Curvelet domain Radon transform noise suppression method for loess tableland region
A noise suppression and wave domain technology, applied in the field of Radon transform noise suppression in the curvedlet domain, can solve the problems of poor effect, unsatisfactory denoising effect, difficult to achieve seismic data processing, etc., and achieve the effect of improving the signal-to-noise ratio.
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0053] Example 1. The Radon transform noise suppression method in the curve wave domain for the loess plateau area includes the following steps:
[0054] Step 1: Perform curvelet forward transformation on the original seismic data to obtain the matrix of curvelet coefficient fields in different directions and scales. At this time, the interference wave with better linear characteristics in the curvelet coefficient field still maintains a good Linearity, while the effective waves are spread out in the matrix;
[0055] Step 2: Perform Radon forward transformation on each scale-angle matrix one by one, transform the data in the curvelet coefficient domain to the Radon domain, and perform threshold filtering on each scale-angle matrix one by one; since the interference wave appears as a point with strong energy or energy in the Radon domain Random noise shows unfocused energy, while effective wave does not focus, but still shows greater focused energy than random noise. Through thr...
Embodiment 2
[0093] Example 2. The Radon transform noise suppression method in the curve wave domain for the loess plateau area includes the following steps:
[0094] ① Firstly, curvelet forward transformation is performed on the original seismic data to obtain matrices of different directions and scales. In the curvelet domain, the interference wave with better linear characteristics still maintains better linearity in the matrix, while the effective wave is dispersed;
[0095] ② Perform Radon transformation on different small matrices, transform the data in Curvelet domain to Radon domain, and set the threshold, set the data greater than the threshold to zero, and keep the data less than the threshold unchanged. Since the interference wave appears as a point or energy group with strong energy in the Radon domain, the random noise appears as unfocused energy, while the effective wave is not focused, but still shows a larger focused energy than random noise. Through processing, the interf...
Embodiment 3
[0125] Example 3. In this embodiment, a two-dimensional seismic data is taken as the target area, and the method is used to process the data, so as to verify the effect of the method. The actual data is collected by two-dimensional single line, the observation system is 6090-110-20-110-6090, the seismic data time length is 6000ms, the time sampling interval is 2ms, the number of sampling points is 3000, and the number of each shot is 408. Use the above method to process the data.
[0126] (1) First step 1, analyze the typical single shot in the work area and select appropriate processing parameters.
[0127] Curvelet forward transform was performed on the single-shot records at different positions in the work area ( figure 1 It is a schematic diagram of curvelet transform, in which the left figure is a detailed division of seismic signals in multiple dimensions during the process of curvelet transform, in which the four sides represent the scale, and the angle represents the...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More