【摘要】 针对计算机辅助测试的数字采样信号向模拟信号转化时出现的拟合失真现象,对Bezier样条插值法进行优化,提出一种基于Bezier样条插值的采样数据光滑曲线拟合方法,并针对个别采样数据失效,提出了一种可直观交互的降噪处理方案.在降噪处理中,依据数字采样信号出现异常的物理特点,从待降噪控制点集合中计算出最佳降噪控制点,并对最佳降噪控制点的临近采样点按照合理的幅度进行降噪,在高效降噪的同时保证了曲线的光滑,保证了拟合曲线的质量,更加准确地反应了模拟信号所代表的物理意义.通过在基于WPF的MVVM模式设计的测井数据处理系统中应用,验证了方法的有效性,取得了很好的效果.更多还原

【Abstract】 Aiming at the distortion problem generated by curve fitting in the process of D/A, this paper optimize the method of bezier spline interpolation and put forward a method of fitting smooth curve based on bezier spline interpolation and sample data. To deal with the failure sample data, we propose a method for noise reduction which is intuitionist and interactive. According to the physical characteristics of sample data, the most efficient point for noise reduction can be found. Noise reduction can be achieved by adjust the closest points of the most efficient point reasonable. These methods can make noise reduction efficiently, and guarantee the fitted curve quality. The physical meaning of the analog signal can be expressed rightly. By using those in the well logging processing and interpretation system which is based on the WPF MVVM model, the feasibility and effectiveness are verified.更多还原