基于机器视觉的瓦斯继电器油位异常检测

Abnormal Oil Level Detection of Gas Relay Based on Machine Vision

摘要:

针对传统的瓦斯继电器油位异常时机械故障引起的误报警情况，文中提出基于机器视觉的瓦斯继电器油位测量方法，为瓦斯继电器油位异常检测提供一种新的方法，有效提高其工作可靠性和安全性。由于瓦斯继电器窗口内部背景复杂干扰较大，使得传统的基于图像处理液位提取方法误差很大，无法得出满意的结果。为此提出颜色空间域转换和Canny检测边缘结构复杂度相结合的算法，提取出瓦斯继电器油位。实验结果显示，所提算法在瓦斯继电器油位提取方面优于传统方法，满足实际工程需求。

Abstract:

Aiming at the false alarm caused by mechanical fault when the oil level of traditional gas relay is abnormal, a method for measuring the oil level of gas relay based on machine vision is proposed, which provides a new method for detecting the abnormal oil level of gas relay and effectively improves its working safety. Due to the complex background and large interference in gas relay window, the traditional liquid level extraction method based on image processing has a large error and can not obtain satisfactory results. Therefore, an algorithm combining color space domain conversion and structure complexity of Canny edge detection is proposed to extract the oil level of gas relay. The experimental results show that the proposed algorithm is better than the traditional method in the extraction of oil level of gas relay, and can meet the actual engineering needs.