针对传统边缘检测算子适应性差的缺陷,提出一种基于全局优化相关系数的边缘检测算法。首先在综合多种常用边缘检测算子的基础上生成不同等级的边缘密度图;然后使用全局相关系数作为度量函数,计算边缘密度图与不同边缘检测算子的边缘输出图集之间的相关性;最后优化得到较为理想的边缘检测结果。实验结果表明该算法能融合多种边缘检测算子的优点,有较强的开放性与目标适应性,自动化程度高,工程实用性广。
According to weak suitability for traditional edge detectors,this paper proposed a new edge detection algorithm based on global optimization correlation coefficient.Firstly,obtained different level of edge density maps by combining output images of edge detection algorithms.Then using global correlation coefficient as metric function,calculated similarity between edge density maps and edge output images.Finally,extracted ideal edge detection result by optimizing the global correlation coefficients.The experimental results show that the algorithm integrates the merit of different edge detectors,possesses algorithm openness,object suitability and high automation,and has broad practical implementations.