Parameter selection on a multi-exposure fusion method for measuring surfaces with varying reflectivity in microscope...

Area of Science:

• Metrology
• Optical Measurement
• Surface Characterization

Background:

• Accurate three-dimensional (3D) shape and surface measurement is crucial for advancing industries and scientific research.
• Measuring surfaces with varying reflectivity presents significant challenges in 3D reconstruction.
• Existing multi-exposure fusion methods offer high-quality results but rely on experience-based parameter selection.

Purpose of the Study:

• To improve the multi-exposure fusion method for enhanced 3D measurement accuracy and completeness.
• To introduce a guided approach for selecting optimal parameters in multi-exposure fusion.
• To experimentally validate the influence of key parameters on 3D reconstruction integrity.

Main Methods:

• Development of an improved multi-exposure fusion algorithm.
• Introduction of a guided parameter selection strategy.
• Creation of a comparative model to analyze Gaussian window variance, optimal grayscale range, and attenuation factor variance.
• Experimental validation on high-dynamic surfaces.

Main Results:

• The guided parameter selection significantly enhances the completeness of 3D measurement results.
• Experimental validation confirmed the impact of Gaussian window variance, optimal grayscale range, and attenuation factor variance.
• Restoration coverage for high-dynamic surfaces improved from 86% (bright areas) and 50% (dark areas) to over 99%.

Conclusions:

• The proposed guided parameter adjustment method optimizes multi-exposure fusion for precise 3D measurements.
• This approach effectively addresses challenges in measuring surfaces with varying reflectivities.
• The study provides a robust strategy for parameter selection in multi-exposure based 3D metrology.