Loading processed FMS images from logDB into TechLog


    The data in the LogDB processed DLIS files have all been corrected for tool motion and depth matched to other data, but the various arrays need to be recombined in order to be able to display and interpret the images. The procedure depends on the software used for the original processing (GeoFrame until 2017; TechLog after):
    For data processed with GeoFrame (last IODP expedition: Expedition 368), the required variables need to be concatenated into image arrays, which are then used to create Pad images, that are finally concatenated into oriented images.
    For data processed with TechLog (starting with IODP Expedition 371), the Pad images are already part of the processed dlis file, and they only need to be oriented.
    For all data, the recommended final step is the equalization, or normalization, of the images, which typically enhances the images and makes the interpretation easier.
    The first version of this document was started with an earlier version of Techlog, and it was later upgraded with newer versions. Some appearances may be different from the current versions, but the procedure is the same.

Importing processed DLIS images
When importing images into TechLog (regardless whther they were originally processed with GeoFrame or Techlog), it is necessary include the execution of a script in the importing process that will group the variables necessary to generate the images into image Datasets. After starting Home/Import, select 'Datasets' in the right side of your import window, bring up the properties editor, select 'Launch the script' under 'Launch a script after loading the file...', and select the 'wbilmport'.


Only then should you import your file, which should create a dataset called FMS_n_-xxx-xxxm - or similar, where n is 4 or 8, depending on how Techlog recognizes the tool, and xxx-xxx is the top and bottom of the interval logged, with 'ft' or 'm' at the end, depending on the field unit. This is the dataset that should be used in the following steps.

Concatenation of FMS variables
The FMS arrays are stored as individual buttons, that need to be regrouped into four pads.
Go into ‘Geology’/AWI explorer and select ‘concatenate the FMS variables into arrays’



This opens a small menu where you only need to select the hole that you are working with.

Then select the image dataset and drop it where it says ‘Please drop your data here’. The arrays are called BA11,…BA28,BB11,..BB28,..BD28. This should create 8 arrays called FCA1,FCA2,…FCD4 into your image data set, which correspond to the two rows of buttons on each pad.

Before going further, you need to check if the orientation and caliper arrays (RB, HAZI, P1AZ, DEVI, C1, C2) are already in the same dataset as the button (BA11,…) and the newly created FC** arrays. If they already exist you can skip this and proceed straight to the Pad Image creation. If not, you need to copy these arrays from the other datasets from the same pass.

Find these arrays (they may be in a DataSet with lower sampling rate), copy them, one by one, or simultaneously with the control key:



and then paste them into the FMS dataset:

A "resampling" window will appear (because the arrays have different sampling rate than the FMS arrays) - select ‘Apply’ to proceed


Pad Image Creation
The next step is to group all the image arrays into 4 pads.
Go under Geology/Processing - and select Pad image creation


Here you have to select change tool and change to FMS_8 (The ODP/IODP tool is actually FMS_4, but the option does not exist at this stage) and select the arrays to use.
To make the data assignment  easier, select only your hole (if there are other holes in the project), display only the variables (blue icons), then select the correct variable, and click in the box near the list of data to actually bring the data in any given list. Then hit ‘create’ to close this window and initiate the workflow in the main techlog window.



Make sure that the central menu says ‘save and display’, not just ‘display’, and hit the blue arrow to run the script

This creates the ‘PAD_*’ arrays, that still need to be combined with orientation data to be displayed properly.
This is similar to the FMS processing results that you can find in the LogDB processed dlis files that were processed with Techlog, and still need one last step: the pad concatenautaion and orientation.

Pad Concatenation and orientation

Finally the pad image arrays need to be combined with orientation data to be displayed properly and interpreted. In the Geology/Processing menus, select ‘Pad concatenation and orientation’:

When this menu opens, you have to change the tool - select FMS_4, again display only the variables from your hole, make sure that the arrays displayed in the data type assignment match your arrays, or change them if necessary, and hit ‘create’

This opens the pad concatenation workflow in the main window. If you have been through the array creation stages before, your DataSet is already there and you just need to hit the blue ‘run’ button.
If you are just loading dlis data that were processed in TechLog, select your dataset with FMS data, drop it in the window, then hit run:


One more time, make sure that ‘save and display’ is selected before hitting the little blue arrow. This should complete the processing and display an FMS image in your logview window. The new array will have a name like ARRAY_WBI, which can be renamed more appropriately, and that you can already display and interpret in LogView.

Image Normalization
One last step can be added to ensure that the images are similar to the ones in LogDB: the ‘Histogram equalization’, or normalization, which enhances the images and helps with the interpretation. Select Geology/Processing/Histogram equalization (normalisation). Once opened, you can choose either static or dynamic normalization: static equalization is over the entire hole, for a broader interpretation, while dynamic normalization is using a moving window to enhance local contrasts.


One Final Note
Images displayed after this stage should look similar to the LogDB gif images. One significant possible source of difference is that in the creation of the gif images, a constant hole size is assumed, to create images that cover about 25% of the hole. This procedure uses the actual caliper, generating wider pads in narrow intervals and thinner pads in enlarged intervals.