绝缘子憎水性图像检测方法研究

Research on Measurement of Hydrophobicity of Insulators

绝缘子表面憎水性检测是判断绝缘子性能的主要手段,也是确保绝缘子安全运行的重要保证。该文把图像处理技术引入到绝缘子憎水性检测中,通过对绝缘子图像的对比度增强、图像的灰度、梯度及度的信息特征的综合考虑、模糊C均值聚类(fuzzy C-means,FCM)和形态学对憎水性图像进行分割,获取大水珠(或水迹)形状信息,最后用最大面积比法判定绝缘子的憎水性等级。实验结果表明,该算法能够有效地提取出水珠(或水迹)的形状信息、克服传统目测法的主观性,检测结果的准确性满足实际检测的需要。

Measurement of hydrophobic is not only a main way to evaluate the performance of insulators, but also an important factor to ensure the safety of insulators. In this paper, image processing technology is used to measure the hydrophobicity insulators. Through considering image intensity, changes of gray value, gradient value and degree, image contrast enhancement, fuzzy C-means （FCM） and morphological algorithms are used to segment the insulators. Finally, the largest area ratio algorithm is adopted to confirm the level of hydrophobicity. Experimental results show that the proposed algorithm is efficient to extract shape information of droplets （or watermarks） and obtain better performance than traditional methods.