基于YCbCr和Ostu算法的电力热故障区域提取

Thermal Fault Zone Extraction of Electrical Equipment in Infrared Images by YCbCr and Ostu Algorithm

针对电力设备热故障区域定位问题,提出一种融合YCbCr(一种色彩空间)、色差法和Ostu(大津法)算法的红外图像提取方法。首先将原始红外图像从RGB(Red Green Blue)色彩空间转换到YCbCr色彩空间,完成初始前景分割;然后对分割完的图像进行红绿色差灰度化,突出故障区域;最后采用Ostu算法寻找最优分割阈值来对灰度图作图像分割,实现热故障区域的提取。借助Matlab对常见的电力热故障红外图像进行了实验测试,实验结果表明:该算法能够有效提取红外图像中的热故障区,为后续的故障特征提取与识别奠定基础;与传统算法对比,该算法具有更加精确、全面、完整等优点。

Aiming at the thermal fault zone in power equipment,a infrared image extraction method with the fusion of YCbCr,aberration and Otsu was proposed.Firstly,the original infrared image was transformed from RGB color space to YCbCr color space so as to finish the foreground segmentation.Secondly,the red-green difference graying method was used to process the segmented image.Finally,the extraction of thermal fault zone was realized by processing the graying image with Ostu algorithm.Matlab was adapted to run the experiment of the stated algorithm.The experimental results showed that,the method has the ability to obtain better extraction performance.It lays foundation of future recognition and extraction of the thermal fault.Compared with traditional algorithems,this method is of more accuracy,comprehensiveness and integrity.