基于图像处理技术的电能表外观检测系统

Electricity Meter Visual Inspection System Based on Image Processing

摘要:

为适应智能电能表自动化检定的要求,设计了由图像处理技术完成外观检测功能的系统,实现了外观检查的自动化,解决了人工检测方法工作量大、效果不佳的问题;系统采用高性能CCD相机完成图像采集,确保图像信息丰富完整；利用图像平滑去噪、形态学处理等技术完成图像的预处理,大幅降低后续比对的难度；运用基于高斯金字塔的混合配准算法完成图像配准,最终实现待检图像与模板图像之间匹配程度的检测；该方法检测时间短,正确率高,已成功应用于省级计量中心电能表自动化检定线,满足生产需求。

Abstract:

To meet the demand of smart electricity meter automatic calibration, a visual inspection system based on image processing is designed, which achieves automatic visual examination and solves the problems of high labor intensity and low efficiency in manual inspecting method. The system uses high-performance CCD camera to collect photos to ensure the plenty and wholeness of images information. Besides, it adopts image preprocessing such as smoothing and morphological technology to decrease degree of difficulty in follow-up comparison. Also, mixed registering algorithm based on Gauss pyramid is employed in image registration, which finally serves for the inspection of matching degree between sensed image and template image. This method costs short inspecting time and has high accurate rate, which has been successfully applied in electricity meter automatically calibrating line of provincial-level measurement center and meets producing need.