谱聚类图像分割在绝缘子憎水性批量检测系统中的应用

Application of spectral clustering image segmentation in insulator hydrophobicity batch detection system

针对现有的复合绝缘子憎水性检测效率较低且判定不准确的问题,设计了一个基于谱聚类算法的憎水性批量检测系统。检测装置主要由环形灯、二轴运动平台及底部的旋转托盘组成。二轴运动机构上安装有摄像头和喷头,底部有一个可放置4个绝缘子的旋转托盘,通过步进电机作分度运动。经过一系列的预处理,有效减少了环境对绝缘子图像质量的影响。根据谱聚类算法对预处理后的图像进行分割,然后获取最大水珠,以改进的形状因子法为基础,供上位机进行憎水性判定。经过仿真实验可知,相比于FCM、K-means等方法,该方法针对绝缘子图像有较好的分割效果,且该系统既能有效地提高检测的准确率和效率,又能实现批量检测。

Aiming at the problem that the existing composite insulators have low hydrophobic detection efficiency and inaccurate grade determination,a hydrophobic batch inspection system was designed.The detection device is mainly composed of a ring light,a two-axis motion platform,and a rotating tray at the bottom.The two-axis motion mechanism is equipped with a camera and a nozzle.The bottom has a rotating tray for placing four insulators.The stepping motor is used for indexing movement.After image graying,histogram equalization,filtering,morphological processing and other pre-processing,reducing the impact of the environment on the image quality of the insulator effectively.The pre-processed image is segmented according to the spectral clustering algorithm,then obtained the maximum water droplet.Based on the improved shape factor method,the host computer is used for hydrophobic detection determination.The simulation experiment shows that compared with FCM,K-means and other methods,this method has better segmentation effect for the insulator image.The system can improve the accuracy and efficiency of detection effectively,and also can realize batch detection.