提取血管与血管中心线的算法研究

发布时间：2018-02-27 00:20

本文关键词： 血管提取 MOC算法动态规划 SOP算法　出处：《长春工业大学》2012年硕士论文　论文类型：学位论文

【摘要】：随着人们生活水平的不断提高,心脑血管疾病已经成为影响人们身体健康的主要病因之一,血管轮廓和血管中心线提取技术的研究发展,为心脑血管疾病的诊断输入了新的血液。血管轮廓和血管中心线的同时提取,确保医生在诊断中获得出更加丰富的血管信息,使人为诊断误差降低到最小。血管轮廓的精确提取对于血管疾病的诊断研究起着决定性作用。传统算法提取出的血管轮廓边缘不连续、混有伪边缘轮廓。针对这一问题,本文结合常规微分法,在动态规划模型的基础上提出MOC——最大优化成本法。采用该算法提取的血管轮廓边缘连续、定位准确、避免伪边缘的提取。血管中心线跟原始图像具有相同的拓扑性和相似的几何形状,因此可以较好的反映血管走向和血管的空间结构,同时佐证了血管轮廓提取的准确性。在以距离变换算法为基础,Hessian矩阵模型的提取脑血管中心线的前提下,进一步通过可视性搜索检测要求,制约最短路径的生成提取方法,然后提出了SOP——最短路径优化法。采用该算法确保血管中心线连续平滑的提取出来。同时提取血管轮廓和血管中心线,使血管轮廓和血管中心线相互佐证,相互制约,减少了虚假判断,克服了传统方法提取出的血管轮廓和血管中心线边缘不连续,具有提取灵敏度高,定位准确等优点,具有广泛的实用价值。
[Abstract]:With the continuous improvement of people's living standards, cardiovascular and cerebrovascular diseases have become one of the main causes affecting people's health, and the research and development of vascular contour and blood vessel centerline extraction technology. The new blood is input to the diagnosis of cardiovascular and cerebrovascular diseases. The blood vessel contour and the blood vessel center line are extracted simultaneously to ensure that the doctor obtains more abundant vascular information in the diagnosis and minimizes the error of artificial diagnosis. The accurate extraction of vascular contour plays a decisive role in the diagnosis and research of vascular diseases. The edge of vascular contour extracted by traditional algorithm is discontinuous and mixed with pseudo-edge contour. Based on the dynamic programming model, MOC-maximum optimal cost method is proposed, which can extract the edge of vascular contour continuously, locate accurately and avoid pseudo-edge extraction. The center line of the blood vessel has the same topology and similar geometry as the original image, so it can better reflect the direction of the blood vessel and the spatial structure of the blood vessel. At the same time, the accuracy of vascular contour extraction is confirmed. On the premise of extracting cerebrovascular centerline from Hessian matrix model based on distance transform algorithm, the method of generating and extracting shortest path is restricted by visual search detection. Then the SOP- shortest path optimization method is proposed, which is used to ensure the smooth extraction of the blood vessel centerline. At the same time, the blood vessel contour and the blood vessel center line are extracted at the same time, which makes the blood vessel contour and the blood vessel center line mutually corroborate, restricts each other, reduces the false judgment, and overcomes the discontinuity of the blood vessel contour and the blood vessel center line edge extracted by the traditional method. It has the advantages of high extraction sensitivity, accurate location and so on. It has extensive practical value.
【学位授予单位】：长春工业大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：R318.0

【参考文献】