Objectives
Identify Defect Types
Comprehensive identification and categorization of various defects found in LCD screens.
Pattern Effectiveness
Evaluate how different display patterns reveal various defects with varying degrees of clarity.
Preprocessing Impact
Study how image preprocessing techniques affect clarity and defect visibility.
Algorithm Analysis
Analyze algorithms suitable for detecting and classifying various LCD defects.
Pipeline Development
Propose an optimized pipeline for automated defect detection and classification.
Quality Assurance
Improve LCD manufacturing quality through advanced detection methods.
Common LCD Defects
Explore common defect types found in LCDs. Understanding these defects is crucial for effective quality control.
Line Defects
Horizontal or vertical lines caused by gate or data line failures in the TFT array.
Pixel Defects
Individual pixels that are permanently on (bright), off (dark), or stuck on a particular color.
Mura Defects
Irregular cloudy patches or stains of uneven brightness across the display.
Backlight Bleeding
Light leakage from the edges or corners of the display, especially noticeable on dark backgrounds.
Proposed Analysis Pipeline
1
Image Acquisition
Capture high-resolution images of LCD panels using controlled lighting
2
Preprocessing
Apply noise reduction, contrast enhancement, and normalization
3
Feature Extraction
Identify key features using CNN and traditional CV algorithms
4
Defect Classification
Classify defects using trained ML models and rule-based systems
5
Reporting
Generate detailed reports with defect location, type, and severity