基于二次加权滤波的热设备故障区域定位研究

Fault area location study of thermal equipment based on quadratic weighted filtering

针对热设备在线故障区域定位对实时性和准确性要求较高的问题,提出一种基于中值滤波的二次加权快速滤波方法对热设备红外图像进行预处理。在较好滤除图像随机噪声的同时,提高了处理速度,更好地保留了图像细节。采用迭代法阈值分割和Roberts边缘提取算子相结合的方法,快速、准确地定位出了热设备故障区域所在,为今后进一步对故障区域识别、判断、分析和评价奠定了基础。

As locating the fault area of thermal equipments online has higher requirements to real-time and accuracy, this paper puts forward a quadratic weighted filter which based on the median filter to process infrared thermal image of ther-mal equipments. Using this algorithm, the random noise of the image has been filtered out effectively, and at the same time, this algorithm greatly enhances the real-time processing and retains the image details better. Through a combination of iterative threshold segmentation method and Roberts edge detection operator, the fault areas of thermal equipments can be located rapidly and accurately. And this lays the foundation for the further identification, judgment, analysis and evalua-tion of the fault areas in the future.