摘要
对重击下颅骨CT图像裂痕区域进行检测,可以提高CT图像的检测精度。重击下的医学颅骨图像裂痕区域较大,骨裂区域会形成体液流入干扰,降低了裂痕与骨头区域的特征差异度,传统算法进行重击下的医学颅骨图像裂痕区域分割时,需要结合裂缝区的特殊差异度完成检测,导致医学颅骨图像裂痕区域检测误差大的问题。提出改进弹性形变算法的重击下的重击下颅骨CT图像裂痕区域检测方法。上述方法先融合于弹性形变算法以三角面片为基本单位,生成重击下的医学颅骨三维模型,并利用医学颅骨图像的灰度信息将其分割成不同的区域,结合像素的边界特性构建各个像素的特征向量,并对特征向量进行聚类,依据聚类的结果实现对重击下的医学颅骨图像裂痕区域检测。仿真结果表明,改进弹性形变算法的重击下颅骨CT图像裂痕区域检测算法可以显著提高检测精度。
It can improve the detection precision of CT image to detect the fissure area of skull CT image under thump. The fissure area under thump is larger and there is disturbance of body fluid influx. Therefore, it reduces the feature diversity factor between fissure and bone area. In this paper, we proposed a detection method for fissure area of skull CT image under the thump based on the modified elastic deformation algorithm. Firstly, the we used the triangular patch as the basic unit integrated with the elastic deformation algorithm and generated the 3D model of medical skull. Then, we used the grey information of medical skull image to divide it into different areas and built feature vector of each pixel integrated with the boundary feature of pixel and clustered the feature vector. Finally, we achieved the detection for fissure area of skull CT image under thump according to the clustering results. The simulation results show that the detection method based on the modified elastic deformation algorithm can improve the detection precision apparently.
出处
《计算机仿真》
CSCD
北大核心
2016年第12期366-369,共4页
Computer Simulation
基金
江西省教育厅基金(GJJ14503)
江西省自然科学基金(20151BAB205050)
关键词
医学颅骨
裂痕区域
检测
Medical skull
Fissure area
Detection
