子束变换是一种在多个尺度上对图像线段求积分的线段检测算法。针对二值图像中的线性特征,在强杂波环境下,研究了子束变换的特征检测问题。综合利用了不同尺度上的子束特性,引入线性插值、新的目标函数和Chebyshev不等式,提出了一种自适应线性特征检测算法。仿真结果表明,该算法对杂波环境下的线性特征检测非常有效,并对短线干扰有一定的抑制能力。
Beamlet Transform is a linear feature detection technique based on multiple scales. Linear feature detection is studied in this paper concerning features in binary images. An adaptive linear feature detection method based on Beamlet transform is proposed which makes full use of the characteristics of beamlets of each scale and utilizing linear interpolation, a Experiment results verify that the proposed algorithm embedded in noisy environment; meanwhile, it's resilient new object function and Chebyshev inequation. is quite efficient in detecting linear features to linear clutters.