Research using X-ray diffraction (XRD) remains to be accelerated in spite of its importance in materials science. Automated noise separation or optimization of measurement time in XRD is beneficial for discovering materials. This study analyzes two-dimensional XRD (2D-XRD) with density-based clustering to accelerate XRD. This clustering technique can separate diffraction pattern signals from noises, even with low signal-To-noise ratio (S/N) 2D-XRD. Moreover, we found that the crystalline degree information in composition spreads is captured based on density. This information requires a long time to be captured with conventional one-dimensional detectors or scintillation counters. Therefore, these findings lead to dramatic reduction and optimization of measurement time to improve S/N. The proposed procedure is applicable with 2D detector measurements.