A Plain Segmentation Algorithm Utilizing Region Growing Technique for Automatic Partitioning of Computed Tomography Liver Images

A Plain Segmentation Algorithm Utilizing Region Growing Technique for Automatic Partitioning of Computed Tomography Liver Images Arica, Sami; Avsar, Tugce Sena; Erbay, Gurcan Medical image segmentation is quite significant, especially for diagnosis and treatment of diseases. In this study, similar and different tissues in computed tomography (CT) images of liver are decomposed by utilizing region growing method. The images are preprocessed before segmentation. First, gray scale CT images are smoothed with a median filter, and a coarse segmentation is done with four level uniform quantization. A pixel from each connected component of the quantized image is selected as a seed point and is employed by region growing algorithm to specify corresponding segment. The number of segments depends on the number of connected components. Experimental results show that this basic method has successfully segmented the liver.

A Plain Segmentation Algorithm Utilizing Region Growing Technique for Automatic Partitioning of Computed Tomography Liver Images

A Plain Segmentation Algorithm Utilizing Region Growing Technique for Automatic Partitioning of Computed Tomography Liver Images Arica, Sami; Avsar, Tugce Sena; Erbay, Gurcan Medical image segmentation is quite significant, especially for diagnosis and treatment of diseases. In this study, similar and different tissues in computed tomography (CT) images of liver are decomposed by utilizing region growing method. The images are preprocessed before segmentation. First, gray scale CT images are smoothed with a median filter, and a coarse segmentation is done with four level uniform quantization. A pixel from each connected component of the quantized image is selected as a seed point and is employed by region growing algorithm to specify corresponding segment. The number of segments depends on the number of connected components. Experimental results show that this basic method has successfully segmented the liver.