【摘要】 由于X射线数字图像光场分布呈抛物面状 ,据此提出了一种基于最小二乘法的实用算法 ,用以解决X射线数字图像光场矫正问题 .该算法首先取空白图像中最亮区的平均灰度作为参考值 ,并将空白图像中所有像素点的灰度值与相应的系数相乘后规则化至该参考值 ,然后采用抛物面函数对系数图像进行最小二乘法拟合生成矫正算子 ,最后利用该矫正算子对目标图像进行矫正 .实验表明 ,该方法所生成的矫正算子不仅稳定、可靠 ,同时能够在不增加图像噪声的情况下对目标图像中不均匀的光场进行矫正更多还原

【Abstract】 Experiment shows that the light field of the X ray digital image is in paraboloidal distribution.A practical algorithm based on the least mean square (LMS) algorithm for the correction of X ray digital image is proposed.Firstly, the average gray level of the highlight region in the empty image is used as a referential value, and all the pixels in the empty image are multiplied by the corresponding coefficient to get the referential value.Then, a paraboloidal function is adapted to fit the coefficient image and to make the correction operator.Eventually, the correction operator is applied to correct the objective image. It is shown that the operator generated by this algorithm is stable and reliable, and it can correct the non uniform illumination in the objective image without increasing the noise.更多还原