摘要
角点是图像目标的重要局部特征,角点检测在目标跟踪、运动估计、相机定标等方面都有重要的应用.曲波变换基于多尺度几何分析的思想,以边缘为基本元素能够同时获得对图像平滑区域和边缘部分的稀疏表达,是一种更适合图像处理特点的多尺度变换.为降低曲波的数据冗余性,采用有限脊波变换对曲波进行改进,并在此基础上提出利用Mallat模极大值原理进行角点检测的算法,实验结果表明该方法即使在有噪声的情况下仍然具有较高的稳定性和鲁棒性.
Corner is a significant local feature of images. Corner detection in images is important for a variety of image processing tasks including tracking, image registration, change detection, determination of camera pose and position and a host of other applications. The aim of curvelet transform is to find a kind of optimal representation of such type of image in the sense of nonlinear approximation. In this paper, proposed on improved curvelet transform through using finite ridgelet transform. And based on the improved curvelet transform, a more robust corner detection approach combined with Mallat local maxima modulus is proposed. Experimental results show the efficiency of this algorithm even in noised scene.
出处
《小型微型计算机系统》
CSCD
北大核心
2008年第11期2141-2144,共4页
Journal of Chinese Computer Systems
基金
国家自然科学基金项目(60453001)资助
关键词
角点检测
曲波变换
有限脊波变换
corner detection
curvelet transform
finite ridgelet transform