Background: Skin capillaroscopy has been proven to be a valid tool to study microvascular function non-invasively in humans. Unfortunately, the analysis of capillary density from video-clips is very time-consuming, since this is done manually. This impedes the use of this technique in large-scale studies. We aimed to develop a (semi-)automated assessment of skin capillary density. Methods: CapiAna (Capillary Analysis) is a newly developed semi-automatic image analysis application. The technique involves four steps: 1) movement correction, 2) positioning of the region of interest (ROI) and selection of the frame range, 3) automatic detection of capillaries, and 4) manual correction of detected capillaries. We assessed accuracy, reproducibility, and time-savings of the semi-automatic method as compared to the manual counting procedure. For this, finger skin capillary density was measured in 10 healthy participants (6 women; mean age 55.7 (49 – 68) years). First, to investigate agreement between the semi-automatic method and the manual procedure, we used linear regression and Bland-Altman analysis. Second, intra- and interobserver reproducibility of the semi-automatic method was assessed by calculating coefficients of variation (CV). For this, an experienced investigator semi-automatically counted the number of capillaries twice with a two week interval. In addition, a second experienced investigator semi-automatically counted the number of capillaries in exactly the same ROI. Finally, the difference in analysis-time between the semi-automatic and the manual method was assessed. Results: Linear regression of the data resulted in a correlation (r) of 0.88 between the semi-automatic method and the manual procedure. The comparison between the semi-automatic method and the manual method by the Bland-Altman analysis, showed a bias of -3.9 capillaries/mm2 (<5%) and limits of agreement (± 2 SD) of +13.6 to -21.4 capillaries/mm2 (approximately 17%). The intra- and interobserver CV of the semi-automatic method were 2.0% and 5.4% respectively. Finally, the mean analysis time for the semi-automatic method was on average 30 minutes versus 90 minutes for the fully manual procedure. Conclusion: In conclusion, we have developed a semi-automatic image analysis application for the assessment of skin capillary density. The results indicate that the semi-automatic method is in good agreement with the manual procedure, with a good correlation and with a small bias and reasonable limits of agreement. In addition, the semi-automatic method is more reproducible as compared to previous data for the manual procedure. Finally, analysis time with the semi-automatic method was significantly reduced. This new analysis technique offers opportunities for skin capillaroscopy as a tool to estimate microvascular function in large-scale studies.