高压输电线路航拍图像目标边缘检测

摘要

In this paper,image features are analyzed for high voltage transmission lines under various circumstances and then a method combining LoG operator and morphology is proposed.The gray image histogram shows that this method can not only present two peaks well,but also exhibit better performance finding the optimal threshold segmentation points.Compared with the improved Otsu (maximal variance between clusters) image segmentation algorithm,this method makes it easier to separate the target from the background.Matlab simulation results show that the proposed method can successfully detect the contour edge of the target in the image regardless of environment impact of high voltage transmission lines.It is still effective even under the influence of certain severe natural environment conditions.%对高压输电线路不同环境下的图像特征进行分析,提出了基于LoG算子与形态学融合的方法.通过灰度图像直方图表明该方法能够较好地呈现二峰性,更好地寻找最佳阈值分割点.该方法较以往提出的改进最大类间方差(Otsu)图像分割算法能更好地对背景和目标物进行分离.Matlab仿真实验表明,在高压输电线路环境及受一定自然环境条件的影响下,该方法能够更好地检测图像中目标物的轮廓边缘.