Active Contour Using Local Regional Information on Extendable Search Lines (LRES) for Image Segmentation

• DOI: 10.1587/transinf.e93.d.1625
• Publication Date: 2010-06-01T06:06:27Z
• Authors: Sopon PHUMEECHANYA; Charnchai PLUEMPI...; Saowapak THONGVIG...
• Keywords: Active contour model; Contour line; Line (geometry)

In this paper, we propose a novel active contour method for image segmentation using local regional information on extendable search line. We call it the LRES contour. Our uses intensity values along set of lines that are perpendicular to front. These used inform front toward which direction move in order find object's boundary. Unlike other methods, none these have predetermined length. Instead, their length increases gradually until boundary object is found. compare performance our existing contours, both edge-based and region-based. The results show provides more desirable outcomes, particularly some images where methods may fail. Not only robust noise able reach into deep concave shape, also has large capture range performs well segmenting heterogeneous textured objects.