摘要
在分析了可变形模型(Deformable models or snakes)的物理背景的基础上,将量子力学中关于粒子运动的观念及规律引入到目标轮廓的提取中,通过估计粒子从一点到达另一点的概率(相对概率),得到了一种新的目标轮廓提取方法-量子轮廓模型,并对提取的轮廓进行了光滑处理。实验表明,我们的方法有较好的效果、较快的速度。
Physical background of deformable model is described. Then, an approach of finding the contour of an object based on quantum mechanics is obtained by introducing the point of view of the quantum mechanical law of motion of a particle into extraction of object contour and estimating the probability (relative probability) that a particle goes from a point to another point. Next, Smoothing of the extracted contour is considered. Finally, this method is applied to both simulated images and real medical images.
出处
《量子电子学报》
CAS
CSCD
北大核心
2005年第3期354-360,共7页
Chinese Journal of Quantum Electronics
基金
湖北省自然科学基金(2004ABA016)湖北省教育厅科学技术研究项目基金(2004D004)资助
关键词
数字图像处理
能量最小化
可变形轮廓
量子力学
量子轮廓
digital image processing
energy minimization
deformable contour
quantum mechanics
quantum contour