Despite recent advancements in the powder bed fusion (PBF) additive manufacturing (AM), in-process monitoring and quality assurance remain insufficient for wide applications in production. One obstacle is the interpretation and evaluation of the intense monitoring data. The objective of this study is to discuss a data processing infrastructure for coaxial sensing towards consistent and repetitive AM. Pioneering data analytics with morphological image processing, imaging field correction, and histogram equalization on an in-house developed coaxial database, we address key issues induced by contaminated optics, inconsistent illumination, and low contrast between raw materials and the built parts in the noisy grayscale raw images. By utilizing texture-based region analysis and graph-based segmentation, our approach leads to an innovation to isolate laser-fused materials from the unmolten powder bed. From there, we provide an estimation of the melting-pool width and identify an overheating defect. These results, along with other significant process signatures, can be further used to bridge the gap between spatially resolved process monitoring and ultimate model-based control for a robust, high-throughput AM.