Multiphase flow characterization is an important task for monitoring, measuring or controlling industrial processes. This can be done by means of process tomography. The use of tomographic techniques has been used within the oil industry. One of the potential applications is flow visualization and measurement in producing wells. Research on industrial process tomography consists in obtaining estimated images of a cross section of a pipe or vessel containing or carrying the substances of the process. One category of process tomography is ultrasonic tomography technique. A simple tomography can be built by mounting a number of sensors around the circumference of a horizontal pipe. This includes acquiring and processing ultrasonic signals from the transducers to obtain the information of the spatial distributions of liquid and gas in an experimental column. Analysis on the transducers’ signals will be carrying out to distinguish between the observation time and the Lamb waves. The information obtained from the observation time is useful for further development of the image reconstruction. To obtain the time easily, the time will be calculated from the starting pulse of transmitter signal until the starting peak of receiver signal. Finally Support Vector Machine (SVM) was employed to distinguish of each phase between water and gas.