基于形态学的不完全性骨折CT图像裂痕定位

针对不完全性骨折的CT图像损伤位置不易明确定位的问题,提出了一种基于数学形态学的裂痕定位算法。为了从CT图像中获得准确的骨质区域,在CIE Lab均匀色彩空间对原始图像进行双边滤波去噪,利用最大类间方差法计算出最佳阈值,从而分离出骨质区域,采用数学形态学中的黑帽运算结合Canny边缘检测算子对裂痕位置进行精确定位。实验结果表明,能够自动提取并定位不完全性骨折CT图像的裂痕区域。算法易于实现,具有较强的鲁棒性和实用性,方便了医学影像诊断。

CT image crack location with incomplete fracture based on morphology

A new algorithm based on mathematical morphological is proposed to solve the problem that how to locate the damage site easily in the incomplete fracture CT images.In order to obtain accurate bone region from CT image,image noise is removed through bilateral filtering in the CIE Lab well-distributed color space.The segmentation threshold is estimated by the method of the maximum between class variance, which can be used to separate the bone region from the image.The crack area is accurate located by making use of the black cap operation in the Mathematical morphology and the Canny edge detection operator.Experimental results show that, the algorithm can automatically identify and extract the crack area in the incomplete fracture CT image.It''s easy to implement and has strong robustness and practicability, which greatly facilitates the medical image diagnosis.