该算法以原始图像与分割后图像的相关系数最大为准则，选取最优阈值对磨粒图像进行二值化。
An optimum threshold is selected by maximizing the correlation between the original halftone image and the thresholded bi-level image.
该准则能有效抑制图像中高频信息的存在对相关系数计算的影响，获得准确的最优停止时间，并无需预先知道图像中的噪声水平。
This criterion can suppress the affection of minutiae to correlation coefficient and get a more accurate stopping time without knowing the noise variance.
此外提出另一类分割判定准则——最大相关系数判定准则。
Based on the three different image segmentation results, this paper finally shows that the proposed algorithm and maximum correlation coefficient have significan.
提出一种基于相关系数矩阵判断准则的经验模态分解去除趋势项的方法。
This paper proposes a method to remove trends based on correlation matrix by empirical mode decomposition(EMD).
实验结果表明该方法与最小均方误差准则、信噪比准则和相关系数最小准则相比，具有更好的稳定性和视觉特性。
The experimental result shows that the new method has better stability and visual features compared with MSE, SNR and minimum correlation coefficient.
实验结果表明该方法与最小均方误差准则、信噪比准则和相关系数最小准则相比，具有更好的稳定性和视觉特性。
The experimental result shows that the new method has better stability and visual features compared with MSE, SNR and minimum correlation coefficient.
应用推荐