高空悬挂式相机拍摄中的拉索提取方法

Method for extracting cable in aerial image

为了解决高空悬挂式相机拍摄航拍图像中的拉索对图片产生的影响,提出一种拉索提取方法。该方法首先分别用Canny算子和小波变换处理图像,然后用小波变换提取的垂直方向的高频分量与Canny算子得到的边缘图像做交集,对得到的交集图像经形态学膨胀处理后,叠加到Canny检测得到的边缘图上;在此基础上,通过形态学腐蚀滤除复杂的图像环境中的大量背景图案,保留直线方向的特征,再在概率霍夫直线检测的基础上,利用聚类拟合更精确地找到待提取的拉索内容。实验结果表明,所提出的拉索提取方法能从复杂的背景中完整地提取出目标拉索。

In order to solve the influence of the cable captured in the aerial image on the picture,a cable extraction method is proposed.This method processes images with Canny operators and wavelet transforms firstly,then uses wavelet transform to extract the high-frequency component in the vertical direction,and intersect it with edge images obtained by Canny operators.After morphological dilated,the obtained intersection image is superimposed on the edge image detected by the Canny operator;on this basis,the morphological erosion is used to filter out a large number of background patterns in a image which contains complex environment,thereby preserving the features of the straight line direction.Based on the probabilistic Hough straight line detection,the content of the cable to be extracted can be found more accurately by using the cluster fitting.The experimental results show that the cable extraction method proposed can completely extract the target cable from a complex background.