Start with a 3D-model of a fibrous material that has been loaded in memory and is displayed in the Visualization area. In the FiberFind section, select Estimate Fiber Orientations from the pull-down menu and click the Options’ Edit... button.

In the opening Fiber Diameter Estimation dialog, enter the desired Result File Name (.gdr), and select the Material to analyze as explained for Estimate fiber diameters.

For homogeneous materials, without variations in orientation, the domain can be analyzed as a whole and, therefore, the Number of orientation tensors in X direction, in Y direction, and in Z direction can be set to 1, 1, and 1.

For non-homogenous fibrous materials, where fiber layers with varying orientations are present, the domain must be subdivided into blocks and an orientation tensor can be estimated for each block. The partition allows to analyze fiber orientation in these diverse areas of the domain, instead of having to cut the structure in parts and to analyze them separately.

The size of the blocks is defined by setting the Number of orientation tensors, for which values can be set, e.g. to 3, 3, and 2, producing 18 blocks.

These 18 blocks are labeled later in the result file as Block 0,0,0, Block 1,0,0, Block 2,0,0, Block 0,1,0, Block 1,1,0, etc.

Next, choose a method for the analysis of fiber orientation: Star Length Distribution (SLD) or Principal Component Analysis (PCA, Automatic or Manual window size).

The Star Length Distribution (SLD) method works on a per-voxel basis. For each voxel, it analyzes chord lengths through the voxel for a pre-defined set of directions.

The relative length of the chords gives the per-voxel orientation tensor. Then, the tensors are averaged over all voxels included in a block.

The PCA method is an alternative approach that subdivides the domain into windows of a given size. The size of the window can be automatically determined based on maximum fiber diameter (Automatic window size) or manually entered in voxels (Manual window size).

The subdivision into windows is applied to eliminate the influence of fiber crossings. Our PCA-based algorithm identifies fiber fragments and analyzes the direction tensor within each window. Then, it averages the direction tensors over all windows within a block to obtain the orientation tensor for that block.

When needed, check Write GOF file to save a GeoDict Orientation File (GOF) with the per-voxel fiber orientation information.