Designed especially for neurobiologists,
OverviewFluoRender is an interactive rendering tool for confocal microscopy data visualization. It combines the rendering of multi-channel volume data and polygon mesh data, where the properties of each dataset can be adjusted independently and quickly. The tool is designed especially for neurobiologists, allowing them to better visualize confocal data from fluorescently-stained brains, but it is also useful for other biological samples.
10/09/2014. Ver 2.14 released !!
- Video output
- Feature tracking
- 3D measurement tools
- GPU-based volume rendering ensures fast response for real-time user interaction.
- Multiple render modes for multi-channel confocal data.
- Rendering properties of individual channels can be adjusted independently.
- Intuitive and easy-to-use parameter adjustment enhances workflow efficiency.
- Easy generation of animations and movie exporting.
- Trimming of individual channels with clipping planes.
- Data channels can be grouped. Gamma and brightness can be set independently for each group in layered and composite modes.
- 4D time sequence data support.
- 3D batch rendering for RGB TIFF files.
- Scalar value dependent colormap.
- RGB and 16-bit grayscale multi-layer TIFF file format support.
- 64bit Windows systems are supported.
- Maximum Intensity Projection (MIP) supported.
- Clipping planes can rotate independent of the object.
- FluoRender Twitter.
- Equalization for 3D and 4D confocal data.
- Shadow effects.
- Importing volume data from Olympus and Zeiss formats.
- Volume paint selection and segmentation.
- Better support of Zeiss format, including larger-than-4GB data support.
- Volume baking.
Release Notes: 10/09/2014 FluoRender 2.14
This version has expanded to MacOSX, includes improved user interface, issue fixes, and other program improvements.
1. Simplified and more user-friendly Recorder Dialog Box.
2. Icon-based options, Program File Menu, Optional Toolbar, and more intuitive controls.
3. More support for various TIF file formats.
4. Output of Quicktime MOV format available for recording.
5. Support for larger data sets. (Option available in Settings).
6. New 3D Measurement and Ruler Tools (for lengths, position, etc.)
7. Key Framing available for Advanced recording.
8. Continuous perspective angle adjustment.
9. Additional shortcut buttons for clipping planes/slabs.
10. New tracking tool to track structures in 4D time sequence data.
This version includes some improvements and issue fixes:
1. File loading speed for TIFF format is improved. For 4D playback, the speed increase can be 100% on certain systems. For the best speed performance, save a 4D sequence in separate, uncompressed channels.
2. When saving projects, there is an additional option for embedding datasets into the project directories. However, complete 4D sequence will not be saved using this feature.
3. Rendering contrast is increased for fine details with less opacity, when a bright background color is used.
4. It also includes other minor fixes with user interface, etc.
This version includes some improvements and issue fixes:
1. The colormap legend reflects transparency settings, which is more accurate to visualized results.
2. Fixed an issue in gradient magnitude calculations.
3. Added interpolation option in the render view settings. It turns on/off interpolations on the current selection.
4. Alpha in the volume property settings can be turned on/off.
5. Selection in the workspace panel is linked in the datasets panel.
6. A compression option is available in the open volume dialog. It reduces graphics memory usage and improves rendering speed.
7. Fixed some issues for Olympus format reading.
8. In the edit window, an edge detection option is added to the selection adjustment section. It enables the brush to use gradient magnitude for edge detection.
9. In the edit window, a select group option is added to the selection adjustment section. It allows painting on all channels in one group simultaneously.
10. Other minor user interface improvements and issue fixes.
1. In the edit dialog, removed 2D adjustment influence slider.
2. When growing region is turned off, selection is based on intensity thresholding.
3. Re-organized the property window for polygon mesh, with lighting being a feature that can be turned on/off.
4. Added a setting in the settings dialog. It allows ignoring voxel size from metadata.
1. When voxel size of XY plane is not set to 1, The 1:1 button won't zoom current dataset correctly. This is now fixed.
2. After loading polygon mesh files several times, they may all disappear. This is now fixed.
3. In certain situations, the workspace panel become empty after loading a project. This is now fixed.
4. The setting for saving LZW compressed TIFFs is not saved properly. This is now fixed.
1. Volume baking.
2. A link to FluoRender's facebook page.
1. Support of LZW compressed Olympus formats (OIB and OIF).
2. Save TIFF files (screen captures or volume data) with LZW compression.
3. Polygon mesh can be selected by mouse click in the render views.
4. Name of current selected data can be displayed in the render view.
5. Many UI improvements.
1. Low intensity voxels can be selected by setting a low falloff value of volume paint selection.
2. Larger paint stroke can have stronger grow strength.
3. Faster rendering of time sequence data through fixing a bug in using graphics memory.
This version has some issues fixed and improvements on usability.
1. Fixed a sampling rate issue when a volume is anisotropic and user sets XY spacings much greater (or smaller) than 1.
2. Users can choose a gradient background from the settings dialog.
3. Progress is shown when reading data or projects.
4. Improved user interactions when user chooses to use the workspace toolbar for volume painting functions.
5. Volume data loaded in the datasets panel can be saved as a sequence of 2D TIFF files.
1. Interactive volume segmentation.
2. Calculations between two volumes (Addition, Subtraction, Division, and And).
3. Derived volumes from the above operations can be saved as multi-layer Tiffs from the Datasets panel.
1. A 2D Tiff sequence can be opened as a volume. The new option is in the open volume dialog.
2. Users can set the 4D time sequence identifier in the open volume dialog.
3. Better LSM format support, including support of larger-than-4GB 4D data.
4. Excitation wavelengths can be read from raw format meta data and used for default color settings.
5. Improved polygon mesh rendering speed.
6. Better OBJ format support.
7. Automatic search for broken-link files when project and data are copied to a different directory.
8. Improved shadow effects.
9. Shadow effects in Depth render mode.
10. Instant render view update for most numeric inputs.
1. Support of importing several manufacturer-specific confocal formats, including Olympus Image Binary (*.oib), Olympus Original Imaging Format (*.oif), and Zeiss Laser Scanning Microscope (*.lsm).
2. Variable sample rate for mouse interactions.
3. Equalization for equalizing uneven brightness of 3D and 4D data. It also enhances details at the same time.
4. Shadow effects and directional shadows.
1. Improved user interface. Panels can be hidden or shown with a toolbar button.
2. Installation packages for FluoRender setup.
3. FluoRender project files (*.vrp) are associated with the executables if FluoRender is installed with the installation package. Double-clicking the project files will automatically launch FluoRender.
4. Project files as well as volume datasets can be drag-and-drop to either the FluoRender icon or the main interface to open.
1. Rendering core rewritten, for better rendering quality and performance.
2. True 16 bit TIFF support. Parameter settings reflect the actual scalar values.
3. Real-time smoothing filters for reducing the artificial patterns.
4. Data within groups can be set to depth mode, even the viewport is set to layer or composite mode.
5. Micro Blending. It can render accurate colors of multiple channels in depth mode, with an arbitrary number of channels.
6. User can set the zoom to reflect 1:1 voxel to pixel ratio.
7. One dataset can be added multiple times for renderings, and each instance can have independent parameter settings.
1. Several user interface and parameter setting improvements, for better usability.
2. In MIP mode, and colormapped MIP mode, some parameters are disabled, in order to keep the color accuracy.
3. Rendering quality and speed for mesh and mesh with volume improved.
1. The four modes for gradient magnitude calculation settings are removed.
2. User can set the mesh rendering quality by adjusting the number of layers to render.
1. A 64bit version can run on 64bit Windows systems. It supports loading larger
2. Maximum Intensity Projection (MIP) mode can be selected for any volume
channels from the volume properties. Shading effects, intensity-dependent
colormap, and most volume properties are also supported in MIP mode.
3. The clipping planes can be rotated independent of dataset rotations. The
user can set the clipping plane rotations either by aligning to view, or by
adjusting explicit rotation angles.
4. The clipping planes are color-coded.
5. The confocal voxel size can be retrieved from TIFF metadata, if presented.
6. A new option for saving a project file during image/movie export.
7. The time range can be set for 4D sequence movie export.
8. FluoRender now has a Twitter page, which can be accessed from the main
1. Realtime multi-channel 4D rendering with full playback control.
2. 3D batch rendering for RGB multi-layer TIFF files. Switching between multi-channel datasets is easy and fast.
3. Scalar value dependent colormap. The colormap is a blue-green-red gradient for low to high scalar values.
4. RGB multi-layer TIFF format support. RGB multi-layer TIFF files are added to the render view automatically after opened.
1. Property settings of datasets within the same group can be synchronized through the Tools panel.
2. Camera position and rotation are saved before switching to Free Fly mode, and they are restored after switching back to either Perspetive or Orthogonal projections.
3. Tools panel has less blinking when volume data are selected.
1. The visibility status of the groups are saved in the project files.
2. Solved an OBJ mesh file reading problem.
3. Solved a clipping plane setting by numeric input problem.
1. 4D multi-channel rendering.
2. New rendering modes that calculate the gradient magnitude differently. Choose different modes in the settings dialog can balance between rendering speed, data loading speed, and memory usage.
3. Saved volume properties can be set by the new reset button in the Tools panel.
4. The number of reset buttons of the 2D adjustment panel is reduced to three.
1. When new data are added to the group with existing volumes, the voxel size of the newly added volume is set to the existing ones.
2. Less crash when open project file with existing volumes.
3. Transparency of the volumes is corrected when open project files with Depth mode.
4. Color icons of the volumes in the scene view are corrected to always follow their color settings.
5. Same dataset in different render modes look the same now.
1. Drag & drop support in the tree of the scene view. User can now drag & drop items in the scene view to re-order and re-group them. The order of the volume data will be shown in the layered mode.
2. Groups in the tree of the scene view. Volume datasets can be grouped by creating new groups and dragging into the groups. Volume data in the same group share the same 2D adjustment parameters.
1. Spin buttons added for rotation settings.
2. For movie capture, the value for reverse rotation is always considered as the absolute value, no matter what the user input is.
3. 2D adjustment channel linking is automatically calculated from the color settings. Non-zero RGB channels will be linked in the 2D adjustment settings, when user changes the color of the volume data.
4. It remembers the HSV values of the volume data before user sets luminance. So the hue and saturation of the volume data will not change when user changes the luminance value within the range [0, 255].
1. In previous versions, the scaling of the volume datasets will change after viewport resizing. The problem is now fixed.
2. In previous versions, the program will crash sometimes when user deletes volume datasets from the data view. The problem is now fixed.
3. In previous versions, the program crashes quite often when loading volume datasets with slice resolution equal or larger than 1024x1024. Now the problem is fixed.
1. Gamma and brightness adjustment for final results.
2. Save settings as default for render view.
3. Preview, stop, and reset angel for movie export.
4. XYZ rotation reset button, and spin buttons for finer control.
5. Double-click to trun channel display on/off in the scene view panel.
6. Color representations for channels in scene view.
7. In property panel, volume XYZ voxel ratio are set without hitting "Enter" key.
1. 32-bit transparency blending for better detailed structures.
2. Slicing distance recalculated for consistent transparency during rotation.
1. Scale bar for size reference. The length, unit and text of the scale bar can be set by the user. The length of scale bar is linked to viewport zoom. Though scale bar is available for perspective projection, it's recommended that use it in orthographic mode for accuracy.
2. Clipping planes. Clipping planes can be set for different channels independently, or they can be linked for all the channels.
1. Zoom of render view is calculated according to the relative size of the dataset bounding box to the viewport, rather than previously the projected voxel size to screen pixel size. Thus even datasets acquired with different objectives and scanning resolutions can generate comparable rendered results, with zoom being set to similar values.
FluoRender (ß) 1.1.0
Fixes to lighting problems when rotating the datasets.
FluoRender (ß) 1.0.9
1. Switching render mode when the viewport is empty causes crash.
2. New channels added to the viewport don't show up in "Depth Mode".
3. Viewport flickering when changing the render modes.
1. Rotation animation and movie export.
2. Viewport rendering supports perspective and orthographic projections.
3. Double-click/Enter to add opened datasets to the viewport.
FluoRender (ß) 1.0.8
1. Fixes to some memory leaks.
2. Rendering speed is improved in "Depth Mode".
3. Fixes to image distortion of viewport capturing on some systems.
4. Fixes to program crash when capture viewports.
1. When using large data set option, the paintbrush is not available.
2. Paintbrush and related tools (including tracking) are unavailable for MacOSX.
3. Equalization in the Output Adjustment panel is unavailable in MacOSX.
4. Some large data sets have been known to crash the program on smaller graphics cards.
5. Minor bugs exist in a few tools. Please report any issues that affect your FluoRender experience.
FluoRender is available for free and is open source under the MIT License
The MIT License Copyright (c) 2012 Scientific Computing and Imaging Institute, University of Utah. License for the specific language governing rights and limitations under Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions: The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software. THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
If you use FluoRender in work that leads to published research, we humbly ask that you add the following to the 'Acknowledgments' section of your paper:
"This work was made possible in part by software funded by the NIH: Fluorender: An Imaging Tool for Visualization and Analysis of Confocal Data as Applied to Zebrafish Research, R01-GM098151-01."
- OS: Windows XP/Window Vista/Windows 7 (64 bit), or OSX 10.6
- CPU: Intel Pentium 4 or equivalent processor
- Graphics card: NVIDIA Geforce 9000 series
- System memory: 1GB
- Display resolution: 1280 x 1024
- OS: Windows 7 64 bit or OSX 10.9
- CPU: Intel Core i7
- Graphics card: AMD Radeon R9 290X or NVIDIA GeForce Titan Black
- System memory: 8GB
- Main Disc: Solid state drive, RAID0 or non-RAID
- Display resolution: 1920 x 1200
For volume data of more than 3 channels, strong graphics cards such as AMD Radeon HD6970/7970 or NVIDIA GeForce GTX 580/680 are recommended, in order to get smoother user interaction and better rendering quality. Lower-end NVIDIA and AMD cards may also run, possibly with slower speed and/or occasional crashes.
If you want to install the higher-end graphics cards, please check the power supply reuirement and form factor of the cards. These cards may require at least 500 watt system power supply. Also they are full-height and very likely full-length, so they may not fit into small cases.
A display resolution of at least 1200 x 900 is required to show all of the rendering and parameter setting panels.
On a Mac system, please install Windows with Boot.
A comparison of features between three commercial volume rendering packages (Amira, Volocity, Imaris), a freeware package (Osirix), and FluoRender.
|Amira 5||Volocity 5||Imaris 6||Osirix||FluoRender|
|Layered Render Mode||-||-||-||-||V|
|Depth Render Mode||V||V||V||V||V|
|Composite Render Mode||-||-||-||-||V|
|Save Volume Settings as Default||-||-||-||-||V|
|Low Threshold Setting for Volume Data||-||V||V||*V*||V|
|High Threshold Setting for Volume Data||-||-||V||-||V|
|Setting RGB Values Easily||-||V||V||-||V|
|Saturation Point Setting||-||*V*||V||*V*||V|
|Sampling Rate Adjusting||-||-||V||V||V|
|Extracting Volume Boundary||-||-||-||-||V|
|Displaying XYZ Rotation Angles||-||-||-||-||V|
|Displaying Zoom Ratio||-||-||V||V||V|
|Clipping Channels Independently||-||-||-||-||V|
- In Amira, only Maximum Intensity Projection is effective for visualization of confocal data.
- In Osirix, volume properties cannot be adjusted independently for each channel.
- [*V*] denotes multiple parameter settings that are controlled by one widget, and cannot be set independently.
User DocumentationDownload a printer friendly file of the documentation
After downloading the installation package, double-click the icon of the package, then installation will start. Go through the installation process as shown in the screen captures below.
Before Using FluoRender
FluoRender reads muliple file formats including confocal files from microscope manufacturers such as Olympus and Zeiss. You can also use tiff as a convenient exchange format, which can be converted with ImageJ.
An ImageJ macro allows conversion from all major confocal manufacturers' formats; for details about the macro, visit the plugins tab.
A Quick Start Demo Video (Ver. 1.0):
User Interface of FluoRender (v2.10.0):
Instructions for Basic Operations:
I. Click "Open Volume" icon in the toolbar (A). In the file choosing dialog select the volume dataset (multiple files can be selected at the same time). The selected datasets are loaded in the "Datasets" panel (B). The file choosing dialog contains additional options for loading 3D and 4D sequence. You can load several file formats, including TIFF, OIB, OIF, LSM, and NRRD. The chosen volume data are added automatically into the "Render View" (F) and "Workspace" (C).
II. Click "Open Mesh" icon in the toolbar (A) to choose one or multiple polygon mesh file(s). Only Wavefront's OBJ format is currently supported. If you export OBJ files with textures from a modeling tool, please use TGA as the texture format. Transparency of the OBJ data can be set either within FluoRender or through a texture with alpha channel.
III. When OIB, OIF, or LSM files are loaded, FluoRender will try to acquire meta data from these raw formats and set voxel size and channel colors accordingly. If no meta data are available, colors are assigned automatically as red, green, blue, then white, according to the order that datasets are added to the viewport. Colors can also be set with the color chooser in the "Properties" panel (H), or simply double-click on the RGB color values and type in the desired RGB values, or else R, G, B, W, or P, which are shortcuts for the preset colors.
IV. Viewport interactions are controlled by the mouse; a three-button mouse is ideal. To rotate the view, click and drag with left mouse button; to move the view, click and drag with the middle mouse button, or hold down the ctrl key and drag with the left mouse button; to zoom the view, click and drag with right mouse button.
V. Rendering properties of the volume dataset are displayed in the "Properties" panel, and each channel has its own settings. To switch between the channels, click the channel name in the "Workspace" panel (C). Settings for rendering properties are discussed in detail below.
VI. To switch render mode, click on the radio buttons above the "Render View" (F).
VII. To save the rendering results as an image, click "Capture", and enter an image file name.
VIII. Loaded datasets and their settings can be saved as a FluoRender project file. To save a project file, click the "Save Project" icon in the toolbar (A) and enter a project file name.
IX. To open a project file, click the "Open Project" icon in the toolbar (A), and choose the file. The saved project is loaded with all the datasets and their settings.
- The Project file remembers loaded datasets by their paths, and when the paths to the datasets change, FluoRender will try to search the datasets within the project file folder. So when you copy projects, copy datasets and project files together in order to open them correctly later.
User Interface Elements in Detail
- Open Volume. Choose volume datasets to open. Supported formats: TIFF, OIB, OIF, LSM, NRRD. Manufacturers' raw formats usually contain meta data, which help to determine the data type. However, to load datasets without meta data, you probably need to set the additional settings in order to load 3D and 4D sequence correctly.
- Open Project. Open a saved project file. Starting from version 2.8, project files are associated with FluoRender executables. Double-clicking the project files will launch FluoRender automatically.
- Save Project. Save all current settings to a project file.
- New View. Open a new render view.
- Show/Hide UI. Show or hide the UI panels. The individual panels can be show or hide from the drop-down list.
- Open Mesh. Open polygon datasets. Supported formats: OBJ.
- Edit. Open the Edit dialog. You can customize paint brush behavior and do calculations with two loaded volumes.
- Settings. FluoRender settings.
- Check Updates. Check the latest version of FluoRender (Internet connection required).
- facebook. FluoRender's facebook page.
- Twitter. Follow FluoRender on Twitter.
- About. Version and contact information.
B. Datasets Panel
All loaded datasets are listed in the Datasets panel. Double-click on a dataset, or select it and hit "Enter" to add it to the latest opened render view. Right-click an item to show its contextual menu. You can choose a render view to add the selected dataset, or you can delete, rename and save (for volume datasets only) the selected dataset.
Datasets panel has its own toolbar.
- Add selected dataset to the render view.
- Rename selected dataset.
- Save selected dataset.
- Bake selected volume dataset. It saves the volume with its property settings (gamma, thresholds, etc.) applied.
- Delete selected dataset.
- Delete all datasets.
Workspace lists and organizes all active datasets that have been loaded to the render view. Right-click the name of a render view to show the contextual menu. You can toggle the visibility of the view, collapse or expand part of the tree, or add a new group. Right-click the name of a dataset to show the contextual menu. You can toggle its visibility, remove a dataset from the render view, or remove the dataset from the view. Double-click a render view, group, or dataset to toggle its visibility.
To create a group, choose "Add Group" from the contextual menu, or drag any data channel to the lower empty space of the scene view window. Output adjustments (gamma and brightness) can be set for each group individually in the layered and composite modes. To set these parameters, choose the group or a data channel within the group, and use the Output Adjustments panel (E). In layered and composite modes, the settings only affect the chosen group, while in depth-sorted mode, the settings will affect all channels.
The order and hierarchy of the groups and channels can be changed by dragging and dropping the selected group or channel on the tree. The order of the channels will affect their rendering order in layered mode: channels that are higher in the window will cover those that are lower.
The currently selected volume data is ready for volume paint selection and editing. You can use different types of brushes to highlight voxels of the selected volume data, and you can also extract or remove the highlighted parts from the volume data.
Workspace has its own toolbar.
- Hide/show current selection.
- Add a volume data group.
- Add a mesh data group.
- Delete current selection. Id the current selection is a dataset, it is removed from the view port, and becomes "inactive". You can load it again from the Datasets panel.
- Switch to painting mode and use a brush to highlight voxels of the selected volume dataset. You can also hold down the Shift key on keyboard as the shortcut.
- Set painting mode to diffusion. Diffusion currently highlighted voxels of the selected volume dataset. You can also hold down the Z key on keyboard as the shortcut.
- Set painting mode to reset. Paint on highlighted voxels and remove them from selection. You can also hold down the X key on keyboard as the shortcut.
- Reset all highlighted voxels from selection.
- Erase highlighted voxels from the selected volume data. A new volume data will be created with "_DELETED" in its name. If you apply erase operation again on a volume with "_DELETED" in its name, no new volume will be generated.
- Extract highlighted voxels from the selected volume data. A new volume data will be created and added to the render view in a new group. You can also save the result from the Datasets panel.
D. Movie Exporting
Set movie exporting properties in the "Make Movie" panel. (For 4D naming and file format, see instructions.)
- Capture View. Choose the render view you want to use to capture the movie from the drop down list.
- Help. Click to open the default web browser and display this online manual.
- Movie Modes.
- 3D Rotation - Set the rotation axis and angles to make the rotation movie of the datasets.
- 3D Batch - Make a series of rotation movies for several sets of datasets, using the same settings as 3D Rotation.
- 4D Seq. - Make 4D time sequence rendering without rotations. Each time point is saved separately.
- 4D Rotation - Make 4D time sequence rendering as well as rotations. The rotation settings are the same as those of 3D Rotation mode. Each time point is saved separately.
- Time/file selection. In 3D Batch mode, use the slider, spin buttons and input box to select different files within the same folder for batch processing. In 4D mode, use the same UI to select current time point.
- Rotation Axis. Set the axis of rotation for animation.
- Rewind. Checking this box sets the animation to go back to the initial view after reaching the second angle.
- Angle. Set the range of the rotation angle for the animation. For example, (0, 360) starts from the current view, and rotates the view 360 degrees forward around the chosen axis; (-90, 70) starts from the current view, rotates 90 degrees backward, rotates 90 degrees forward to the initial view, rotates 70 degrees forward, and rotates 70 degrees back to the initial view (rewinding is set automatically when the first entry is nonzero).
- Degrees/frame. Set degrees of rotation angle per frame. This value is linked to "Total frames". Total frames. Set the desired total number of frames for the animation. This value is linked to "Degrees/frame".
- Time range. Set the start and end time frame for movie export.
- Preview. Preview the animation or batch processing.
- Stop. Stops the animation or batch processing during preview or Save Movie.
- Save movie... After choosing a filename for the exported movie, the animation or batch processing will be saved as a tiff file sequence.
- Reset Angle. Goes back to the initial angle.
- Enable framing. Check to enable the clipping frame for movie export. Only the area within the yellow frame is captured.
- Reset frame. Center and recalculate the size of the clipping frame to fit within the current view size.
- Center. XY coordinates of the clipping frame. Values are in pixels.
- Size. Width and height of the clipping frame. Values are in pixels.
E. Output Adjustments
Output adjustments allow you to adjust gamma, brightness and equalization for each RGB channel of the final output images, after the volume data are rendered. This is usually for fine-tuning the output images.
- Red channel link. Check this box if you want the adjustment of the red channel of the output image link to other channels.
- Red channel gamma, brightness and equalization adjustment slider.
- Red channel reset.
- Red channel gamma, brightness and equalization input values.
- Green channel link. Check this box if you want the adjustment of the red channel of the output image link to other channels.
- Green channel gamma, brightness and equalization adjustment slider.
- Green channel reset.
- Green channel gamma, brightness and equalization input values.
- Blue channel link. Check this box if you want the adjustment of the red channel of the output image link to other channels.
- Blue channel gamma, brightness and equalization adjustment slider.
- Blue channel reset.
- Blue channel gamma, brightness and equalization input values.
- Save as Default. Save all the settings as default. Data loaded next time is set to this default settings.
Note 1: In layered and composite modes, since channels are rendered independently, we can group channels into groups and set the output adjustment settings for each group independently. In depth-sorted mode, the settings affect all channels irrespective of how they are grouped. Refer to the instructions on Scene View to see how groups are created and organized.
Note 2: The gamma, brightness and equalization settings can be set independently for the R, G, and B channels, or channels can be linked by checking the checkboxes. Under certain conditions, some R, G, and B channels are linked automatically. For instance, when the R and G channels of one dataset have nonzero values, they will be linked; this auto-linking allows hues to be kept constant when changing the output adjustments.
Note 3: Equalization can be used to equalize uneven brightness for both 3D and 4D renderings. Equalization also enhances details at the same time.
F. The Render View Panel and Its Controls
The render view panel displays the rendered dataset; render view properties are set by the controls arranged around the border of the panel.
- Render Modes. Render modes control how multi-channel volume datasets are rendered. The three modes are: Layered mode: similar to Photoshop. Channels are layered on top of one another, with the rendering order specified by their order in the "Scene View" panel (C). Depth mode: channels are rendered according to their depth values; this mode is spatially faithful to the data. Composite mode: channels are rendered individually and then blended together for the final rendering. For groups, the render modes can be overriden by the depth mode. Turn on "Group: Depth" in the properties panel of any dataset within the group to set the group into depth mode. See H. Rendering Properties for details.
- Capture. Click to take a screen shot of current rendering and save as a tiff file.
- Persp/Ortho/Free Flight. Choose perspective, orthographic views, or free flight mode.
- Background Color. Sets the viewport background color.
- Center. Toggles display of the viewport camera's center of interest.
- FPS. Toggles display of rendering rate (in frames per second).
- Depth Int. The checkbox toggles depth attenuation. The slider sets the intensity of the effect.
- Zoom. Moves the viewport camera towards or away from the center of interest (changing the overall size). This is equivalent to dragging with the right mouse button held down, but with more accuracy. The "1:1" button sets the zoom to reflect the 1:1 voxel to pixel ratio. Click this button and it uses one screen pixel to render exactly one voxel of the volume dataset.
- Reset XYZ rotation angles to zero.
- XYZ Rot. Sets the orientation of the viewport camera. XYZ rotation values can be set by numerical entry, by slider, or by dragging with the left mouse button held down.
- Save as Default. Save all the render view settings as default. The next time opened views are going to use the default settings.
- Clipping frame. The yellow rectangle shows the area of the viewport to be captured when exporting a movie.
- Camera center. The camera center widgets shows the position of the center of camera and its rotations.
- Scale bar. This indicates the physical scale of the dataset, but will only be accurate in orthographic (not perspective) view. The length of the scale bar can be set in the Rendering Properties panel (H).
- FPS. Display the rendering speed if FPS is enabled.
G. Clipping Plane Settings
These settings determine the orientation and position of front and back clipping planes, which delimit the part of the volume to be displayed.
- Sync all chan. Check this box to synchronize the clipping planes of all the channels of the open dataset. If unchecked, each channel can have independent clipping planes.
- X1, Y1, and Z1 clipping plane control sliders. Use these sliders to control the positions of the front clipping planes.
- X1, Y1, and Z1 clipping plane position inputs. Control the positions of the front clipping planes by numerical entry.
- X, Y, Z clipping plane link. Use the check boxes to link the movement of the front and back clipping planes.
- X2, Y2, and Z2 clipping plane position inputs. Control the positions of the back clipping planes by numerical entry.
- X2, Y2, and Z2 clipping plane control sliders. Use these sliders to control the position of the back clipping planes.
- Align to View. Click this button to align the six clipping planes to the view, so that Z1 is facing to the view, Z2 is facing off the view, X1 is on the left hand, X2 is on the right hand, Y1 is on the top, and Y2 is on the bottom.
- Reset to 0. Click this button to set the clipping planes rotation back to zero in the object space.
- Clipping plane rotation angle inputs. These numbers are the relative rotation angles of the clipping planes to the object. Set these numbers to rotate the clipping planes independent of the object.
- Clipping plane rotation angle sliders. These sliders control the rotation of the clipping planes relative to the object.
Clipping Plane Usage in Detail
When the mouse cursor moves over the clipping plane settings panel, the clipping planes of the current selected channel will be displayed in the render view (A). Only the clipping plane being modified will be displayed when moving the sliders (B). The 6 clipping planes act like a box to clip the dataset (C). If the two clipping planes of one axis are linked, one will move with the other, keeping a constant separation, and both clipping planes will be displayed in the render view (D).
H. Rendering Properties
Volume properties are displayed in the "Tools" window, when the name of the dataset is selected in the "Scene View" window. The "Tools" window is dockable, so that the user can rearrange its position by dragging its title bar. Most values can be controlled either by slider or by numerical entry.
Note: Starting from FluoRender 2.9, render view updates instantaneously without the need to hit Enter key for the above numeric inputs. Earlier versions still require user to hit Enter key after input.
- Gamma. Controls the shape of the nonlinear transfer curve. Setting a value greater than 1 accentuates signals of low intensities, while a value less than 1 suppresses signals of low intensities. A low value is usually desirable for confocal datasets, where noise is often present as low intensity signals.
- Extract Boundary. Controls how much boundary-like structures are extracted.
- Saturation point. Sets an intensity threshold above which all values are mapped to the highest output intensity. Moving the slider to the left results a steep transfer curve with more pixels saturated.
- Threshold. Sets an intensity threshold below which voxels are not displayed in the output with the first slider and sets an intensity threshold above which voxels are not displayed in the output with the second slider. This setting is independent of gamma value setting. In MIP mode, this setting is disabled.
- Luminance. Controls overall brightness of the final rendering by changing the highest output intensity.
- Shadow/Light. Check the check box to enable shadow effects. Use the first slider to adjust shadow intensity. Use the second slider to adjust highlight intensity. Shadow effects can be directional. Set the shadow direction in the Settings dialog (see below).
- Alpha. Controls the transparency of the voxels. In MIP mode, this setting is disabled.
- Sample Rate. Controls the quality of the output rendering (higher rate = better quality, slower rendering).
- Shading. Controls the output intensity of the shaded area of protruding structures.
- Colormap. Assigns a colormap instead of a single color to the selected data. The colormap has a gradual change from blue to green to red depending on the scalar values of the data. User can control the blue point (for low scalar value) and red point (for high scalar value) with the sliders and input boxes. When enabled, a legend of the colormap currently used will also be displayed in the rendering view.
- Color. Color used to display this dataset. Double-click to select the entire text window. Shortcuts for commonly used colors: R/r - red, G/g - green, B/b - blue, P/p - purple, W/w - white. Color is overriden when colormap is enabled.
- XYZ Spacing. Sets the voxel size, using the unit of length (e.g. microns) from the scale bar settings. Sizes are linked between channels, so that settings for one channel affect all channels in the same viewport. If the metadata of the TIFF file contains the voxel size of the data, these values are filled automatically by the program.
- Scale bar settings. The checkbox toggles the display of the scale bar on/off. The input box and drop down list set the length of the scale bar and the length unit. The "Show Text" checkbox toggles the text label for the scale bar on/off.
- NOTE: Though the scale bar can be displayed in both perspective and orthographic views, its length is correct only in orthographic view. In perspective view, its length is imprecise due to foreshortening.
- MIP. Turn on MIP mode for the data. In MIP mode (without colormap), alpha and high threshold are disabled and low threshold becomes the threshold for shading only. Use the shading in MIP and adjust the shading to render the surface details properly. In color-mapped MIP mode, gamma, saturation point, luminance, alpha, extract boundary, and high threshold are all disabled. Use the Output Adjustment (E) to fine-tune the brightness and contrast. The MIP mode only works in Layered and Composite render modes. Switching to Depth mode automatically turns MIP mode off, and a warning dialog box is shown when the user attempts to enable MIP in Depth mode.
- Smoothing. Apply filers in real-time and remove the artificial patterns. The results are usually slightly dimmer than without the filtering due to its noise suppression effect, but the renderings should be smoother, especially in the low signal intensity regions.
- Synchronize Group. Enable this option synchronizes the settings of all the data within current group.
- Depth Mode in Group. Enable this option to set the group of the current dataset into depth mode, irrespective of the current render mode of the render view. The datasets within the current group are rendered with the correct depth occulusions, and the group are combined with other datasets according to the current render mode, which should be layered mode or composite mode.
- Reset. Resets all settings as the saved default.
- Save as default. Saves the parameter settings as the default, which are applied to volume datasets loaded later.
Open volume options
There are several options in the Open Volume dialog. Mask volumes, 3D and 4D sequence files are loaded by selecting these options.
- When a volume dataset is saved as a series of 2D images, check this option and FluoRender will try to load the sequence as a volume. Please note that you probably need to edit the names of the image sequence, since FluoRender will try to find the last occurrence of consecutive digits and use them as the index for the sequence.
- This is the identifier for a time sequence. Similar to the above option for 3D sequence, FluoRender uses consecutive digits after the identifier as the index for a time sequence.
Note to previous version FluoRender users: Previous versions of FluoRender always use "_T" as the identifier, if you want to open 4D sequences created previously, please make sure this option is set correctly.
Click the Edit button from the Toolbar (A), and open the Edit dialog. Use the settings to control brush behavior for volume paint selection. There are two ways to edit volume datasets: 1 use the volume paint selection; and 2 use volume calculations. All volume datasets loaded into Workspace and shown in the render view are ready for volume editing.
To use volume paint selection, select the volume dataset from the Workspace, and then start with the first brush in the toolbar. You can also hold down the Shift key. Adjust the brush size with either mouse wheel or the options in the volume edit dialog. Then paint in the view port to highlight voxels you want to select. You can use the diffusion brush to diffuse the highlighted voxels into regions that are connected to the existing highlights. Or you can use the reset brush to remove the highlights. When you think you are done with painting, you can either erase the highlights with the eraser in the toolbar, or extract the highlights out with the extract tool in the toolbar.
The following figure illustrates the usage of the three brushes:
You can select two volume datasets and calculate using four volume calculations:
Subtract. Subtract Volume B from Volume A.
Add. Add Volume A and B together.
Divide. Divide Volume B from Volume A.
And. Find the common parts of Volume A and B.
- Falloff controls how low intensity voxels are selected. Setting a lower number will select more low intensity voxels. However, when you use the diffusion brush, more voxels are considered "connected".
- The influence of the output adjustments. You can use Gamma, Luminance, and Equalization to influence the volume painting results. This is the factor controls the strength of their influence.
- It controls the size of the inner circle of the brush.
- It controls the size of the outer circle of the brush.
- It controls how many iterations are used for calculating the growth. With more iterations, growth is stronger.
- Display this help page.
- Save the painting selection parameters as default.
- Select a volume dataset from the Workspace and click "Load", the selected dataset will be ready for calculations as operand A.
- Select a volume dataset from the Workspace and click "Load", the selected dataset will be ready for calculations as operand B.
- When Volume A and B are both ready, click any one of the four calculations to get the result. The resulting volume will be add to the Render view within a new group.
Click the Settings button from the Toolbar (A), and open the FluoRender Settings dialog. Use the settings to control the global behavior of FluoRender upon all the datasets.
- Auto Save Project. Enable this option so that every time FluoRender captures a rendering or exports a movie, it is going to save the current settings as a project file, in the folder where the files are exported. You can go back and open the project later.
- Variable Sample Rate. Enable this option to automatically lower the sample rate for volume data when using mouse interactions.
- Micro Blending. In depth mode (including the depth mode within a group), enable Micro Blending and FluoRender is going to blend the colors from different channels at voxel level, which usually can generate more accurate colors. This is especially useful for RGB images, since e.g. green and red colocalized at the same place will result in yellow. And the blending is not restricted to three channels in RGB, any number of channels can be blended.
- Note: This feature is going to slow down the rendering speed in depth mode. Depending on the system, the factor of performance drop may vary.
- Mesh Transparency Quality. This options sets the layers of transparent mesh objects to render. It only works when mesh files are opened and added to the render view. The more layers set, the more accurate of the rendering results, but it also slows down the rendering speed quite significantly as the number of layers increases.
- Shadow Direction. Check the check box to enable directional shadows. Use the slider to control the light source position.
- Set default colors for confocal channels with different excitation wavelengths. This only works for confocal raw formats that wavelength info can be acquired from meta data.
- Note: The settings in this dialog are applied to FluoRender instantly when you change the settings. The settings will remain in effect until FluoRender exits, even the settings are not saved. FluoRender is going to use the saved settings upon next start.
ImageJ Macros & Plugins for 3D Confocal File ConversionThe most commonly-used input data format for FluoRender is 8-bit multi-layer TIFF files; manufacturer-specific formats (Zeiss, Leica, Olympus, Nikon) can be converted to multi-layer TIFF using a custom ImageJ macro. There is a guide for basic ImageJ operations available: ImageJ_guide.pdf.
- Download ImageJ from ImageJ website:
- Launch ImageJ.
- From the Edit menu select "Edit->Options->Memory & Thread", then increase the memory allocation. For 32-bit systems, use 1450MB; if a greater value is set, ImageJ fails to start. For 64-bit systems, higher settings are possible--set the value just below the total system memory.
- To use the macro, install the "loci_tool" plugin for ImageJ first.
Download: plugins.zip 5MB
- Unzip the downloaded zip file and copy the files into the folder ImageJ/plugins/. Then restart ImageJ.
- Download the format conversion macro:
This macro converts the original confocal data file (Zeiss, Leica, Olympus, or Nikon) to an RGB multi-layer tiff file. The output files are placed in a "RGB_tif" folder, with "_1ch, _2ch, _3ch" appended to the filenames.
This macro extracts channels from the original confocal data file (Zeiss, Leica, Olympus, or Nikon) and generates an 8-bit multi-layer tiff from each channel. The output files are placed in a "TIFF_Files" folder, with "_B, _R, _G, _W" appended to the filenames.
This macro is similar to the macro above, except that before saving as multilayer TIFF, it resamples the data from 1024x1024 px to 512x512 px. Use this resampling when graphics memory is insufficient. Attention: This macro fails to interpret the confocal data correctly if multiple samples are stored in a single file. Instead, store each sample as a separate file.(The resulting 8-bit multi-layer file is usually mentioned as a volume dataset.)
- Open the macro file with ImageJ (drag-and-drop is the easiest method), then from menu select "Macros->Run Macro".
- In the dialog box, select the folder where the input confocal data are. The macro reads all files in the folder, so keep only confocal data files there.
About the loci tools plugin
- We use the loci tools plugin (website http://www.loci.wisc.edu/ome/formats.html) to read various formats of confocal files into ImageJ. As of April 2010, the latest version has become incompatible with the Save_8bit_multiCH macro, so use the older version of the plugin provided above (plugins.zip).
ImageJ Macro for 4D Confocal File Conversion and File Format Specifications4D confocal files in OIB, OIF and LSM formats can usually be loaded correctly without any conversions. Other formats need to be converted to TIFF and follow certain naming conventions.
For easy converting and naming the manufacturer-specific confocal files, an ImageJ macro is provided for download. It can be downloaded here: 4DmultiCH_FluoRender.txt.zip
To use the macro, install the "loci_tool" plugin for ImageJ first: plugins.zip (5MB)
There is also a guide for basic ImageJ operations available for downloading: ImageJ_guide.pdf.
How to run the ImageJ macro:
- Open the macro file with ImageJ (drag-and-drop is the easiest method), then from menu select "Macros->Run Macro".
- In the dialog box, select the folder where the input confocal data are. The macro reads all files in the folder, so keep only confocal data files in the folder.
After the files are converted, make sure they are saved in the same folder and named properly. Just open one time point from the whole sequence. If the dataset has multiple channels, load all the channels from one time point. Choose "4D Seq." from the "Make Movie" panel (D in the manual), and the whole time sequence should be recognized. Use the time slider to browse through the data, and use "Save..." to export the whole sequence.
1. Spaces " " and hyphen "-" in the full file path or file name cause the macro to read data files incorrectly. Use underscores "_" instead.
2. In order to convert the 4D file format, ImageJ will read all the dataset into system memory. So ImageJ 64bit version and system memory more than 6 GB are recomended.
4D file naming conventions:
In order to render 4d movies correctly, certain naming and formating conventions need to be followed, which is described below. If you have any problems of 4D rendering, please first check if the file names follow these convetions.
For multiple channel 4D data, each channel of each time point is saved in a separated file. The file name should contain a pattern as "_T###", and "###" denotes the number of current time point, which should increase continuously and has the same number of digits for all time points. Different channels of the same time point can be distinguished by any additional words, but the same channel of different time points should be identified by the same word.
For example, a 4D dataset contains 120 time points and 3 channels, and we can name each individual file like:
For time point 1,
For time point 2,
For time point 120,
In the above example, "dataset_name" can be any thing that identifies all the files within the same 4D dataset, and it should be the same for all files. "_T###" is used to identify the time point number and should be the same for all the channels of the time point. The number to identify the time point is increased by 1 for each time point and all the numbers have 3 digits with zero padding. "_R", "_G", and "_B" are used to identify channels and they should keep the same for different time points.
1. I cannot start the software
Q: I downloaded FluoRender and double-clicked the FluoRender icon to run the program, but it popped an error dialog window, saying "The program can't start because teem.dll is missing from your computer. Try reinstalling the program to fix the problem.". What's the problem and how can I fix it?
A: Probably you didn't extract (or unzip) the entire folder from the downloaded zip file, so you were running FluoRender without the necessary library files. Under Windows, some commonly used zip/unzip programs include WinZip and WinRAR. If you are not using any zip/unzip programs, you have to copy the entire FluoRender folder inside the downloaded zip folder to a certain directory of your hard drive, such as the C:\ root directory or your desktop.
Q: When I run FluoRender I get the error message "This application has failed to start because the application configuration is incorrect. Reinstalling the application may fix this problem." Any ideas?
A: It's very possible that you run FluoRender on a newly installed system and the Microsoft Visual C++ 2005 Redistributable Package has not been installed. We currently compile and build FluoRender with Microsoft Visual C++ 2005 and FluoRender needs the Redistributable Package in order to run. Most systems already have this package installed, but if you get the error message you can try to download the package form Microsoft. If you are using 32 bit Windows (x86), you should download the 32 bit version of the Redistributable Package, and if you are using 64 bit Windows (x64), you should download the 64 bit version of the Redistributable Package.
Q: When I start FluoRender, I barely see the user interface and it disappears immediately, or it won't start at all. What is the problem?
A: FluoRender uses certain graphics hardware features to render the volume data. Currently only a limited number of graphics harware is compatible with FluoRender. We recommend a recent discrete graphics card from nVIDIA. (GTX400 series)
2. I can't use FluoRender to generate any imagesQ1: I can run FluoRender and see the user interface, but after I load volume data into the program, it doesn't display anything. Why?
Q2: I can run FluoRender and see the user interface, but after I load volume data into the program, it stops responding and crashes. Why?
A: If you have these problems just occasionally, please refer to the Memory problem. FluoRender uses certain graphics hardware features to render the volume data. Currently only a limited number of graphics harware is compatible with FluoRender. We recommend a recent discrete graphics card from nVIDIA. (GTX400 series)
3. FluoRender crashes when loading files/projectA: Currently FluoRender loads all the data you open into both main system memory and graphics memory, and there is a limit the total memory FluoRender can use. If FluoRender crashes, 1) you can try to load less channels or resample the data (preferably with ImageJ). 2) please use 64 bit version of FluoRender (64 bit Windows system required).
4. Does FluoRender have a MacOSX/Linux/UNIX version?A: Currently we are only developing FluoRender on Windows platform, and we recommend running FluoRender on a Windows based PC desktop. Intel based MacBook Pro can run FluoRender by installing a Windows system on Bootcamp.
Sample Data1. Zebrafish Head
Confocal dataset of 5 day old transgenic zebrafish larva, stained for anti-actin, anti-GFP (recognizing the isl1:gfp transgene), and ToPro3 (nuclear stain). Voxel size = 1.25 x 1.25 x 3.0 microns.
Brig Bagley, Research Software Engineer
Scientific Computing and Imaging Institute, University of Utah
Yong Wan, Ph.D
Scientific Computing and Imaging Institute, University of Utah
GUI & Function Design; ImageJ Macro Coding:
Hideo Otsuna, Ph.D
Dept. Neurobiology & Anatomy, University of Utah
Scientific Computing and Imaging Institute, University of Utah
Chi-Bin Chien (decesed)
Dept. Neurobiology & Anatomy, University of Utah
Yong Wan*, Hideo Otsuna*, Chi-Bin Chien and Charles Hansen, "Interactive Extraction of Neural Structures with User-Guided Morphological Diffusion", in Proceedings of 2012 IEEE Symposium on Biological Data Visualization (BioVis), 2012.
Yong Wan*, A. Kelsey Lewis*, Mary Colasanto, Mark van Langeveld, Gabrielle Kardon and Charles Hansen, "A Practical Workflow for Making Anatomical Atlases in Biological Research", IEEE Computer Graphics and Applications, vol. 32, no. 5, 2012, pp. 50-60.
Yong Wan*, Hideo Otsuna*, Chi-Bin Chien and Charles Hansen, "FluoRender: An Application of 2D Image Space Methods for 3D and 4D Confocal Microscopy Data Visualization in Neurobiology Research", in Proceedings of 2012 IEEE Pacific Visualization Symposium (PacificVis), 2012, pp. 201-208.
Yong Wan*, Hideo Otsuna*, Chi-Bin Chien and Charles Hansen, "An Interactive Visualization Tool for Multi-channel Confocal Microscopy Data in Neurobiology Research", IEEE Transactions on Visualization and Computer Graphics, vol. 15, no. 6, 2009, pp. 1489-1496.
Joining the FluoRender Users Mailing List
- Nothing needed in the Subject line. Just send.
- You will receive a response from the list manager requesting confirmation. Just send back a reply message to confirm your subscription (no added text needed).
- Finally, you will receive a welcome message confirming that you have been added to the mailing list.
- Nothing needed in the Subject line. Just send.
- You will receive a response from the list manager notifying you that the command has succeeded and you have been removed from the list.