Source Code – Executable

This application was based on the paper Lowest-Variance Streamlines for Filtering of 3D Ultrasound by Šoltészová et al., 2012. A novel 3D volume filtering approach targeted at ultrasound data was presented. The filtering focuses on removing noise without changing the structure of the relevant information. To achieve that, two steps have to be taken: the generation of streamlines in direction of the lowest variance and the integration along the streamlines concluding in filtering.

Usage

Load data

File > Open...
A sample dataset is provided in /data.

Filtering Modes

Off: No filtering, dataset as it was loaded
Streamline: Lowest Variance Streamline Filtering algorithm
Avg Gauss: Simple Average Gauss filtering

Side by Side view

By activating the side by side view per default no filtering and the last Streamline filtering are rendered side by side. By changing the parameters on the right, the left image can be adapted.

Variables

In the Variables section the three variables of the Streamline Filtering algorithm can be adapted:

delta: The sampling value of the directions the lowest variances are computed of in degrees
n: The number of steps taken for calculating the variance in positive and negative direction
m: The number of steps for integrating along the positive and negative vectorfield

Camera

The camera can be controlled by mouse input:

Left click + drag: Rotate camera
Right click + drag: Strafe camera
Vertical scroll: Zoom in/out
Horizontal scroll: Roll camera