The ability to accurately measure velocity within the embryonic zebrafish heart, at high spatial and temporal resolution, enables further insight into the effects of hemodynamics on heart development. Unfortunately, currently available techniques are unable to provide the required resolution, both spatial and temporal, for detailed analysis. Advances in imaging hardware are allowing bright field imaging combined with particle image velocimetry to become a viable technique for the broader community at the required spatial and temporal resolutions. While bright field imaging offers the necessary temporal resolution, this approach introduces heart wall artifacts that interfere with accurate velocity measurement. This study presents a technique for cardiac-phase filtering of bright field images to remove the heart wall and improve velocimetry measurements. Velocity measurements were acquired for zebrafish embryos ranging from 3 to 6 days postfertilization. Removal of the heart wall was seen to correct a severe (3-fold) underestimation in velocity measurements obtained without filtering. Additionally, velocimetry measurements were used to quantitatively detect developmental changes in cardiac performance in vivo, investigating both changes in contractile period and maximum velocities present through the ventricular-bulbar valve.