Computer Automated Measurement System (CAMS)

Author(s):

Gael Mota, Clara Mock, and Stephen Cluff

Abstract:

Additive manufacturing (AM) provides unparalleled design freedom for valuable applications, from medical prosthetics to military equipment. However, the quality of AM products is critically dependent on processing parameters like layer thickness and scanning speed. While artificial intelligence and machine learning (AI/ML) are powerful tools for optimizing these parameters by learning from measured data, its application is hindered by the manual characterization of printed samples needed to build large datasets. This work presents the Computer Automated Measurement System (CAMS), a computer vision-based solution for high-throughput automated characterization of AM samples. Validation revealed that the difference between CAMS and manual measurements by FIJI was less than 2% for area measurements and less than 4% for aspect ratio measurements, while processing samples 198% faster than manual methods. This demonstrates the feasibility of fully automated high-throughput characterization, thereby enabling the rapid dataset generation required to accelerate AI/ML-guided refinement of AM processes.