My name is Joel Andres Granados. I am a PhD candidate at the IT University of Copenhagen. I am currently studying the application of computer vision and machine learning techniques in the automation of current plan phenology processes.
I received my bachelor degree in Systems Engineering from EAFIT University (Colombia). At that time I worked with topology control in WSN. More specifically I was trying to evaluate the performance of a known topology control algorithm in Mobile Ad-hoc Networks (MANETs) called XTC. I graduated with a mention of honor in late 2006.
After my studies in Colombia I moved to Czech Republic to work for Red Hat. I had the opportunity to experience one of the fastest growing engineering offices (at the time) in Red Hat. I spent a total of 3 years working with the installer team. Though it was a very difficult decision, I left to pursue my Ph.D. in computer science in Copenhagen.
I began my Ph.D. on October of 2009 at the IT University of Copenhagen and am currently part of the software development group. My work is mostly related to automating arctic phenology measurements in a research facility on the mid-eastern cost of Greenland.
We are working with Biologists to automate and support plant phenology measurements in the arctic. The research is in Zackenberg. Our objective is to use state of the art image recognition processes in a Wireless Imager Sensor Network (WiSN) to detect specific flower types in previously selected plots of land.
Great constraints come from working in this part of the globe. Weather conditions make the deployments inaccessible for months. Electronic components are prone to damage, not only by the weather, but by the wild animals that reside in these places. While we are very interested in creating a working WSN system, our main research question has to do with robustness (or reliability) and how we can incorporate these concepts into our system and into WSN in general.