Abstract— We present a novel technique using active contour model (ACM) incorporating with edge flow for texture segmentation. The edge flow magnitude was used as an external force of active contour model. The active contour model will be guided by the edge flow vector. In our method, the gradient vector flow was replaced by the edge flow vector. Using edge flow vector, texture image can be segmented efficiently. The results of texture segmentation using active contour model with edge flow vector and gradient vector flow will be compared. It was observed that the active contour model with edge flow vector shown better texture segmentation capability. It should be pointed out that the texture image cannot be segmented using the traditional ACM and gradient vector flow. We also found that it is suitable to be applied in the MRI image segmentation.
Index Terms— Texture segmentation, active contour model, gradient vector flow, edge flow vector.
T. Boonnuk and T. Sripramong are with The Electrical Engineering Graduate Program, Faculty of Engineering, Mahanakorn University of Technology, Bangkok, Thailand (e-mail: boonnuk2002@hotmail.com, thanwa@mut.ac.th).
S. Srisuk is with Faculty of Industrial Technology, Nakhon Phanom University, Nakhon Phanom, Thailand (e-mail: srisuk.s@gmail.com).
[PDF]
Cite: T. Boonnuk, S. Srisuk, and T. Sripramong, " Texture Segmentation Using Active Contour Model with Edge Flow Vector," International Journal of Information and Electronics Engineering vol. 5, no. 2, pp. 107-111, 2015.