Segmentation of blurred objects by classification of isolabel contours
作者：
Highlights：
• The algorithm is suitable for segmentation of grayscale blurred homogeneous objects.
• The objects are segmented by SVM classification of isolabel contours.
• Labeling training images is reduced to specifying contours on isolabel-contour map.
• Classification of isolabel contours allows using of global object features.
• High performance is demonstrated for segmentation of brain lesions in MRI.
摘要
Highlights•The algorithm is suitable for segmentation of grayscale blurred homogeneous objects.•The objects are segmented by SVM classification of isolabel contours.•Labeling training images is reduced to specifying contours on isolabel-contour map.•Classification of isolabel contours allows using of global object features.•High performance is demonstrated for segmentation of brain lesions in MRI.
论文关键词：Blurred objects segmentation,Isolabel-contour map,Support vector machine,Supervised machine learning,Brain lesion segmentation
论文评审过程：Received 7 October 2013, Revised 7 May 2014, Accepted 11 June 2014, Available online 19 June 2014.
论文官网地址：https://doi.org/10.1016/j.patcog.2014.06.007
