论文部分内容阅读
目的 对X线头影测量分析中标志点实现自动识别。方法 在自动定点过程中,对于不同的标志点采用不同的方法。选择合适的模板匹配方法应用于头影图像,可以确定孤立标志点。对于依附于边缘的标志点采用边缘检测的方法。首先对标准图像的特征边缘利用仿射变换确定被测图像特征边缘的初始位置。然后采用基于仿射变换的模板匹配方法,将模板匹配和仿射变换结合起来,调整图像的初始边缘。最后采用跟踪虫技术对边缘作进一步的修正。根据提取的边缘曲线的特征和标志点的几何特性能够确定依附于边缘的标志点。结果避免了由于头影图像中有些目标在不同图像中的位置存在差异或形变而不能识别目标的缺陷,能更准确地判断目标。结论 该技术能较好的确定X线头影图像中常用的测量标志点。
Objective To automatically identify the marked points in X-ray cephalometric analysis. Method In the automatic fixed-point process, for different signs using different methods. Select the appropriate template matching method applied to the cephalometric image, you can determine the isolated mark points. Edge detection method for edge markings. Firstly, affine transformation is applied to the feature edge of the standard image to determine the initial position of the feature edge of the image under test. Then, a template matching method based on affine transformation is used to combine template matching and affine transformation to adjust the initial edge of the image. Finally, tracking bug technology for further amendments to the edge. According to the characteristics of the extracted edge curve and the geometrical characteristics of the marker points, the marker points attached to the edge can be determined. As a result, the target can be judged more accurately because of the defects that some targets in the cephalometric image have different positions or different shapes in the different images and can not recognize the target. Conclusion This technique can determine the commonly used measurement mark points in X - cephalometric images.