Accuracy Evaluation of Dense Matching Techniques for Casting Part Dimensional Verification

Area of Science:

• Manufacturing Engineering
• Metrology
• Computer Vision

Background:

• Product optimization is essential for competitiveness in global manufacturing.
• Casting design, simulation, and verification tools reduce costs and maintain functionality.
• Traditional 3D scanning requires active light sources, limiting application in some scenarios.

Purpose of the Study:

• To evaluate the accuracy of dense matching techniques for casting part verification.
• To compare dense matching accuracy against established optical measurement methods.
• To assess the suitability of dense matching for industrial casting processes.

Main Methods:

• Utilizing dense matching techniques for 3D surface reconstruction of casting parts.
• Comparing dense matching accuracy with certified optical measurement systems.
• Conducting uncertainty evaluation using artificial targets and natural features.
• Testing in both laboratory and workshop environments.

Main Results:

• Dense matching techniques provide accurate 3D surface point reconstruction.
• The method's accuracy is comparable to validated optical measuring systems.
• Uncertainty evaluations confirm the reliability of the approach.
• Performance is consistent across lab and workshop conditions.

Conclusions:

• Dense matching is a viable and accurate method for casting part verification.
• This image processing technique meets required measurement tolerances in manufacturing.
• It offers a cost-effective alternative to traditional 3D scanning methods.