Automatic Liver Segmentation Method Based on a Gaussian Blurring Technique For CT Images

Due to abdominal CT images´ specific characteristics, a novel automatic liver segmentation method is introduced in this paper based on a Gaussian blurring technique for binary images. This paper utilizes a simple line search method for plane domain segmentation at first to extract binary image composed of isolated white pixel clusters mainly from the liver part based on the histogram distribution and spatial characteristics of the liver region in obtained CT images; In the following step, a Gaussian blurring technique is introduced to connect the isolated pixel clusters; Threshold the blurred image and after the post-processing step of mending holes and size filter, the liver part can be finally extracted from the original CT images. The validity of the method has been proved by the experiment results presented at the end of the paper.