基于可伸缩型注意力机制的神经网络地震数据去噪方法

张敏, 许一卓, 易继东

A method for seismic data denoising based on the neural network with a retractable attention mechanism

ZHANG Min, XU Yi-Zhuo, YI Ji-Dong

图7 3种去噪方法在合成地震数据中的测试结果 a—DnCNN去噪结果;b—DnCNN去除的噪声;c—a和b的局部相似图;d—Unet去噪结果;e—Unet去除的噪声;f—d和e的局部相似图;g—STCNN去噪结果;h—STCNN去除的噪声;i—g和h的局部相似图

Fig.7 Results of three denoising methods in synthetic seismic data a—DnCNN denoising result;b—noise data removed by DnCNN;c—similarity map between panels fig.a and fig.b;d—Unet denoising result;e—noise data removed by Unet;f—similarity map between panels fig.d and fig.e;g—STCNN denoising result;h—noise data removed by STCNN;i—similarity map between panels fig.g and fig.h