I always had in mind to implement a couple of filters for image restoration, image manipulation etc. Instead of coding all possible filters myself, I decided to use one of the open source available imaging libraries. My finally decision was to use Image Magick, more accurately Magick++, to do this task. Now all image filters availabe in Magick++ are available in SMIT. Those filters only work in 2D, slice-by-slice, I am not sure if 3D versions are necessary. Any comments? I decided to use Image Magick since it is available as Open Source, for different platforms, and seems to be a standard in Open Source Image Processing. If you have suggestions about using other libraries, please let me know. Enjoy!!
Finally, I had some time and the need for more SMIT functionality. For my project I needed to resample the image slices of a 3D data set. I wrote a loop that accurately resamples each slice, by splitting pixels in fractions and integrating them. This method is tested for downsampling, e.g. 128x128 pixels into 50x50, 12x12 etc. I hope this function is also useful for you.
Hello, I am happy to announce that I just added a function that allows to rotate the object within the data volume. It uses linear interpolation, so some quality is lost. I hope to be able to implement better interpolation schemes in the future.
I have uploaded a brief introduction in how to compile SMIT on Linux & Windows. It includes all the required preinstalled software, with some installation instructions.
Currently, I am developing SMIT on my own. I have many ideas how to improve it, and what additional functions would be desired by me and others to make this a comprehensive and competitive imaging software for academic and medical research. I am planning to add automated image registration, improve VoI analysis, add saggital and coronal views of 3D medical data, and in the future have a volume rendering engine. These are just a few main ideas, but they already exceed my time available for this project. So I am hoping for co-developers that are interested in working on this fun project that will help people working in the medical field.
After more than 2 years development of this Imaging Tool for medical and biomedical research, I have decided that it is ready for publication. SMIT is a platform independent (Linux, Mac OSX & MS Windows) tool which helps researchers adn M.D.'s to perfrom RoI/VoI Analysis, Image Registration, Data Fitting tasks. I started developing this software during my dissertation, and continue to improve it's functionality. The usage of Qt 4 makes it possible to distribute it to multiple platforms, and OpenGL makes the rendering engine as powerful as it is and opens more opportunities for improvements. SMIT is not nearly finished, but it is already well usable.