Automated Optical Inspection for iPhone User Side Glass Logo

Segmentation-based inspection · Logo region defects · Reflective surface handling

Overview

Specialized inspection system for user-side glass logo area with precision segmentation to detect subtle defects. The system handles challenging reflective and low-contrast logo regions with robust defect localization and classification.

🔬 Segmentation 👁️ AOI Python PyTorch OpenCV

Problem

  • Logo region has reflective/low-contrast characteristics.
  • Subtle defects difficult to detect with traditional methods.
  • Need for precise localization in small logo area.
  • Variations in lighting and viewing angles affect detection.

Solution

  • Segmentation-based inspection for reflective/low-contrast regions.
  • Improved defect localization and classification robustness.
  • Custom training with augmentation for various lighting conditions.
  • Production-ready optimization and validation.

My Role

  • Designed segmentation-based inspection for reflective/low-contrast logo regions.
  • Improved defect localization and classification robustness.
  • Production-ready optimization and validation.
  • Tuned model for stable performance across manufacturing variations.

Implementation

Language: Python

Libraries:

PyTorch OpenCV

Approach: Deep learning segmentation for logo-region defects

Results Gallery

Segmentation results showing Apple logo defect detection.

Training and evaluation demonstrations are omitted due to data confidentiality.

Demo Video

Apple logo inspection system demonstration.

Training and evaluation demonstrations are omitted due to data confidentiality.

End Result

  • Reliable detection of logo-region defects with stable performance.
  • Robust handling of reflective and low-contrast conditions.
  • Production-ready system deployed in manufacturing.