|
<< Click to Display Table of Contents >> Navigation: Velocity > Smooth 3D EXT |
Smooth 3D EXT applies a three-dimensional running-average (box) smoothing filter to seismic gather data. The filter operates independently along the inline, crossline, and vertical (depth/time sample) axes, with configurable half-widths in each direction. An optional adaptive (dynamic) window mode allows the smoothing extent to vary spatially, and special handling is provided for null or ignore values to prevent them from contaminating the smoothed output.
This module is commonly used to smooth velocity models, attribute volumes, or migrated seismic cubes to suppress high-frequency spatial noise before display or further analysis. The Information panel reports the number of inlines and crosslines that were processed.
The standard seismic data container. Connect from the upstream module whose output requires smoothing.
SEG-Y file handle for reading data when operating on file-based input.
Trace header vector providing inline and crossline coordinates used to organise traces into the 3D grid for smoothing.
Gather data to be smoothed. Connect from the upstream module.
Stack line geometry definition for 2D line processing context.
Crooked line geometry for 2D surveys with non-straight acquisition lines.
3D bin grid defining the inline/crossline layout used to map traces into the smoothing volume.
Pre-sorted trace index used to access data in inline/crossline order required by the 3D smoothing algorithm.
Half-width of the smoothing box in the inline direction, measured in traces. Default: 50 traces. The total inline smoothing aperture is 2 × this value + 1. Larger values produce smoother output but blur lateral detail. Reduce for tighter lateral resolution.
Half-width of the smoothing box in the crossline direction, measured in traces. Default: 50 traces. Set independently from the inline direction to accommodate anisotropic acquisition grids.
Half-width of the smoothing box in the vertical direction, measured in samples. Default: 50 samples. Set to zero to disable vertical smoothing and apply only lateral smoothing.
First sample index (0-based) from which smoothing begins in the vertical direction. Default: 0 (from the top of the trace). Increase to protect shallow samples (e.g., water column or near-surface noise zone) from being affected by the smoothing filter.
Number of samples over which the smoothing gradually tapers on from zero to full strength at the start of the smoothed zone. Default: 0 samples (no taper). Apply a taper to avoid a hard amplitude step at the Smooth start sample boundary.
Container holding the ignore value handling parameters. Samples that equal the Ignore value are excluded from the smoothing average so they do not bias the result, and optionally filled with interpolated values.
When enabled (default), locations that originally held the ignore value are filled with the smoothed value computed from their valid neighbours. Disable to leave null sample locations unchanged in the output.
Amplitude value used to mark null or undefined samples in the input data. Samples equal to this value are excluded from the smoothing calculation. Set to the actual null value used in your velocity model or attribute volume (e.g., 0.0 or a large negative sentinel value).
When enabled, the smoothing window size adapts spatially based on data density rather than using fixed half-widths. Default: off. Enable when the data has highly variable trace coverage and a fixed window would produce artefacts near gaps.
Maximum number of expansion steps allowed when Use dynamic window is active. Default: 5. Larger values allow the adaptive window to grow further in data-sparse regions to find enough valid neighbours.
When enabled, input data items are automatically connected from the nearest upstream module with matching output types.
Selects the processing device. CPU is the standard choice for this smoothing operation.
Configuration for distributed (cluster) execution.
Minimum number of gathers sent per work chunk during distributed execution.
When enabled, restricts CPU thread count per cluster node to the value in Number of threads.
Optional text appended to the distributed job name.
When enabled, applies the CPU affinity mask specified in Affinity.
CPU affinity mask applied when Set custom affinity is enabled.
Maximum number of CPU threads used during execution.
When enabled, this module is bypassed and data passes through unchanged.
The smoothed seismic data container. Pass to downstream modules for display or further processing.
SEG-Y file handle for writing the smoothed output to disk.
Trace headers corresponding to the smoothed output traces.
Gather containing the 3D-smoothed trace data.
Output stack line geometry passed through from input.
Crooked line geometry passed through from input.
3D bin grid passed through from input.
Sorted header index passed through from input.
Number of inline lines processed by the smoothing filter. Read-only, updated after each run.
Number of crossline lines processed by the smoothing filter. Read-only, updated after each run.