fbpicker

NAME

SYNOPSIS

DESCRIPTION

fbpicker attempts to locate the onset of energy on input trace data by utilizing one of three algorithms. Option One is to calculate the ratio of standard deviations from a pair of running windows separated by a sample. The first break is take from the position of the ratio minimum over the zone analysed. Option Two is to decompose the data at a user defined frequency using running window spectral decomposition, derive a threshold from an user defined zone and return the first break where the spectal amplitude first exceeds the trace dependent threshold. Option Three is to calculate a running window AR coefficient series and threshold that series in a similar fashion to that used in Option Two.

fbpicker gets both its data and its parameters from command line arguments. These arguments specify the input, output, the start and end trace, start and end record, The output trace header mnemonic for the raw first break pick, the upper window size in the units of the data, the lower window size, the trace header mnemonic from which to obtain a trace variable start time, the start time at zero offset, the maximum time below the start time defined for the current trace for a pick to be made, the maximum time [or depth] to be considered, The hyperbolic velocity at which to vary the onset time for detection, the optional number of traces over which to perform spatial median smoothing of the first break picks, the trace header mnemonic into which to store the smoothed first break pick times, the frequency at which to perform the running window decomposition of the trace, the size of the running decomposition window in the units of the data, the threshold multiplier for the running window spectral decomposition detection, the auto regressive technique flag, the auto regressive technique coefficient threshold multiplier, the starting AR coefficient to consider, the last AR coefficient to consider, the AR predicition direction indicator, the AR residual calculation method flag and verbose printout, if desired.  

Command line arguments

-N ntap [default: stdin]

Enter the input data set name or file immediately after typing -N unless the input is from a pipe in which case the -N entry must be omitted. This input file should include the complete path name if the file resides in a different directory. Example -N/b/vsp/dummy tells the program to look for file 'dummy' in directory '/b/vsp'.

-O otap [default: stdout]

Enter the output data set name or file immediately after typing -O. This output file is not required when piping the output to another process. The output data set also requires the full path name (see above).

-ns nstr [default: 1]

Enter the starting sequential trace in each record to use in processing. Sequential traces less than this number will not be output.

-ne netr [default: last trace of record]

Enter the last sequential trace in each record to use in processing. Sequential traces greater than this number will not be output.

-rs nrst [default: 1]

Enter start sequential record number. Records prior to this entry will be skipped.

-re nred [default: last record]

Enter end sequential record number. Records after this entry will be skipped.

-hw c_RawWrd [default: Horz08]

Enter the trace header Mnemonic in which to store the raw first break pick. This value will be stored in the units of the data. If you detect a floating point header, such as the default, the value will be stored as a floating point number. If you choose an integer location the nearest integer to the floating point solution will be stored. You can convert this data to an xsd pick file for overlay on your data using the USP routine hdr2xsdpick. You can convert these values to USP trace time series entries using the USP routine usp2hdr.

-win1 iwin1 [default: 20]

Enter the upper window length in the units of the data. The function performed by this entry varies accoding to the algorithm being employed. When using the running window standard deviation ratio algorithm, this entry defines the size of the upper running window from which the divisor in the standard deviation ratio is to be obtained. This entry is not used with the running window spectral decomposition option. When using the running window auto-regressive coefficient algorithm, this entry defines this upper window from which the median coefficient is derived for use in the threshold calculation.

-win2 iwin2 [default: 20]

Enter the lower window length in the units of the data. Again the function performed by this entry also changes with algorithm used. When using the running window standard deviation ratio algorithm, this entry defines the size of the lower window from which the divident is of the ratio of standard deviations is calculated. Run time goes up linearly with increase in this parameter.

In both of the other algorithms, this entry defines the extent below the starttime to be examined on any given trace for the existence of the first break. If no appropriate pick is found in this zone then a pick at the base of the trace will be inserted.

-hws c_TzoWrd [default: not used]

Enter the trace header Mnemonic in which is stored the trace-wise start time for analysis. This entry replaces the functionality of the -t0 entry on the command line.

-t0 it0 [default: start of trace]

Enter the start time for the analysis at zero offset.

-twin iwin [default: from start time to end of trace]

Enter the length of the analysis window below the start time for a given trace. This is the window of data that will be examined for the possible location of the first break. If the first break is not located in this window then the max time for the trace will be entered as the first break.

-tmax imax [default: end of trace]

Enter the maximum time to be analysed on the dataset. Trace time series entries greater than this time [depth] will not be considered in any of the calculations.

-vel rvel [default: infinity]

Enter the hyperbolic velocity desired to adjust the analysis start time as a function of offset.

-sm ism [default: no smoothing done]

Enter the number of traces to use in the spatial median smoothing of the raw first break picks. The picks are smoothed within record boundaries only.

-hwsm c_SmoWrd [default: Horz07]

Enter the trace header Mnemonic in which to store the smoothed first break pick. This value will be stored in the units of the data. If you detect a floating point header, such as the default, the value will be stored as a floating point number. If you choose an integer location the nearest integer to the floating point solution will be stored. You can convert this data to an xsd pick file for overlay on your data using the USP routine hdr2xsdpick. You can convert these values to USP trace time series entries using the USP routine usp2hdr.

-freq ifreq [default: not used]

Enter the decomposition frequency in Hertz to used when doing a running window spectral decomposition of the input trace. If an entry is present here then the frequency decomposition algorithm is triggered unless -AR is also present on the command line. The heirarchy goes -AR -freq[]. If neither are present, the routine defaults to the dual running window standard deviation ratio technique.

-fthresh rfthresh [default: 30.0]

Enter the threshold mutliplier to use in concert with the running window spectral decomposition technique. The threshold is calculated as the median spectral coefficient between the start of the trace and the time [depth] designated by the -t0[] entry on the command line. This median is then multiplied by -fthresh[] to form the detection coefficient threshold used to attempt to localize the first breaks. usually a smaller value results in earlier picks and a larger value in the picks being pushed deeping in time or depth. When you get in the ball park you will probably notice that some picks are early and some are late. Hopefully a smoothing of the picks at this point will provide a reasonable result.

-AR

Enter the command line argument '-AR' to get invoke the auto regressive coefficient search algorithm.

-arthresh rarthresh [default: 3.0]

Enter the threshold mutliplier to use in concert with the running window auto regressive coefficient technique. The threshold is calculated as the median AR coefficient between the start of the trace and the time [depth] designated by the -t0[] entry on the command line. This median is then multiplied by -arthresh[] to form the detection coefficient threshold used to attempt to localize the first breaks. usually a smaller value results in earlier picks and a larger value in the picks being pushed deeping in time or depth. When you get in the ball park you will probably notice that some picks are early and some are late. Hopefully a smoothing of the picks at this point will provide a reasonable result.

-arwin iarwin [default: 50]

Enter the size of the AR window used in calculation of the AR coefficients. This entry is in the units of the data.

-ms ims [default: 2]

Enter the starting AR coefficient to examine.

-ml iml [default: nsamp / 4]

Enter the last AR coefficient to examine.

-icode icode [default: 0]

Enter the direction of prediction for the AR coefficient determination. For a forward prediction enter 1. For a backward prediction, enter -1. To do both, enter 0.

-rscode irscode [default: 1]

Enter the type of calculation used to determine the AR residual sum of squares. An entry of 0 uses a fast algorithm. An entry of 1 results in a slower, but more accurate determination to be made.

-V

Enter the command line argument '-V' to get additional printout.

-? or -h or -help

Enter the command line argument '-?' or -h or -help to get online help. The program terminates after the help screen is printed.

DISCUSSION

fbpicker -N[] -O[] -win1 60 -win2 60 -sm3

Option Two: Threshold on mono-frequency running window spectral decomposition. In this option the trace is decomposed using the running window algorithm for the frequency -freq[] using a decomposition window of size -fwin[]. A median spectral coefficient is computed from the zone between the first sample and the sample designated by the -t0[] entry on the command line. That median coefficient is then multiplied by the -fthresh[] multiplier. The result is used as the threshold of detection for that trace. The decomposed trace is examined from the start sample determined by -t0[] -vel[] and the trace offset down to either the end of the trace, or a maximum time -tmax[] or for some distance below the start sample determined by the -win2[] entry. Whichever of these occurs first will end the threshold test. If no threshold has been detected by that point the maximum time of the trace will be used as the first break pick.

fbpicker -N[] -O[] -t0 1100 -vel 8000 -win2 3000 -freq 35 -fthresh 30 -sm 11

Option Three: Threshold on running window AR coefficient determination. In this option a running window of length -arwin[] is used to compute the AR coefficient for a prediction direciton dictated by -icode[]. This is initially carried out for the zone between sample 1 and the sample corresponding to -t0[]. The median AR coefficient from this zone is computed and multiplied by the -arthresh[] command line entry to form the AR threshold coefficient value. The AR coefficient determination process is continued from the start time of the trace, as determined by -t0[] -vel[] and the trace distance, to either the end of the trace, the max time -tmax[] or the user defined time to hang below the computed start time -win2[].

fbpicker -N[] -O[] -AR -arwin 60 -arthresh 3.3 -t0 1100 -vel 8000 -win2 3000 -sm 11  

BUGS

SEE ALSO

AUTHOR

     copyright 2004, Amoco Production Company 
              All rights reserved
        an affiliate of BP America Inc.

Index

NAME

SYNOPSIS

DESCRIPTION

Command line arguments

DISCUSSION

BUGS

SEE ALSO

AUTHOR