基于GLOH描述子和GVF Snake的绝缘子覆冰检测算法

摘要

利用计算机视觉和图像处理技术对绝缘子覆冰情况进行检测,在保障电力系统的安全运行方面具有非常重要的作用.提出了一种基于GLOH描述子和GVF Snake的绝缘子覆冰检测算法,该算法首先对输入图像进行预处理;其次提取感兴趣的目标区域,该区域内包含绝缘子信息;然后利用GLOH描述子来定位绝缘子,再利用GVF Snake模型来精确检测绝缘子轮廓;最后对绝缘子轮廓进行分析,通过计算最大径向距离来确定绝缘子是否存在覆冰情况,并在覆冰情况下计算覆冰的厚度.实验结果说明了此方法可以有效地判断出绝缘子是否存在覆冰现象.%There is a very important role in the protection of the safe operation of the power system by using computer vision and image processing technology to detect the situation of insulator icing.In this paper,an insulator icing detection al-gorithm based on visual dictionary and GVF Snake is proposed.Firstly,the input image is preprocessed.Secondly,the tar-get region of interest is extracted,and the insulator should be included in the region.Then,the GLOH descriptor is used to locate the insulator,and then the GVF Snake model is used to accurately detect the insulator contour.Finally,the contour of the insulator is analyzed,and the maximum radial distance is calculated to determine whether the insulator is icing or not. The thickness of icing is calculated in the case of icing.The experimental results show that the method can effectively judge whether there is icing phenomenon in the insulator or not.