欢迎您！东篱公司
退出
-
申报数据库
-
立项数据库
-
成果数据库
-
项目获奖数据库
中文摘要：
针对电网中绝缘子破损与夹杂异物的不良状态图像检测问题。提出一种利用二代曲波变换和双正交小波变换检测图像的曲线状奇异性和点状奇异性的绝缘子不良状态定位检测方法。首先通过Radon变换确定图像中高灰度值直线的大致方向并以此方向分别旋转原图像：然后采用二代曲波变换定位绝缘子并微调图像旋转角度：最后利用二代曲波变换检测图像的曲线状奇异性判断绝缘子破损状态．并利用双正交小波变换检测图像的点状奇异性判断绝缘子夹杂异物状态。测试结果表明，该方法绝缘子识别效果良好，判断结果准确可靠，能达到绝缘子不良状态检测的目的。
英文摘要：
For the bad state image detection problem of breakages and ibreign body of the insulators in the grid, the bad state location detection method based on two generation curvelet transform and biorthogonal wavelet transform to detect the curve and point singularity of image is proposed. Firstly, Radon transform is adopted to find the picture＇s general direction of high gradation line, and the picture is rotated by that direction. Then, the insulators are located by the two generation curvelet transform, and the angle of picture is slightly adjusted; finally, breakage of the insulators is judged by the two generation curvelet transform which can detect curve singularity. The foreign body of insulators is judged by the biorthogonal wavelet transform which can detect punctiform singularity. The experiments show that the bad state of insulators can be correctly recognized with the proposed method, and the goal of insulators bad state detection can be achieved.
同期刊论文项目
同项目期刊论文
A New Classification Method for Transient Power Quality Combining Spectral Kurtosis with Neural Netw
期刊信息
位置：