基于统计的显微图像边缘检测阈值分析方法

针对复杂噪音干扰的显微图像, 提出一种新的基于统计原理的边缘检测阈值分析方法. 应用统计函数提取图像中梯度大的一个区域, 再由非极值抑制算法提取边缘像素. 在选择阈值前, 对每一像素的梯度值进行局部标准化, 去除模糊和全局阈值选择的不合适性. 所做分析的统计特性使得输入参数的选择对噪音图像具有鲁棒性, 能有效处理图像中的随机噪音. 实验结果表明, 所提出的算法具有稳定性强、 鲁棒性好的 特性.

Threshold Analysis of Micro Image in Edge Detection Based on Statistical Approach

An edge detection method of noisy micro image was proposed where threshold was performed via statistical principles. The region whose change rate of intensity is maximum was extracted by applying statistical functions. Then an edge pixel was detected from the extracted region with the non-maxima suppression algorithm. The gradient strength at each pixel was standardized locally before threshold was selected so as to eliminate the ambiguity and inappropriateness in choosing global threshold values. However, the values of the input parameters providing the appreciable results in the proposed detector were found to be more stable than other edge detectors and possess statistical interpretation, which can efficiently handle random noise present in an image. The results of the proposed algorithm were compared with those of many well-known edge detectors. The results suggest it can produce reliable, robust, and smooth edges.