摘要
为了使彩色图像的边缘检测更符合人眼的视觉特征,避免出现过检、漏检等问题,文中构建了一个适应人眼视觉特性的改进型CIELab色彩模型,结合局部背景亮度以及不同空间频率的颜色对人眼视觉效果的影响,对比灵敏度函数构建动态阈值公式,使得所设计的算法具有人眼的视觉特性,抑制噪声等特点。最后通过对算法的时间复杂度、边缘的4-连通成分数和8-连通成分数等参数来评价边缘的连续性,测试表明该算法能准确检测出彩色图像的边缘且效果更佳。
Considering the effect on the color of the sensitivity and spatial frequency to human visuals,this paper constructs a new CIELab color model,combined with the effects of local background brightness and color of different spatial frequencies,as well as using contrast sensitivity function to build dynamic threshold formula,so the algorithms can suppress noise,avoid over-inspection and undetected problems.Finally the tests show that the algorithm can accurately detect the edge better,which evaluates continuity of the edge by analyzing time complexity,4-connecting factor and 8-connecting factor of the edge.
出处
《信息技术》
2015年第8期204-207,210,共5页
Information Technology
关键词
边缘检测
动态阈值
对比灵敏度
edge detection
dynamic threshold
contrast sensitivity
