Single target detection split beam
Method 1
This operator detects single targets using a modified version of the single target detection single beam (method 1) algorithm that applies compensation estimates (based on split-beam angle data) to the peak selection criteria.
Echoview accepts operands of the following data types as input:
|
Operand 1 |
Operand 2 |
|
|
Note: Prior to Echoview 4.30, this operator also accepted the Sv data type as input. Sv data requires a Sv to TS conversion and appropriate calibration settings. To avoid confusion and incorrect data, this operator has been simplified. EV files with old versions of the operator will appear with [deprecated] in the operator name and behave as they previously did.
For algorithm information see Split beam (Method 1).
Method 2
This operator detects single targets from split beam data using a modified version of the single target detection single beam (method 2) operator that applies compensation estimates (based on split-beam angle data) to the peak selection criteria.
Echoview accepts operands of the following data types as input:
|
Operand 1 |
Operand 2 |
|
|
For algorithm information see Split beam (Method 2).
Settings
The Single Target Detection Split Beam (Method 1 or Method 2) Variable Properties dialog box pages include (common) Variable Properties pages and these operator pages:
Operands page
Filter Targets page
Single Target Detection page
General parameters
|
Setting |
Description |
|
TS threshold (dB re 1m2) (uncompensated TS) |
Method 1 operators apply an uncompensated TS threshold before detection. Uncompensated TS threshold represents the TS of the smallest single target of interest to you. Determine what TS you are interested in (the TS for fish rather than the TS for plankton) and set TS threshold to a value that is a bit smaller. The minimum value for this setting is -120 dB re 1m2. This minimum value will detect spike noise, especially if you also use a short minimum pulse length. |
|
TS threshold (dB re 1m2) (compensated TS) |
Method 2 operators apply a compensated TS threshold after detection. Compensated TS threshold represents the TS of the smallest single target of interest to you. Determine what TS you are interested in (the TS for fish rather than the TS for plankton) and set TS threshold to a value that is a bit smaller. The minimum value for this setting is -120 dB re 1m2. This minimum value will detect spike noise, especially if you also use a short minimum pulse length. |
|
Pulse length determination level (dB re 1m2 for method 1) Pulse length determination level (dB re 1W for method 2) |
This is theoretically and in-practice 6 dB (and rarely changed). Method 2 operators remove TVG before searching for peaks. PLDL is calculated relative to the pulse's TS peak. Samples above the PLDL are included in the detection. Normalized pulse length and Standard deviation of angles are based on included samples. |
|
Minimum normalized pulse length |
Minimum normalized pulse length together with Maximum normalized pulse length are used to reject pulses that fail the estimate for expected echo length. Too short a pulse length allows too many false detections. Too long a pulse length allows the loss of too many real detections. (Pers. comm. I. Higginbottom). The minimum value is 0.01. Simrad EK60Samples are separated by 1/4 pulse length. Apparent Echoview normalized lengths will be 0.25, 0.5, 0.75, 1, 1.25, 1.5. These values are due to Simrad's sampling and the fact that Echoview does not interpolate.
Simrad EK80 CW dataFor Simrad EK80 CW data, the Method 2 operator uses the EffectivePulseLength with the Minimum and Maximum normalized pulse length settings. BioSonicsBioSonics data has many more samples in a pulse. The estimate for echo length is not as constrained as for Simrad data. |
|
Maximum normalized pulse length |
Minimum and Maximum normalized pulse length are used to reject pulses that fail the estimate for expected echo length. See Minimum normalized pulse length. Too short a pulse length allows the loss of too many real detections. Too long a pulse length allows too many false detections. (Pers. comm. I. Higginbottom). The maximum value is 10. |
Beam compensation
|
Setting |
Description |
|
Beam compensation model |
Beam patterns can be modeled and can be used to estimate what the TS value would be if the target was central in the beam. Echoview offers echosounder-specific beam compensation models. Many models cite the limits of validity.
|
|
Beam patterns can be modeled and can be used to estimate what the TS value would be if the target was central in the beam. Higher values of beam compensation result in more single targets. Many beam compensation models cite limits of validity. "... In shallow lakes (western Lake Erie) or in lakes with few targets (e.g. Lakes Superior and Huron), we recommend using a larger beam compensation to increase the number of detected single-fish echoes. Our analysis shows less than a 1 dB difference when beam compensation is increased from 3 to 12 dB. This potential bias has to be weighed against the increase in the precision of in situ TS measures due to more identified "In effect the higher the value the further from the beam center targets will be accepted. The default value of 4 dB is optimized for accurate TS estimation, because there is more uncertainty in the compensation the further off axis. For just detecting targets, especially strong targets that might be in the edge of the beam, it is fine to use a larger settings ... More noise and false single targets are detected as Maximum beam compensation is increased." (Pers. comm. I. Higginbottom). The maximum value is 35 dB re 1m2 (split beam and dual beam only). Note: The value for Maximum beam compensation is important when using the export variables Beam_volume_sum and Wedge_volume_sampled. For further information refer to Beam angle parameters. |
Exclusion
|
Setting |
Description |
|
Maximum standard deviation (degrees) |
Applies to samples in the single target peak as specified by the Minimum and Maximum normalized pulse lengths. The standard deviation of minor and major axis angles relates back to angular position and where the target is in the beam. When there is more than one target then phase is noisy and TS is not accurate. Use Maximum standard deviation settings to limit noise with respect to phase or in cases of low signal-to-noise. As Maximum standard deviation is increased more false detections may occur. "... Phase-jitter (angle standard deviation) should be smaller than a certain value... " (Parker-Stetter et al: Section 9.2 Single-echo detection). The maximum value is 45 degrees (split beam only). The maximum value for this setting effectively directs means the detection ignores the angle deviation. (Pers. comm. I. Higginbottom). |
Line-Exclusion
The Line Exclusion section is used to select lines to limit targets and minimize processing.
|
Setting |
Description |
|
Exclude targets above line |
These are analogous to the Exclude-above and Exclude-below lines used on the Analysis page. The exclusions on the Single target detection page are applied before the echogram is displayed (affecting the number of targets displayed) and the exclusions on the Analysis page are applied after the echogram is displayed (affecting the on-screen and exported analyses). See About setting variable properties for an illustration of which pages apply when. |
|
Exclude targets below line |
The correct configuration of calibration settings (environment and echosounder system) and transducer geometry for the operands is important. You may need to configure raw variables prior to using the operator. For more information refer to Creating single target detection variables from split beam data.
If you are familiar with Single target detection (or TS-Detection) on Simrad echosounders, see Simrad and Echoview single target detection terminology for a comparison of the terms used.
See also
About virtual variables
Operator licensing in Echoview
Tuning single target detection