GDL

A Novel Image Segmentation Algorithm Based on Graph Cut Optimization Problem

Journal from gdlhub / 2016-11-16 06:49:55
By : Zhang Guang-hua, Xiong Zhong-yang, Li Kuan, Xing Chang-yuan, Xia Shu-yin, Telkomnika
Created : 2015-12-01, with 1 files

Keyword : Image Segmentation, Graph Cut, Energy function, Pixel
Url : http://journal.uad.ac.id/index.php/TELKOMNIKA/article/view/1178

Image segmentation, a fundamental task in computer vision, has been widely used in recent years in many fields. Dealing with the graph cut optimization problem obtains the image segmentation results. In this study, a novel algorithm with weighted graphs was constructed to solve the image segmentation problem through minimization of an energy function. A binary vector of the segmentation label was defined to describe both the foreground and the background of an image. To demonstrate the effectiveness of our proposed method, four various types of images were used to construct a series of experiments. Experimental results indicate that compared with other methods, the proposed algorithm can effectively promote the quality of image segmentation under three performance evaluation metrics, namely, misclassification error rate, rate of the number of background pixels, and the ratio of the number of wrongly classified foreground pixels.

Description Alternative :

Image segmentation, a fundamental task in computer vision, has been widely used in recent years in many fields. Dealing with the graph cut optimization problem obtains the image segmentation results. In this study, a novel algorithm with weighted graphs was constructed to solve the image segmentation problem through minimization of an energy function. A binary vector of the segmentation label was defined to describe both the foreground and the background of an image. To demonstrate the effectiveness of our proposed method, four various types of images were used to construct a series of experiments. Experimental results indicate that compared with other methods, the proposed algorithm can effectively promote the quality of image segmentation under three performance evaluation metrics, namely, misclassification error rate, rate of the number of background pixels, and the ratio of the number of wrongly classified foreground pixels.

Give Comment ?#(0) | Bookmark

Property	Value
Publisher ID	gdlhub
Organization	Telkomnika
Contact Name	Herti Yani, S.Kom
Address	Jln. Jenderal Sudirman
City	Jambi
Region	Jambi
Country	Indonesia
Phone	0741-35095
Fax	0741-35093
Administrator E-mail	elibrarystikom@gmail.com
CKO E-mail	elibrarystikom@gmail.com