基于多尺度Gabor滤波的造影血管中轴线的自动提取

Automated Centerline Extraction of Angiogram Vessels Based on Multiscale Gabor Filters

提出了一种自动提取数字造影血管图像中2维血管树中轴的新方法。作为血管树3维重建的基础,2维中轴提取的准确性显得格外重要。该方法充分利用了多尺度实Gabor滤波灵活的频率带宽和理想的增强效应,对不同粗细的血管进行中轴增强和去除背景噪声,再利用Hessian矩阵计算结果提供的方位信息进行非最大值抑制求取响应图的局部极值点,最后通过双阈值分割得到血管中轴线。实验结果表明,该方法鲁棒性好,提取中轴线的质量高。

In this paper, a new automated approach is proposed to extract the centerlines from 2D angiogram images. The centerline extraction is the basis of 3 D reconstruction of the blood vessels, so the accurate localization of centerlines counts is extremely important. The characteristics of multiscale Gabor even filters, flexible frequency bands and enhancement effects,are fully utilized to enhance centerlines of the blood vessels in various size, which greatly reduce the background noise. Then the non-maximum suppression algorithm is adopted, guided by the directions provided by calculating the Hessian matrix of the image, to get the local maximum points on the response map. Finally, vessel centerlines are extracted by a hysteresis-thresholding segmentation. The experimental results demonstrate the robustness of the algorithm and good quality of the extracted vessel centerlines.