Color flow mapping systems have become widely used in clinical applications. It provides an opportunity to visualize the velocity profile over a large region in the vessel, which makes it possible to diagnose, e.g., occlusion of veins, heart valve deficiencies, and other hemodynamic problems. However, while the conventional ultrasound imaging of making color flow mapping provides useful information in many circumstances, the spatial velocity resolution and frame rate are limited. The entire velocity distribution consists of image lines from different directions, and each image line is estimated using multiple emissions. Therefore, it is very difficult to acquire a full volume of data for the blood flow in the heart in real-time. A radical break with this has been the synthetic aperture technique. This technique makes it possible to increase the frame rate, and the reconstruction also makes it possible to improve significantly the focusing and frame rate. However, it requires a large amount of calculations to fulfill the performance because the signal from each channel is stored and processed simultaneously. The implementation of the full synthetic aperture would be very expensive on the current commercial ultrasound scanner. The motivation for this project is to develop a method lowering the amount of calculations and still maintaining beamforming quality sufficient for flow estimation. Synthetic aperture using a dual beamformer approach is investigated using Field II simulations, phantom measurements and in vivo measurements. Firstly, the method is used to estimate the velocity along the ultrasound beam, which is the axial component. The results all show good quality of color flow mapping in terms of standard deviations and bias. The results of in vivo measurements show the capability of acquiring color flow mapping with a high frame rate. Secondly, the new method is extended to the vector velocity estimation using directional beamforming, which beamforms data in the flow direction. The magnitude of the flow can be obtained and results of simulations and phantom measurements show good agreements with the truth. With directional beamforming, the velocity in the transverse direction can be achieved, which is impossible for the conventional method. Comparing the amount of calculations shows a reduction in number of calculations for the new method compared to full synthetic aperture.