Support Center Email Support Knowledge Base Videos
Geoscience > Main Launcher > Horizon Menu > 3D Auto Picker

3D Auto Picker

The 3D Auto Picker dialog allows you to perform rapid auto picking on an entire 3D volume. The picker is a single seed auto picker, so requires only a single pick to run. Superior performance is generally obtained by picking on several seed lines. It is also possible to modify your seed picks and then quickly rerun the auto pick if you keep the dialog open. The data is only reloaded if a new 3D survey has been selected, or if the vertical start or end have been modified. Multiple iterations can be completed very quickly by setting an appropriate start and end time and then using the Apply button. Changes can be made without exiting the Auto Picker, so in areas of poor results, you can add or modify the seed picks to improve the final result.

You can auto pick either using a Seed Horizon, or a Seed Wavelet. If setting up the auto pick with a Seed Horizon, do not check on the Wavelet Correlation. If using a Seed Wavelet and the Wavelet correlation is checked on, the Seed Horizon will not be used. 

When using a seed horizon, you can use all of the parameters (other than Wavelet Correlation) to restrict the picking. Then using the picks on the seed horizon, the event will be flood-filled using the restrictions set. When using a seed wavelet, only one wavelet  can be selected and the pick will be flooded out from the selected wavelet, as far as it can go using the picking parameters.

General Information

3D Seismic Files

Select from a list of the 3D seismic datasets in the project. The working set version is listed, but alternate datasets can be selected by clicking on the Dataset cell. To help you make your line selection, you can sort any of the columns by clicking on the column header, you can apply a filter (), or you can type a search string into the search bar to limit the information in the list based on Line, or Dataset information. 

When you select a file the Data Loading Extents will be automatically populated with the full extents of the dataset.

For additional selection and navigation options right click on the 3D Seismic Fileslist to access the shortcut menu.

Data Loading Extents

These are the extents of data to load. They represent a cube of data which is loaded into memory for the picking operation. Once data is loaded into memory, the auto picking is performed very quickly. If overwriting an existing horizon, it will only overwrite the horizon within the defined extents and any picks existing outside of the extents will be left untouched.

Start Trace: The lowest trace value of the volume to load.

End Trace: The highest trace value of the volume to load.

Start Line: The lowest line value of the volume to load.

End Line: The highest line value of the volume to load.

Vertical Start: The minimum vertical (i.e. time) value of the volume to load. Note: It is best to modify this value to be closer to your zone of interest. This will reduce the memory requirements of the auto picker which will improve performance.

Vertical End: The maximum vertical (i.e. time) value of the volume to load. Note: It is best to modify this value to be closer to your zone of interest. This will reduce the memory requirements of the auto picker which will improve performance.

Depths are Subsea: Check this item to specify depth extents (vertical start and vertical end) as subsea values. When this item is not checked you will specify depths as true vertical depth. This parameter has no effect on time volumes.

Select From Map: This provides an easy way to select the 3D volume from the map. Click on the button and then go over to the map and select a 3D dataset just as you would select a slice for display. After selection the Data Loading Extents will be filled in. This provides a quick way to auto pick a subset of the full 3D. If you have hit this button in error, click on it again and you will cancel the selection process.

Constraining Polygons: Open the Select Polygons dialog and select inclusion or exclusion polygons to restrict the geographical extents of the 3D auto pick.

Stop at fault surfaces: This directs the auto picker to stop trying to pick across a fault triangle boundary.

Auto Picking Input and Output Parameters

Seed Horizon: This specifies the horizon to use as input for the auto pick process. All picks in the seed horizon will be used as input to the process, and none of the original seed picks will be modified by the process.

Output Horizon: Specify the horizon to write out after the auto pick has been performed. By default  the horizon has the same name as the seed horizon with a " AP" suffix added to it. You may select another existing horizon from the list if you choose, and you can even choose to overwrite the seed horizon with the resulting picks.

Search Gate: Specify the distance in vertical units (typically milliseconds) above and below an existing pick to try and perform the auto pick on the adjacent traces. For example, a value of 20 will search 20 ms above the pick and 20 ms below the pick for the adjacent pick.

Search Mode: Specify how to decide between multiple events within the search gate. This is only relevant for peak and trough picking modes.

    Largest: Find the largest event within the search gate as the resulting pick.

    Closest: Find the closest event within the search gate as the resulting pick.

Event Type: Set the type of event to look for.

    Peak: A peak event is an event surrounded by lower amplitudes. It need not be a positive value.

    Trough: A trough event is an event surrounded by higher amplitudes. It need not be a negative value.

    Zero Crossing: A zero crossing event is an event that crosses from negative to positive or positive to negative values. The zero crossing is picked sub-sample.

    Peak/Trough Zero Crossing: This is an event that crosses from a positive to negative value. The zero crossing is picked sub-sample.

    Trough/Peak Zero Crossing: This is an event that crosses from a negative to positive value. The zero crossing is picked sub-sample.

Amplitude Correlation

Search Direction:  This specified the direction to search out from a seed pick to attempt an auto pick.

    Perpendicular Only: Search only in the inline and crossline direction from a seed pick. This is often the best choice when you have already picked the horizon on several inlines or crosslines.

    Perpendicular & Diagonal: Search in the inline and crossline direction, as well as diagonally out from a given seed pick. This is a more suitable choice when picking from a single seed.

Amplitude sorting: If this item is checked, the auto picker will sort amplitudes after each pick, and choose the peak or trough event with the highest value as the next pick. This is very effective when trying to follow a very strong event. It has no effect on zero crossing events.

Amplitude Correlation: Specifies how to correlate the amplitude of a prospective pick against existing picks before performing the pick. Using this option helps prevent the auto picker from jumping to a different event. This is only meaningful for peak and trough events.

    None: Do not perform any correlation, just make the pick.

    Seed Trace: Correlate against the seed trace only.

    Trace Group: Correlate against the 8 picks surrounding a potential pick. The amplitude cutoff must be met for all 8 picks for the pick to be made. This is much more restrictive than the Seed Trace method, but is less likely to jump to another event.

Amplitude Cutoff: Specify the lower cutoff for rejecting potential picks when using amplitude correlation. For example, a value of 80 means that any pick that has an amplitude value less than 80 % of the picks it is correlating against will be rejected.

Wavelet Correlation

When selected, a wavelet is used for the auto picking. The seed horizon is not required since the picking will start at the position of the wavelet selected and expand out from there. After you have selected the wavelet using the Select/View Wavelet dialog options the parameters are show on the main window. The additional information needed is the Search window, which is how large of a window to use, when looking for the waveform shape. The match is based on similar wavelet character for the defined wavelet but you can search a larger window in the case of dipping events.

Select/View Wavelet:Click to select a wavelet or to view the selected wavelet. 

Wavelet Trace/Line: The trace used for the wavelet.

Pick Location: The vertical position of the pick in the trace.

Wavelet Above/Below:Vertical amount above and below the pick location which defines the wavelet.

Search Above/Below: Size of the search window when looking for the wavelet in another trace. This window can be larger than the picked wavelet. When the wavelet is being compared, it will using a sliding window to ensure that matching wavelet has the same size.

Minimum Confidence:Minimum correlation confidence needed to pick the wavelet in another trace.

Select Wavelet Dialog  

SelectViewWavelet.jpg

Location: The Line/Trace location within the 3D the wavelet is from.

Picking: The default picking parameters: similar to the Auto picking Parameters.

Wavelet: The actual time/depth value where the wavelet was picked, as offsets above and below where you picked.

Trace Correlation

When selected, a cross correlation is performed from trace to trace to ensure the seed event is propagated to the correct event. Useful for steeply dipping events, faulted data or marine data.

Correlation Top Offset: Offset added to the horizon defining the top of the zone.

Correlation Bottom Offset: Offset added to the horizon defining the lower part of zone.

Minimum Confidence: The correlation result must be higher than this value for the pick to be made.

Also See

MORE INFORMATION