The Denoising Algorithm Based On Compressed Awareness In The Application Of The Seismic Random Noise Suppression

As the increasing global demand for oil and gas and mineral resources of the unknown oil and gas fields and the development of mineral resources become the key points and difficulties in seismic exploration.However, uncertain factors such as buried underground structure and resource conditions for exploration work has brought many difficulties.This requires that we have effective means of seismic data processing, can guarantee of seismic data collected from seismic exploration process found more effective information, so as to provide beneficial basis for the development of new resources, unknown buried conditions and exploration in complex environment makes the collected seismic data usually contain a lot of noise, with low signal-to-noise ratio, effective information which is difficult to identify, we usually seismic noise can be divided into two categories: random noise and the noise.The research object of this paper is the random noise in land seismic exploration, random noise is irregular, no law, and between adjacent channel is mutually related to each other, they do not have a fixed frequency, distribution almost the whole frequency band, the serious influence the signal-to-noise ratio of the seismic record, and eventually affect the judgment of the oil and gas.Actual seismic data due to noise interference, seriously affect the processing and interpretation of the work, including data reconstruction.At the same time, the traditional denoising method has some deficiencies in the process of random noise suppression, such as: transform domain threshold method, the seismic wave near the former leads to abnormal phenomenon is not smooth, affect the quality of seismic data.To effectively improve the signal-to-noise ratio of seismic records, this paper will be based on compressed sensing theory of gradient projection sparse reconstruction(GPSR) algorithm is introduced into the random noise suppression of seismic exploration. This method will first with noise signal sparse representation by compression perception theory, on this basis, through the GPSR algorithm to reconstruct the signal. Refactoring process can be regarded as morbid matrix solving process, due to the regularization constraint, in the process of reconstruction can effectively suppress the noise signal in noise. Test results show that the experimental data compression perception denoising algorithm can effectively suppress random noise in seismic records, the restoration of the submerged by noise in phase axis information, at the same time, the denoising algorithm is proposed in this paper with the classic wiener filtering algorithm is analyzed, on the denoising performance is better than that of wiener filtering algorithm.