Microscopy Image Browser System Requirements
Back to Index
Contents
- Computer
- MATLAB
- Toolboxes
- LOCI Bio-Formats
- Compiled Frangi mask filter
- Compiled Image Edge Enhancing Coherence Filter
- Connection to OMERO server
- DipLib library
- Fiji: volume rendering and connection
- Membrane Click Tracker
- Read NRRD format
- Connection to Imaris
- SLIC superpixels, supervoxels and maxflow the Brush tool with supervoxels and for Graph-cut and Classifier
- Random Forest Classifier
- Volume Rendering
Computer
Microscopy Image Browser is a program written under Matlab environment and it was tested to work under Windows/Linux/Mac installation of Matlab.
In addition compiled (standalone) versions of MIB are available for Windows and Mac 64-bit OS. The standalone versions of MIB may not have complete functionality of the original program but do not require to purchase Matlab license. The standalone MIB requires MATLAB Compiler Runtime (MCR) that will be automatically installed during installation.
It is highly recommended to use 64 bit operating system with sufficient amount of memory.
MATLAB
MATLAB, Release 2011a, for saving MPEG movies R2012a is required. MIB was tested with Matlab R2011a - R2015b.
Toolboxes
- Matlab Image Processing Toolbox, V7.0 (R2010a) or newer is REQUIRED
- Statistics and Machine Learning Toolbox (optional) for alternative to the Random Forest classifiers
- Statistics Toolbox (optional) for alternative to the Random Forest classifiers
- Optimization Toolbox (optional) for alternative to the Random Forest classifiers
LOCI Bio-Formats
The Bio-Formats library brings support of multiple microscopy image formats. To use the library check the Bio checkbox in the Directory Contents panel of MIB.
LOCI Bio Formats java library (optional) is provided in the FILE_FORMATS/BioFormats folder.
Compiled Frangi mask filter
Compiled Frangi Mask filter is recommended for faster run. Please compile it for your OS. Most of C-functions can be compiled using a single script:
- In Matlab command window change directory to im_browser\Tools\, where im_browser is path where MIB is stored, for example c:\MATLAB\Scripts\im_browser
- To compile, type in Matlab command window mib_compile_c_files
Note! These files should be already pre-compiled for win32, win64 and mac64.
Compiled Image Edge Enhancing Coherence Filter
Compiled files are required for Image Edge Enhancing Coherence Filter. Please compile them for your OS. Most of C-functions can be compiled using a single script:
- In Matlab command window change directory to im_browser\Tools\, where im_browser is path where MIB is stored, for example c:\MATLAB\Scripts\im_browser
- To compile, type in Matlab command window mib_compile_c_files
Note! These files should be already pre-compiled for win32, win64 and mac64.
Connection to OMERO server
Connection to OMERO server requires the download of OMERO API bindings for Matlab.
- Download Matlab plugin for OMERO from here. It should be listed in the OMERO->OMERO Downloads, Plugins/Matlab section. (Note! Make sure that the version of Matlab plugin corresponds to the version of OMERO server you are going to login. Old omero downloads are listed in the Previous versions section at the bottom of hte page.
- Unzip the file to your scripts directory, for example C:\Matlab\Scripts\OMERO_5
- FOR MATLAB VERSION Add this directory (C:\Matlab\Scripts\OMERO_5) with subfolders to Matlab path (Matlab->Home tab->Set Path...->Add with Subfolders...) or run pathtool in Matlab command window
- FOR DEPLOYED VERSION Modify path to the omero_client.jar library in mib_java_path.txt (located in the first folder on the search path. Uusually [username]\Documents\MATLAB; path to this file will be displayed in the command prompt during start of MIB). For example, C:\Matlab\Scripts\OMERO_5\libs\ for Omero version 5, or C:\Matlab\Scripts\OMERO_4\libs\ for Omero verison 4
DipLib library
When DipLib is installed, Microscopy Image Browser can use several additional methods for anisotropic diffusion filtering available from DipLib. Please refer to the following step for installation.
- Download diplib from www.diplib.org/download (free for non-commercial use)
- Install the package, For example, start DIPimage 2.3 installer win 64.exe for automatic installation on Windows 64-bit operating system
- Start Matlab
- Add directory where the diplib/dipimage was installed into the Matlab path: Matlab->Home tab->Set path...->Add folder...->Move to Bottom->Save. For example, c:\Program Files\DIPimage 2.3\common\dipimage\
Fiji: volume rendering and connection
Microscopy Image Browser can use Fiji 3D Viewer plugin for visualization of volumes and models. In addition, there is a Fiji Connect panel that allows interaction between im_browser and Fiji.
!!! Warning !!!
It seems that Fiji 3D viewer is not compatible with Matlab on MacOS, please do not change write permission for the sys\java\jre\win64\jre\lib\ext on the MacOS
Below are the datails of the installation process.
- Download Fiji, Please use the Fiji Life-Line version, 2015 December 22 [link]
- Unzip and start Fiji (fiji-win32.exe or fiji-win64.exe)
- Update Fiji: Fiji->Menu->Help->Update Fiji (it may be required to repeat this step one more time after restart of Fiji)
- In the Fiji folder should appear a folder scripts with the Miji.m file
- The scripts sub-folder containing Miji.m should appear within the Fiji folder
- Start Matlab
- Add Fiji.app/scripts folder to the Matlab path and save the path: Matlab->Home tab->Set path...->Add folder...->Move to Bottom->Save. For example, c:\Program Files\Fiji.app\scripts\. For Mac OS use the addpath command from the Matlab command window, for example: addpath('/Applications/Fiji.app/scripts').
Note 1, the following Matlab path should be open for writing:
...\Matlab\sys\java\jre\win64\jre\lib\ext and ...\Matlab\sys\java\jre\win64\jre\bin
Note 2, if the Failed to retrieve Exception Message error appears, please increase the heap space for the Java VM in Matlab, see details here. For example, rendering of 1818x1022x717 volume requires 4Gb heap size.
Note 3, the Fiji 3D viewer may not work when started for the first time. In this case, Matlab should be restarted.
FOR DEPLOYED VERSION, to use Fiji with the deployed version of MIB please add path to your Fiji application to the mib_java_path.txt file located in the first folder on the search path. Usually [username]\Documents\MATLAB; path to this file will be displayed in the command prompt during start of MIB.
Membrane Click Tracker
Compiled files are required to use Membrane Click Tracker tool. Please compile them for your OS. Most of C-functions can be compiled using a single script:
- In Matlab command window change directory to im_browser\Tools\, where im_browser is path where MIB is stored, for example c:\MATLAB\Scripts\im_browser
- To compile, type in Matlab command window mib_compile_c_files
Note! These files should be already pre-compiled for win32, win64 and mac64.
Read NRRD format
Microscopy Image Browser uses an own function for saving data in the NRRD format, but relies on Projects:MATLABSlicerExampleModule by John Melonakos for reading it. On Windows OS the files should be already pre-compiled, but for Linux it may be needed to compile them.
Please refer to details in im_browser\ImportExportTools\nrrd\compilethis.m.
Connection to Imaris
Microscopy Image Browser can be used together with Imaris. This functionality is achieved with IceImarisConnector written by Aaron C. Ponti, ETH Zurich.
Please note, the connection is not available in the compiled version of MIB.
Requirements:
- Installed Imaris and ImarisXT
- Add path of Imaris installation to a system environment variable IMARISPATH. Start->Computer->right mouse click->Properties->Advanced system settings->Environment Variables...->New.... For example, IMARISPATH = c:\Tools\Science\Imaris\
- Restart Matlab
Note: it is recommended to put ImarisLib.jar to the static Java path of Matlab. To do that:
- Start Matlab and note the start-up (home) directory. For example: c:\Users\UserName\Documents\MATLAB
- Create javaclasspath.txt in this home directory and add path to ImarisLib.jar to this file (for example, c:\Program Files\Bitplane\Imaris x64 8.0.2\XT\matlab\ImarisLib.jar). One way to do that (Windows) is to type in the Matlab command prompt: system('notepad javaclasspath.txt'); add the path; and save the file
- Restart Matlab
SLIC superpixels, supervoxels and maxflow the Brush tool with supervoxels and for Graph-cut and Classifier
The brush tool can be used to select not individual pixels but rather groups of pixels (superpixels). This functionality is implemented using the SLIC (Simple Linear Iterative Clustering) algorithm written by Radhakrishna Achanta et al., 2015. In addition the SLIC superpixels and supervoxels are used for the Graph-cut segmentation and Classifier.
For the Graph-cut segmentation MIB is utilizing maxflow 2.22 written by Yuri Boykov and Vladimir Kolmogorov.
- In Matlab command window change directory to im_browser\Tools\, where im_browser is path where MIB is stored, for example c:\MATLAB\Scripts\im_browser
- To compile, type in Matlab command window mib_compile_c_files
Note! These files should be already pre-compiled for win32, win64 and mac64.
Random Forest Classifier
Compiled files are required to use Random Forest Classifier. Microscopy Image Browser uses randomforest-matlab by Abhishek Jaiantilal which is already compiled for win32, win64.
For all other OS the files have to be compiled manually: Please refer to details in
- im_browser\ImageFilters\RF_Class_C\README.txt
- im_browser\ImageFilters\RF_Reg_C\README.txt
Volume Rendering
Compiled affine_transform_2d_double.c function is required for volume rendering. Please compile it for your OS. Most of C-functions can be compiled using a single script:
- In Matlab command window change directory to im_browser\Tools\, where im_browser is path where MIB is stored, for example c:\MATLAB\Scripts\im_browser
- To compile, type in Matlab command window mib_compile_c_files
Note! This file is already pre-compiled for win64.
Back to Index