In the early 1930´s, researchers were inspired by sonar to use ultrasonics for noninvasive examination. Even after 75 years, ultrasonics still is a big research area and new applications are coming up in a steady pace. The use of ultrasonic imaging has primarily been within medical applications, but with cheaper and more power-efficient equipment the industry has started to benefit from its advantages. This thesis evaluates spatio-temporal inverse filters to compensate the blurring effect when the transducer size increases. The increase in contrast will make the statistical properties of the image closer to the real statistics of the media. This is valuable when stochastic analysis methods are applied to estimate the porosity of materials or finding faults in composites. To evaluate the performance of the minimum mean square error estimator method, it is compared with the classical delay and sum beamformer. The comparison is made with simulations in MATLAB. They shows that the minimum mean square estimator method has a superior performance compared to delay and sum method. This is particulary evident when the examined object is in the near field of the transducer. To verify the simulations, measured data were used in the beamformers. Unfortunately the result was far from those obtained in the simulations. The reason for this is probably because of differences in the spatio-temporal impulse response between the simulation and the measurement. One possible solution, is to measure the impulse response of the transducer instead of simulate it.