Validation of powder flow characteristics in laser deposition using image correlation

Additive Manufacturing (AM), particularly Laser Direct Metal Deposition (LDMD), enables precise layer-wise metal deposition; however, powder stream stability critically influences part quality and dimensional accuracy. This study introduces a non-contact, image-based methodology integrating Digital Image Correlation (DIC) and Structural Similarity Index Measure (SSIM) to quantitatively assess powder distribution in LDMD. High-speed imaging combined with Python-based OpenCV processing is employed to measure convergence diameter and uniformity index for different materials (aluminum, iron, and lead) under varying process conditions. Pixel-to-length calibration and structural similarity analysis allow accurate capture of transient powder flow characteristics. Results show that convergence diameter increases with particle size and gas pressure, reaching 4.1 mm for 300 µm lead particles, while uniformity indices between 0.9585 and 0.985 indicate stable flow behavior. The proposed approach provides a validated framework for optimizing LDMD process control, reducing material wastage, and improving component performance in high-precision manufacturing sectors.