Title: An adaptive and selective segmentation model based on local and global image information

Authors: Xueqin Wang; Shurong Li; Jiaojiao Li; Jiayan Wang

Addresses: College of Information and Control Engineering, China University of Petroleum, Qingdao, 266580, China; College of Electronic, Communication and Physics, Shandong University of Science and Technology, Qingdao, 266590, China ' No. 66 , Changjiang West Road , Huangdao District , Qingdao Shandong Province, 266580 China ' College of Information and Control Engineering, China University of Petroleum, Qingdao, 266580, China ' College of Information and Control Engineering, China University of Petroleum, Qingdao, 266580, China; College of Electronic, Communication and Physics, Shandong University of Science and Technology, Qingdao, 266590, China

Abstract: This study investigates the application of partial differential equations in image segmentation field. A novel selective segmentation mode is proposed for the existing selective segmentation model which cannot segment the intensity inhomogeneity and fuzzy edge image. In this novel model, a weighting function based on local information is constructed. This weighting function can introduce the global and local information of the image into the novel model which can realise the adaptive segmentation of the image. Compared with the existing selective segmentation model, the novel selective segmentation model proposed in this paper can realise the adaptive segmentation of intensity inhomogeneity and fuzzy edge images. Experimental results show that the novel model is more effective and adaptive to segment images with intensity inhomogeneity or fuzzy edge, and less sensitive to the location of initial contour, without choosing the weighting parameter between global and local information by manual method.

Keywords: selective segmentation; intensity inhomogeneity image; adaptive segmentation; level set method; weighting function; global and local information.

DOI: 10.1504/IJMIC.2017.085943

International Journal of Modelling, Identification and Control, 2017 Vol.28 No.2, pp.114 - 124

Received: 12 Aug 2016
Accepted: 03 Oct 2016
Published online: 18 Aug 2017 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article