Corner detector based on global and local curvature properties
Corner detector based on global and local curvature properties
复制标题
DOI：
10.1117/1.2931681
复制
发表时间：
2008-05-01
影响因子：
1.3
通讯作者：
Yung, Nelson H. C.
中科院分区：
文献类型：
--
作者：
He, Xiao Chen;Yung, Nelson H. C.
本文提出了一种基于曲率的角点检测器，可以以较低的计算成本准确地检测精细和粗糙的特征。首先，它从Canny边缘图中提取轮廓。其次，在低尺度下计算轮廓上每个点的曲率绝对值，并将曲率绝对值的局部极大值作为初始角点候选。第三，使用自适应曲率阈值从初始列表中去除圆角。最后，由于量化噪声和琐碎的细节，通过评估角点候选人的角度在一个动态的支持区域，消除假角点。提出的检测器进行了比较，与流行的角点检测器的平面曲线和灰度图像，分别在主观的方式，以及与特征对应性测试。结果表明，所提出的检测器在这两个领域都表现得非常好。(c)2008年，由光学仪器工程师协会（Society of Photo-Optical Instrumentation Engineers）主办。
This paper proposes a curvature-based corner detector that detects both fine and coarse features accurately at low computational cost. First, it extracts contours from a Canny edge map. Second, it computes the absolute value of curvature of each point on a contour at a low scale and regards local maxima of absolute curvature as initial corner candidates. Third, it uses an adaptive curvature threshold to remove round corners from the initial list. Finally, false corners due to quantization noise and trivial details are eliminated by evaluating the angles of corner candidates in a dynamic region of support. The proposed detector was compared with popular corner detectors on planar curves and gray-level images, respectively, in a subjective manner as well as with a feature correspondence test. Results reveal that the proposed detector performs extremely well in both fields. (c) 2008 Society of Photo-Optical Instrumentation Engineers.