【摘要】 由于传统基于拟合的亚像素边缘算法的模型较复杂且求解时间较长,文章提出了一种基于反正切函数拟合的亚像素边缘检测算法。首先,利用根据提出的一维反正切边缘模型,通过最小二乘拟合给出一维边缘检测方法;随后,利用矩的旋转不变性理论计算边缘投影方向,可以将二维边缘曲面投影为一维边缘曲线,从而将二维边缘检测问题转化为一维边缘检测。实验结果表明,在不影响精度的前提下,文章提出的方法的运算时间要低于传统基于拟合方法,如基于高斯模型的检测算法。因此,基于反正切函数拟合的亚像素边缘检测算法基本满足图像测量的稳定可靠、高精度、强实时性要求,具有较好的实用性。更多还原

【Abstract】 Because of the complexity of the traditional sub-pixel edge algorithm, a sub-pixel edge detection algorithm based on the fitting of the tangent function is presented in this paper. Firstly, the one-dimensional edge detection method is given using the least square fitting algorithm based on one-dimensional tangential edge model; then, the edge projection direction is solved by using the rotation invariant theory of moment, and the 2D edge curve can be projected into a one-dimensional edge curve. By this way, the two-dimensional edge detection problem is transformed into a one-dimensional edge detection. The experimental results show that the run time of the sub pixel edge detection algorithm based on the arctangent function fitting is faster than that based on the Gaussian model. Therefore, the proposed algorithm can basically meet the requirements of stability, reliability, high-precision, and strong real-time.更多还原