Liver Focus Detections Based on Visual Attention Model
The detection of region of interest (ROI) in medical images has played a very important role in computer aided diagnose. With respect to liver-focus pixels having weak textural and similar intensities with their neighborhood, a novel detection algorithm of abnormal regions in liver CT images has been proposed in this paper by visual attention model. Firstly, a set of statistical texture features for strengthening finer textures and some salient factors based on directional fractals are selected. Then, a saliency map is generated by combinations of several sub-saliency maps. Finally, regions with liver focus are located by labeling the saliency map. Experiments shows the ROI locations could be extracted by the proposed method with accuracy and efficiency for ROI detection of medical images using saliency maps.
Saliency maps Visual attention model ROI eztraction Liver computer tomography
Li Ma Wenfeng Wang Shaofang Zou Juan Zhang
School of automation,Hangzhou Dianzi University,Hangzhou,310018,P.R.China Department of Radiology,Zhejiang Cancer Hospital,Hangzhou,310022,China
国际会议
北京
英文
1-5
2009-06-11（万方平台首次上网日期，不代表论文的发表时间）