Deep learning image reconstruction (DLIR) and Adaptive Statistical Iterative Reconstruction-V (ASIR-V) has been used for cardiac computed tomography imaging. However, DLIR and ASIR-V may influence the quantification of coronary artery calcification. This study aimed to investigate the effects of DLIR and ASIR-V on coronary calcium quantification compared to traditional filtered back projection (FBP). CT images of 96 patients were reconstructed by FBP, ASIR-V 50%, and three levels of DLIR (low [L], medium [M], and high [H], respectively). Image noise decreased significantly with ASIR-V 50% and increasing DLIR levels from L to H in comparison with FBP (all P < 0.001). There is a significantly decline with ASIR-V 50% and incremental DLIR levels in Agatston calcium score, volume score and mass score as compared to FBP (all P < 0.001). For all CAC score risk categories, Severity classification shows no significant differences among five reconstructions (all P > 0.05). DLIR-L has the minimal effect on coronary calcium quantification as compared to ASIR-V and DLIR at medium and high levels. it may be considered as an alternative to FBP for routine clinical use.