At present, human beings are increasingly dependent on oil and gas energy. Due to the influence of a series of factors such as exploration technology and natural environment, in the process of actual oil and gas exploration technology, the collected seismic data usually has the problem of low sampling rate. The lower sampling rate of seismic data may cause spatial aliasing in migration imaging, resulting in inaccurate exploration. Therefore, in order to carry out more accurate oil and gas exploration, the seismic signal enhancement method also puts forward higher requirements. In this paper, the seismic signal enhancement will be realized by seismic data channel interpolation. A seismic trace interpolation method based on generative adversarial network is proposed to realize it. The traditional seismic trace interpolation is generally carried out based on complex mathematical transformation or certain assumptions, in order to solve the problems existing in the traditional seismic trace interpolation method. This paper will use the generative adversarial network model in deep learning to combine the generative adversarial network with the residual network, and the loss function is the sum of the Wasserstein distance and the content loss. The optimized generative adversarial network can better realize the interpolation of seismic traces, and the spatial aliasing frequency has been suppressed. Therefore, the enhancement of seismic signals can be better achieved, and theoretical support for accurate exploration can be provided.

Wang Jinhuan, Ang Ling Weay, Sellappan Palaniappan. Research on Seismic Signal Enhancement Method Based on Deep Learning. Academic Journal of Computing & Information Science (2022), Vol. 5, Issue 11: 46-49. https://doi.org/10.25236/AJCIS.2022.051107.

[1] Zha Quanheng. Oil and gas exploration and development efforts need to be improved [N]. China Energy News, 2019-02-18 (004)

[2] Jiang Jinsheng, Ren Haoran, Li Hanye. Seismic trace interpolation method based on convolutional autoencoder [J]. China Petroleum Society 2019 Proceedings of the Annual Geophysical Technology Seminar, 2019.

[3] Y. Li, S. Liu, J. Yang, et al. Generative face completion [C]. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2017: 3911-3919.

[4] C. Ledig, L. Theis, F. Huszr, et al. Photo-realistic single image super-resolution using agenerative adversarial network [C]. Proceedings of the IEEE conference on computer vision and pattern recognition. 2017: 4681- 4690.

[5] H. Zhang, T. Xu, H. Li, et al. Stack GAN: Text to photo-realistic image synthesis with stacked generative adversarial networks [C]. Proceedings of 2017 IEEE International Conference on Computer Vision. Venice, Italy. 2017. 5908-5916.