论文部分内容阅读
针对复杂背景下钻有圆孔的钢结构板图像,为实现圆孔中心的精确定位,根据近景摄影测量中二维直接线性变换的方法确定各圆孔对应的感兴趣区域,并分别对其进行二值化、形态学等处理,之后采用Canny算子检测圆孔像素级边缘,同时根据边缘轮廓所占面积作为条件,剔除错误边缘;为提高边缘点的定位精度,在像素级边缘点的基础上使用Zernike矩获取圆孔的亚像素级边缘点,最后对边缘点集用最小二乘的方法进行多次椭圆拟合,从而确定圆孔中心在图像中的精确位置.试验结果表明,该方法可以获得较高的孔心影像坐标量测精度,点位“均方差”达到0.182mm.
For images of steel structural plates drilled with holes in complex backgrounds, in order to achieve accurate centering of circular holes, the corresponding regions of interest for each circular hole were determined according to the method of two-dimensional direct linear transformation in close-range photogrammetry, and they were separately performed. Binarization, morphology, etc., then use Canny operator to detect the pixel-level edge of the hole, at the same time according to the area occupied by the edge contour as a condition, remove the error edge; in order to improve the positioning accuracy of the edge point, the basis of pixel-level edge point The Zernike moments are used to obtain the sub-pixel edge points of the circular hole. Finally, the least squares method is applied to the edge point set to perform multiple ellipse fittings to determine the exact position of the center of the hole in the image. The experimental results show that this method Achieve higher accuracy of the coordinates of the image of the hole center, and the point “mean square deviation” reaches 0.182mm.