摘要
水平集方法(Level Set Method)是一种用于计算曲线演化位置的有效方法。该文针对传统Level Set Method进行图像检测中需要重复初始化水平集函数的缺点,研究了不需要重复设置水平集函数的图像检测方法,实验表明,该方法能更好的检测多目标图像轮廓。
Level set method is numerical techniques for computing the position of propagating fronts. Image contour detecting by traditional level set methods Needs to re-initializing the level set" function.Then, Without Re-initialization Level Set Method is discussed is proposed.The experimental results show that it can accurate converge to contour of Multi-objectives image.
出处
《电脑知识与技术》
2008年第11期944-946,共3页
Computer Knowledge and Technology
关键词
水平集方法
非重复初始化
轮廓检测
Level Set Method
Without Re-initialization
Contour Detecting