均值漂移带宽选取新方法及其在分割肺结节中的应用

针对肺结节与周围组织相连且边缘模糊造成分割困难的问题,提出一种新的均值漂移(meanshift)带宽自动选取方法并采用均值漂移算法解决结节分割.与基于统计分析规则的带宽选择方法相比,该方法时间复杂度低,且能得到符合实际问题的正确带宽参数.应用带宽选择定理确定带宽参数的初始值,利用尺度空间滤波聚类理论的最稳定尺度准则确定最佳的自适应带宽参数.该方法对毛玻璃型、粘连血管型、贴胸壁型和各向异性型进行评估实验,都取得了正确的分割结果.结果表明,该方法对分割结节是有效的.

New Selection Method by Bandwidth Mean Shift and Its Application in Segmentation of Lung Nodule

To solve the segmentation problem of the blurred image edge and the connection of lung nodule to peripheral tissues,a new selection method is proposed by adaptive bandwidth mean shift and its mean shift is applied to nodule segmentation.Comparing it to the selection method by bandwidth on the basis of statistical analysis,it has the advantages of low complexity at time cost and getting correct bandwidth adaptive to actuality.According to the principle of bandwidth selection,the initial parameters of bandwidth are determined,and the most stable scale criterion of multi-scale filtering clustering theory was used to determine the optimal parameters of bandwidth.The proposed method was evaluated and tested for the clinical chest CT images including the nodules of different types,such as the ground glass opacity,the nodules lung walls and vessels adhesion,and anisotropic nodules.The results revealed that the proposed method is successful in segmentating lung nodules.