摘要
介绍了一种用于X线头影片进行边缘检测的多重分形方法,并对模拟X线头影侧位片进行了多重分形性质的分析。通过图像的多重分形谱与权重因子相关性的分析,进一步判定线性区间及适当的权重因子。得到了X线头影片图像的多重分形的线性区间,验证了X线头影片图像的多重分形的几何性质,在分析图像各像素点多重分形谱的基础上获取图像的边缘信息。用该算法进行图像边缘提取,保持和加强图像的局部特性,突出图像的细微处,并且该算法克服了多重分形谱对噪声敏感的缺点,能有效抑制图像中的噪声。分析结果表明,利用多重分形对X线头影片图像进行硬组织内部轮廓的边缘提取是可行的。
Muhifractal approach to detect the edge of ceohalogram is introduced, The multifractal spectrum property of cephalogram is analyzed and the relationship between the multifractal spectrum and its weight is discussed, so the multifraetal linear region of cephalogram is determined, The geometrical properties of its multifractal are verified. The edge information used in the above approach shows that the noise could be repressed effectively, the local property of an image could be remained as much as possible, and the fine property could be stressed, It is shown that the edges of inner hard tissues can be detected with multifractal approach.
出处
《控制工程》
CSCD
2006年第6期609-612,共4页
Control Engineering of China
基金
黑龙江省自然科学基金资助项目(2004F0303)
黑龙江省教育厅基金资助项目(10541173)
黑龙江大学重点实验室基金资助项目(HDXH0505)
