Cyber-green. Has that term been coined? If not, I am excited to do so here as my first contribution to the Imagery Speaks blog.
Think of this: repurposing data you already have, or using new approaches to extract meaningful information from the vast amount of data that already exists out there just waiting to be downloaded. This is a trend I’ve heard from many of the geospatial imagery users I speak with regularly and it brings phrases to mind like “Don’t reinvent the wheel”, and “Reduce, reuse, recycle.”
Take feature extraction for example. One time-tested way to extract a feature is to identify its unique spectral signature and look for that signature in the image (aka: spectral-based feature extraction). Enter object-based feature extraction, where not only does a spectral band have values in each pixel, but these values have minimums, maximums, means, and other attributes. And further, we can also consider spatial features such as rectangular fit, area, elongation, and even texture attributes. Now results that were previously not optimal due to low spectral resolution in the data can be improved by recognizing some of the other feature attributes like shape and size.
In the world of remote sensing data visualization and processing, are we approaching the realm where science and technology are enabling us to solve our problems using resources we already have? Does a “cyber-green” movement take us to a place of updated algorithms, optimized processing routines, and a myriad of other new and exciting approaches?
I’m not sure of the answer but have had a lot of fun exploring the data sources available. I’ve included some of these great data sources in this white paper. How do you think a “cyber-green” could positively impact your geospatial analysis?