Tacos and Data Fusion
My wife and I have a favorite restaurant in Winter Park, CO called Pepe Osaka's. If you think about the name for a minute you might guess it's a Mexican-Asian fusion restaurant. If you're not into guessing games, but wonder what they serve, here's the menu http://pepeosakas.com/menu/
As I was eating my spicy tuna fish taco, the sushi preparation of the fish drizzled with chili garlic salsa, somehow reminded me that I owe my management a definition of data fusion. I like sushi on its own, I like taco makings too. I could put them together at home, but not achieve the same result. Clearly, Pepe Osaka's is doing something special which makes their fusion so successful. I don't know exactly what it is, but the taste is transformed to a new space that exceeds simply Mexican + Asian.
I believe the same type of thing needs to happen to achieve geospatial data fusion. It's not as simple as overlaying two georeferenced data sets of the same region of interest. It's not just assembling physical properties of an object from multiple sensors. Fusion has to go further than this. I believe fusion must be able to extract the most essential information elements from the inputs before or during the process such that noise or irrelevant data is reduced from the solution. A fusion algorithm, at its most sophisticated level, should enable intelligent data sources that request from each other the most important contributions to the answer.
Are all these things are happening with a Pepe Osaka taco? Not sure. Maybe in their infused tequilas.
Would love to hear what you're doing with data fusion and what your wish list is for the future.