User Guide
(WRITING IN PROGRESS)
- Kongsberg .all
- Konsgberg .kmall
To use Iskaffe, you need to first convert the raw data files, and then load those converted files in the software.
Converted files are saved on your drive, in the same location as your raw data, in a Coffee_files folder. There is one subfolder per converted file. You are only limited by your drive capacity here. This may take a lot of space eventually if you convert a lot of data, but you can simply delete the Coffee_files folder whenever you need. Or you can delete the individual subfolders for each file you don't want to keep.
Loading converted files in the software takes up space in your system's live memory, so you will be limited by your machine’s RAM capacity about how many files you can load at a time.
Conversion takes much longer than loading, so you may find it useful to just convert everything first, when you are not using the software, and then later load whichever files you want to look at.
Last note: as new features are developped for CoFFee and Iskaffe, the format of the converted data (fData) may be changed. If there is no more backward compatibility between a new version of Iskaffe and old converted data, Iskaffe will no longer recognize converted files as "converted" and you will need to re-convert them.
You convert raw data files and load converted files from the Raw files tab.
In the raw files tab:
- Files in
$\textcolor{gray}{\textsf{normal gray}}$ are not converted yet - Files in
$\textcolor{black}{\textsf{bold black}}$ are converted but not loaded - Files in
$\textcolor{green}{\textsf{bold green}}$ are converted and loaded
You can use the “Convert” button to convert files that are not converted yet, or re-convert files that are already converted.
You can use the “Load” button to load converted files. If you try to load unconverted files, Iskaffe will let you know that they need to be converted before being loaded.
To select which files to convert/load, simply check the corresponding boxes. You can right-click to "select all", "deselect all", "inverse selection" or "selected converted files only".
If there are multiple modes among all loaded lines, you will receive a warning at this stage.
Loaded lines are also listed in the Loaded lines tab. This separation of the Raw files tab and Loaded lines tab allows you to load lines from multiple folders at ease.
Here you can unload lines by right-clicking on lines, and choose "Unload selected lines" from the context menu.
From the context menu, you can also reorder the lines (for now only chronologically).
Set the grid size and click on "Grid"
- (coverage)
- Bathymetry
- Slope
- Modes
When displaying the modes, you can click on the colour bar on the right side to get a popup window for a description of each mode. A bit better to read than in the “analysis results” section.
After clicking “Analyse”:
- The yellow stars represent soundings flagged as invalid in the system.
- Not showing on your data are “bad pings” when you have them. They show as red lines.
I am still developing this “bad pings” algo to try to be more consistent across all sonar models, but for now a ping is flagged as “bad” if:
- It has more than 10% of bad soundings, OR
- It has more than 50% of the soundings experience a drop of more than 3dB compared to the median level in the 10 previous pings (after a gate of 5 pings).
- This part of the algo may be tripped if the seafloor type changes perpendicular to the direction of surveying.
Since it is under development, no need to pay too much attention to this for now, but it will give you an estimate of the rate of weather dropouts. In my experience so far with the data I have tried, when you reach about 1% of bad pings, it usually means the conditions are quite bad for surveying.
All the “analyse survey” does for now is finding cells that are not covered with data within the “survey limits”. Eventually, I will get it to figure which cells have bad data too. Because the bad pings in the “line analysis” completely ignores if you have acquired good data over bad data.
When you hover the cursor over Map view, the location of the cursor (in the local UTM projection) are shown in the bottom part of the Display section of the Process tab. This information will change with mouse movement.
If you click on the navigation line, the location of the cursor at the time of the click, the line that was clicked on, and the depth at this location are shown in the upper part of the Display section of the Process tab. This information will stay on the screen until you click somewhere else on the map, allowing you to copy this information.
In the top-right corner of the Map view, there is a toolbar with icons for common interactions: Pan, Zoom-in, Zoom-out, and Restore view (zoom-all-the-way-out).
These two tools allow running the “parameter change analysis” and the “line analysis” (aka % bad soundings and % bad pings) on a folder of raw data without loading them in Iskaffe, aka without overloading the memory.
So you should be able to run these tools on an unlimited number of files (provided there are no issues in the data reading and application of the algorithm).
Both tools work the same: specify the folder of raw data, specify the file to contain the results, and let run.