高级检索

改进的非线性全局映射灰度化方法

The Improved Color-to-Gray via Nonlinear Global Mapping

  • 摘要: 彩色图像灰度化的一个重要目标就是尽量保持原始彩色图像的边缘特征.基于三角级数逼近的非线性全局映射灰度化方法虽能较好地保持颜色顺序,且在很多情况下也保持了特征区分度,但在某些情况下仍会丢失或模糊边缘特征.为此,基于三角级数混合多项式逼近的思路,提出了改进的非线性全局映射灰度化方法.首先在三角级数逼近的基础上继续引入参数tt2组合成最终的混合逼近函数,当tt2前面的系数均为0时即为传统的灰度化方法;然后同时优化三角级数、tt2前面的系数,以增强灰度图像的边缘特征.实验结果表明,在传统方法丢失或模糊边缘特征的情况下,采用文中方法也可以保持较好的边缘特征.

     

    Abstract: One of the important goals for transforming a color image to a gray one is to try to save as more visual characteristics of the original image as possible.The nonlinear global mapping method utilizes the trigonometric series approximation method, and achieves good results in many cases.By introducing t and t2 to mix with the trigonometric series where t is the corresponding parameter, this paper presents an improved nonlinear global mapping method.When the parameters of tand t2 are zero, the new method is degenerated into the traditional one.The new method achieves the final result by simultaneity optimizing the parameters of trigonometric series, t and t2.Experiments show the obvious improvement of the new method, i.e., even in the case that some feature is blurred or lost by using the traditional one, the new method can also achieve a fairly good result.

     

/

返回文章
返回