Image Segmentation Based on Graph-Cut Models and Probabilistic Graphical Models: A Comparative Study

Abstract

Image Segmentation has been one of the most important unsolved problems in computer vision for many years. Recently, there have been great efforts in producing better segmentation algorithms. The purpose of this paper is to introduce two recently proposed graph based segmentation methods, namely, graph-cut models (deterministic) and unified graphical model (probabilistic). We present some foreground/background segmentation results to illustrate their performance on images with complex background scene.

Design and Validate a Dual-Modality Characteristic Information Fusion System Based on Probabilistic Graphical Models

“…The current state di(fi) is the likelihood that executes individual penalty for assigning a pixel label fi (eg, the lesion or the normal) to the corresponding pixel i. The smoothness term wi−1,i(fi−1, fi), meaning the interaction potential between adjacent pixels (fi−1, fi) [43], estimates the loss of assigning diverse pixel labels to two adjacent pixels fi−1 and fi in the set of Np.…”

A Novel Lung Nodule Accurate Segmentation of PET-CT Images Based on Convolutional Neural Network and Graph Model

“…The current state d i (f i ) is the likelihood that executes individual penalty for assigning a pixel label f i (eg, the lesion or the normal) to the corresponding pixel i. The smoothness term w i−1,i (f i−1 , f i ), meaning the interaction potential between adjacent pixels (f i−1 , f i ) [46], estimates the loss of assigning diverse pixel labels to two adjacent pixels f i−1 and f i in the set of N p .…”