机构地区: 湖南大学电气与信息工程学院
出 处: 《湖南大学学报(自然科学版)》 2004年第4期28-33,37,共7页
摘 要: 提出了一种基于多重分形频谱分析的图像边缘检测方法.该方法首先计算出图像像素点的粗粒化H lder指数,并用核估计方法估计出图像与该粗粒化指数相对应的多重分形奇异频谱,然后根据图像的多重分形频谱值就可以检测出图像的边缘点.实验结果表明该方法是有效的,而且与传统方法相比具有更好的局部性. An image edge detection method based on multifractal spectrum analysis was presented. The coarse exponent of the image pixels was first computed. Then its multifractal spectrum was estimated in the kernel estimation method. Finally the image edge detection was done by means of different multifractal spectrum values. Experiment results have shown that this method is efficient and more local compared with traditional methods.
领 域: [自动化与计算机技术] [自动化与计算机技术]