A new denoising method for seismic planar attributes
Yang Yu-jie1, Qu Shou-li2, Yin Xing-yao1
1. College of Geo-Resources and Information,China University of Petroleum(East China),Dongying City,Shandong Province,257061,China;
2. Exploration and Production Research Institute,Sinopec,Beijing City,100083,China
Abstract:Regarding the denoising for the seismic planar attributes in the seismic facies analysis and reservoir prediction,this paper proposes a new denoising method called Factorial Kriging method.By calculating the statistical correlation,spatial correlation of regionalized variable to optimize the seismic attributes,the random noise and linear noise in the seismic planar attributes can be eliminated.The key idea for the new method is that the authors regarded the regionalized model as consisting of primary signals and noise factors and each factor has its own corresponding variogram.Based on the difference between primary signal and noise in variogram domain the noise is eliminated.Real data test shows that the new method works very well and the processing results from the new method are much better than using other regular denoising methods.