOLOv8 with engineered features, complemented by a probability-map ensemble to produce defect heatmaps. Key findings show that the GLCM, Fourier, and Wavelet feature set often outperforms the YOLO baseline and that a well-tuned feature ensemble can match or exceed deep-model performance in this domain, demonstrating the value of blending classical image analysis with modern deep learning for industrial QA. The work advances practical tire quality assurance by offering a real-time, hybrid framework that leverages both established texture analysis techniques and contemporary detection methods.

Abstract

This paper introduces a robust approach for automated defect detection in tire X-ray images by harnessing traditional feature extraction methods such as Local Binary Pattern (LBP) and Gray Level Co-Occurrence Matrix (GLCM) features, as well as Fourier and Wavelet-based features, complemented by advanced machine learning techniques. Recognizing the challenges inherent in the complex patterns and textures of tire X-ray images, the study emphasizes the significance of feature engineering to enhance the performance of defect detection systems. By meticulously integrating combinations of these features with a Random Forest (RF) classifier and comparing them against advanced models like YOLOv8, the research not only benchmarks the performance of traditional features in defect detection but also explores the synergy between classical and modern approaches. The experimental results demonstrate that these traditional features, when fine-tuned and combined with machine learning models, can significantly improve the accuracy and reliability of tire defect detection, aiming to set a new standard in automated quality assurance in tire manufacturing.
Paper Structure (15 sections, 2 figures, 3 tables)

This paper contains 15 sections, 2 figures, 3 tables.

Table of Contents

  1. Introduction
  2. Related Works
  3. Methodology
  4. Feature Extraction
  5. YOLO with Augmented Features
  6. Model Selection, Optimization, and Pre-Processing
  7. The Probability Map Ensemble
  8. Experiments and Results
  9. Dataset Overview
  10. Metrics
  11. Experiment 1: Comparing Engineered Features vs. Deep Models
  12. Experiment 2: Ablation Study on the Feature Sets
  13. Experiment 3: Effect of Dataset Generation Parameters
  14. Discussion
  15. Conclusions

Figures (2)

  • Figure 1: Examples of two major tire defect types (blister and wire) with inherent challenges. Note: annotations (green) and predictions (red) are marked.
  • Figure 2: The three channels used as input for the YOLO detection model.