I'd like to share a story with the community about an application of ImageJ that really speaks to the spirit of open-source and its world changing potential. Three years ago, I joined a small Canadian company, Spectral Applied Research, which in recent years had switched its business almost exclusively to making products relating to biomedical optics (confocal microscopy, flow cytometry, etc.). Prior to that, the company had been involved in the space optics and remote sensing industry. By the time I joined, one final space optics project remained to be completed, that being the construction and testing of a micro-satellite commissioned by the Canadian Space Agency. This micro-satellite, called the Near Earth Object Surveillance Satellite (NEOSSat), was launched into orbit a few days ago on February 25, 2013 from India. NEOSSat is a space telescope designed for the purpose of discovering and tracking asteroids of the inner Solar System, within a region that is difficult to image by ground-based telescopes because of the brightness of the sun (more information about the micro-satellite's mission can be found at www.neossat.ca).
One task carried out in the final year of the NEOSSat project involved the characterization of the CCD detectors and testing of the system's overall imaging performance, a process known to those in the astronomy field as photon transfer analysis. At one point the team ran into some trouble trying to track down a noise source in the imaging system, and they were limited by a spreadsheet analysis program that was slow and lacking in real-time visualization features of the images generated by the cameras. The company president was aware of my use of ImageJ from my PhD days which I used primarily for microscopy applications, and one day he asked if ImageJ would be able to open and view one of the NEOSSat Images. Fortunately, ImageJ was able to recognize the image format, and immediately it was possible to visualize periodic noise patterns in the raw images. I taught the NEOSSat team the basics of ImageJ and how to manipulate and analyze the images using the simple image calculator and the basic measuring/statistics tools. Once they saw the power, speed, and ease of use of this program compared to the other one that they had been running, they asked me to write some basic macros that would automate the key steps of the photon transfer analysis. I am by no means an expert ImageJ programmer, and I still consider myself a novice, but the requirements of the macro were within my ImageJ macro writing capabilities.
I have been told by the company president that these ImageJ macros were critical for the completion of the NEOSSat project and meeting the launch date. Photon transfer data analysis trials which normally would have taken weeks to complete were reduced to a few hours. Eventually, other members of the NEOSSat team began using ImageJ to perform more complex types of near real-time analysis. It is my understanding that ImageJ is now even popular within the Canadian Space Agency itself.
I was compelled to tell the ImageJ community this success story because I feel it is a success that owes much thanks to those who use and develop ImageJ (not just the NEOSSat team), and the fact that it is open-source. Of course, much thanks is to be given to Wayne Rasband, the creator of ImageJ, but what makes the tool special is that its users can build upon its basic features through the macro and scripting functions, and that these macros/scripts can be shared openly, thus allowing even more users to use, share, and build upon them further. When I think about it, there must be a long line of successive users that had to recommend ImageJ that eventually led to a recommendation by my PhD advisor to learn ImageJ as well and create the NEOSSat analysis macros that came thereafter. Much of my macro writing skills, meager as they are, come from observing discussions on this email list-server and clipping parts of code found on the internet that did the job for me. ImageJ has previously been used by others for Astronomy applications, but it seems to have not found wide-spread use in this field. My hope in sharing this story, is that all users of ImageJ can take some sense of pride in the fact that the policy of openness and sharing that pervades ImageJ really can lead to good (and potentially world-changing) things. I think it's a great example of that hidden network of ideas and people (past, present, and future) that it is at the heart of science and engineering. Methods, and discoveries openly shared today which might seem insignificant to you now, could lead to something very useful for someone else tomorrow.
Sincerely,
John Oreopoulos
Staff Scientist
Spectral Applied Research
Richmond Hill, Ontario
Canada
www.spectral.ca
--
ImageJ mailing list:
http://imagej.nih.gov/ij/list.html
Locked
PastedGraphic-1.tiff (223K)