Improved imaging of extremely-slight transparent aesthetic defects using a saturation level-guided method

Area of Science:

• Materials Science
• Optical Engineering
• Image Processing

Background:

• Polarizer production quality control is challenged by extremely-slight transparent aesthetic defects (ESTADs).
• Saturated imaging with structured backlights enhances ESTAD contrast but requires precise manual saturation control.
• Current methods lack robustness and efficiency in defect detection.

Purpose of the Study:

• To develop a simple, deployable saturation level-guided image enhancement method for industrial polarizer quality control.
• To improve the imaging contrast and detection accuracy of ESTADs.
• To automate the selection of optimal saturation levels for structured backlit imaging.

Main Methods:

• Proposed a novel, invariant definition for saturation level in structured backlit imaging.
• Established an empirical model correlating image contrast with saturation level.
• Utilized parameter optimization based on contrast data from five saturation levels to find the optimal setting.

Main Results:

• The proposed method effectively enhances imaging effects for transparent thin-film defect detection.
• Demonstrated the method's ease of use and effectiveness in industrial settings.
• Achieved improved contrast for detecting subtle defects.

Conclusions:

• The saturation level-guided enhancement method offers a significant improvement over existing techniques for polarizer defect detection.
• This automated approach simplifies quality control and enhances the reliability of defect identification.
• The method is practical for industrial deployment, addressing limitations in manual saturation selection.