Results
The result file (*.gdr) is opened in the Result Viewer after the computation is finished.
The Results tab is the central point for the analysis of the identified grains. It is grouped in three subtabs: Report, Plots and Map. The Report tab shows statistics about the estimated grain diameters and the Plots tab contains a histogram of the estimated grain diameters. The Map tab contains all resulting data from the Estimate Grain Diameters run. This data is the basis for the tables in the Report tab and for the plot in the Plots tab. On the left, the Post-Processing Widget allows you to classify the grains into the specified Number of Grain Types by fitting a Gauss function to the estimated diameters. Collapse the Post-Processing Widget by pulling it to the left.
Report
General analysis of the size distribution (*.gsd)
Here, the computed Average inner diameter of all grains and the Standard deviation are given
|
Grain type analysis
The statistical parameters (Average diameter, Standard deviation, and Median diameter) of the fitted Gauss function for each class are displayed in this table. Additionally, the volume fraction for each class is given.
If more than one class is reported, the used Thresholds to determine the types are given below the table.
|
Note! These values depend on the entered Bin Size!
|
|
Characteristic diameters
The characteristic diameters D10, D50, D90 are the diameters such that 10%, 50%, or 90% of all grains have a smaller estimated diameter than the corresponding value.
|
Histogram
The histogram table contains the values from the histogram plot, where each row contains the data of one bin. The diameter of grains in one bin is between the minimum and maximum diameter. The volume fraction shows how many grains are contained in a bin. Of course the data in the table changes, when entering another Bin Size.
|
Plots
The Plots tab shows the histogram of the estimated grain diameters. Additionally, the fitted Gauss functions for the different classes are plotted.
Post-processing widget
In the post-processing panel on the left, you can choose the Bin Size to define the granularity of the histogram. Use the Thresholding Methods and Number of Grain Types to classify the grains into different classes based on their diameter. Clicking Apply immediately changes the results in the Map subtab, and therefore the tables in the Report subtab and the histogram in the Plots subtab.
Bin Size
Changing the Bin Size effects the resolution of the histogram in the Plots tab and the histogram table in the Report tab. A too large Bin Size leads to a “blocky” histogram, while a too small Bin Size leads to a volatile and noisy histogram. The Bin Size should be chosen so that the histogram is smooth and detailed. The effect of the Bin Size on the histogram is illustrated below.
|
Thresholding Method and Number of Grain Types
You can classify the grains in the structure into different types based on the estimated diameter.
To determine the threshold value, three Thresholding Methods are available:
- k-Means: uses the k-Means algorithm to find thresholds for the chosen Number of Grain Types.
- Otsu: uses the Otsu algorithm to find thresholds for the chosen Number of Grain Types.
- Manual: define your own thresholds by writing a comma-separated list of thresholds into Threshold(s). You implicitly define the number of grain types by the number of thresholds you enter, e.g, if you enter two threshold values this will result in three grain types.
After clicking Apply, the Grain Type Analysis table is updated and shows for each type the relevant parameters. In the Plot tab, the Gauss fit of each grain type and the threshold values are added to the histogram plot. This allows you to control if the computed thresholds fit to the grain diameter distribution or need to be adapted.
Based on these diameter classes, the structure is automatically segmented and can be loaded from the Data Visualization tab
|
