This paper presents a method of automatically measuring peritoneum thickness in ultrasound images. In our previous work, a method of manually selecting the region of interest (ROI) area has been developed. To achieve an automatic ROI area selection, two phases: Gaussian high-pass filtering and bilateral filtering, are used in the proposed method. In the bilateral filtering phase, the ultrasound image is enhanced for obtaining more details of the peritoneum so that probable areas can be extracted. In the other phase, the ultrasound image is processed with a Gaussian high pass filter, and the result is used to locate the precise area of peritoneum in the first phase result. The experimental results show that the proposed method has high accuracy and fast processing speed in determining the peritoneum area and its thickness distribution.