基于形态学的复合绝缘子憎水性图像检测技术

Composite Insulator Hydrophobicity Image Detection Method Based on Morphology

为能快速、准确检测绝缘子憎水性等级，提出了一种基于形态学的图像处理技术的检测方法。该方法在复合绝缘子进行可见光图像分析中，首先通过形态处理方法实现对复合绝缘子表面覆水图像的优化处理，再利用改进形状因子法判断绝缘子憎水性等级。试验结果表明，利用该方法进行去噪处理可以填充由于反光造成的水珠图像孔洞及去除水膜边缘的毛刺，消除现场光线环境对图像的影响，可实现复合绝缘子憎水性的现场在线检测，准确地测量HCl～3、6～7级憎水性的复合绝缘子，提高复合绝缘子憎水性等级检测的正确性，明显增强复合绝缘子憎水性图像分析效果。

In order to test the hydrophobic grades of insulators rapidly and accurately, we proposed a morphology-- based image processing method. The method is for using in analysis of visible light images. The method includes optimizing the surface overlying water drops image of a composite insulator through morphological methods firstly and then determining the insulator＇s hydrophobic grade by using an improved shape factor method. Experimental results show that using this de-noising processing method can fill the holes in bead image, remove burrs on the edge of water, and eliminate the influence from background lights on visible-light-captured images which helps to accurately measure the grades of hydrophobic composite insulator within HCl - 3 and 6 - 7. It is concluded that using the method can improve the accuracy of hydrophobic-level detection of composite insulators and enhance the effect of analyzing the composite insulator hydrophobicity images.