直线特征标志子像素级定位方法的研究

Sub-pixel localization of target based on linear feature

在近景摄影测量中,常常需要对直线特征的标志进行定位,而标志定位的精度会影响到整个测量的精度。提出一种新的混合方法,即采用Hough变换对直线特征标志进行直线检测,得到直线初值;然后采用灰度矩边缘直线拟合定位法,对标志点中心进行子像素定位。在三维重建实验中采用所提出的灰度矩边缘直线拟合定位法和高精度直线定位法进行精度比较,实验证明经过Hough变换得到粗值的灰度矩边缘直线拟合定位法达到较高的定位精度,且计算更简单。

Targets based on linear feature often need to be localized in close-range photogrammetry. Howew er, the precision of localization will influence the precision of measurement at large. A new integrated algorithm is proposed in the paper and the procedure that is taken in the algorithm can be divided into two steps. Firstly, the rough location of the lines was extracted with Hough algorithm. Then straight line fitting based on Gray Moment Operators for Edge Location were adopted to extract the centre of target respectively. In the experiment for reconstructing object, the proposed method and High-Precision location algorithm for straight line were used, and the results of two algorithms were compared. The results of experiments show that the method for sub-pixel localization locates the target and obtains high precision and computes easily, after rough location using Hough transformation.