29 June 2020
We are pleased to present FMGT 7.9.4
Cascading Backscatter Normalization Tool
We have built a new tool that lets users normalize the backscatter across an entire survey to compensate for occasions when different survey lines have different backscatter levels. The tool is semi-automatic in that the user must select one or more reference lines and then the dB offsets are calculated for all other survey lines relative to the selected reference lines. Once calculated, the user can adjust, tweak or disable offsets for individual lines. This is assisted by a visualization preview of the final mosaic. A snapshot of the tool in action is shown below, with a before and after image showing the result.
Once satisfied with the results, the dB offsets are saved along side each line in the project folder and they are used by default in the initial mosaic that existed prior to launching the tool. They are also used in any subsequent creation of new mosaics. The user can adjust the offsets individually after this by using the existing "Line Brightness Tool", which has been renamed to "Line Backscatter Adjustment" tool. The renaming was done to clarify the use of the tool, especially now with the arrival of the Cascading Backscatter Normalization Tool. The dB offset that is calculated with the Line Backscatter Adjustment tool is now saved so that the offset is re-applied with new mosaics. Previously, it was only applied to the existing mosaic and would have had to be re-calculated anytime a new mosaic was created. The image below shows the new option to save the offset, it is enabled by default.
There is also now a new setting in the Mosaic menu that allows the user to choose to use the line dB offset results in mosaicing or not. By default, it is off, but it is enabled once the user uses either the new Cascading Backscatter Normalization Tool or the Line Backscatter Adjustment tool.
There are a number of other miscellaneous improvements in this release also:
- The "Backscatter Level Adjustment" menu item has been renamed "Intra-Vessel Backscatter Adjustment" and the title of the tool's dialog has been changed to match. This was done to clarify the intended use of this tool and to make a clear separation on the use case for this tool compared to the Cascading Backscatter Normalization Tool and the Line Backscatter Adjustment tool.
- Processing improvements - When adjusting the mosaic cell size, FMGT will only need to re-run the AVG filter. Previously it was re-processing the backscatter, which made re-mosaic operations take longer after changing the cell size.
- The right-click context menu for mosaics has had the order of the top two menu items switched to have 'Rename' listed first instead of 'Remove'. This was done to help avoid accidental removal.
- Added a safety check when pairing .R2S files with GSF to prevent accidental pairing of dual-head data. Dual head data is not supported for .R2S files from Hypack, but the pairing procedure would attempt to pair anyway, resulting in incorrect results.
- Multi-spectral processing files will now report the frequencies available for processing in the raw files tree. This is also reported in the Datagram Viewer.
- Processing / Source Files
- Fixed a bug with multispectral pseudo-color image creation where color bands were sometimes misaligned
- Improved the angle calculation for the Statistics layers for GSF file users in very shallow water with large CoG to sonar head offsets across track. It was previously not accounting for the across-track offset for the Statistics calculations. Normal backscatter mosaics were unaffected by this bug since this type of correction is already made for backscatter mosaic and ARA workflows.
- Fixed a problem where the Coverage processing flag was set even if the Coverage processing did not complete. This affected GSF file users in which the GSF file did not have a sensor ID set, resulting in a need to do a sensor over-ride in the processing settings.
- KMALL processing has a correction to use the full number of beams in a ping, not just the number of valid beams. The beam processing loop was only examinining the number of valid beams, which left one edge of the coverage to be smaller than expected.
- KMALL backscatter processing has had a fix to correct for the sonar head offset, relative to the CoG, so that beam angle calculations are improved. This would have affected users with data in shallow water with large offsets between the sensor and the CoG.
- Importing / Exporting
- Improved handling of HDCS/XTF pairing when encountering an incorrect number of beams in the HIPS process.log file and also situations where beams with zero (0.0) depths were influencing the cell size calculation.
- Fixed a problem with Qinsy db/qpd pairing when ADCP records were present. FMGT's .db reading code was not aware of the newer ADCP packet type, this was causing errors in indexing the file and thus access to the raw multibeam backscatter packets.
- Fixed color map labeling for reference grids, they were reporting depth range of 0 to 1. It now maps to the correct depth range in the reference grid.