A popular topic these days is cloud computing. And the world of remote sensing is no exception. New developments in software, hardware, and connectivity are offering innovative options for performing remote sensing image analysis and visualization tasks in the cloud.
One example of the recent advance in cloud computing capabilities for geospatial scientists is the development of the ENVI Services Engine by Exelis Visual Information Solutions (Exelis VIS). Using what was previously the domain of desktop computing – this software engine brings the image analysis tools of ENVI into the cloud. This translates into an ability to deploy ENVI processing tools, such as image classification, anomaly detection and change detection, into an online environment. Additionally, because the system uses a HTTP REST interface and was constructed utilizing open source standards, implementing the software is feasible across a variety of operating systems and different hardware devices.
This flexibility of the ENVI Services Engine, and cloud computing in general, speaks directly to the “bring your own device” movement. Rather than being limited to certain operating systems or certain types of hardware, users have many more options to satisfy their preferences. Access and processing thus becomes feasible from a variety of tablets, mobile phones and laptops, in addition to the usual array of desktops and workstations.
As an example, consider the ability to access imagery and derived data layers from your favorite mobile device. Now consider being able to adjust your analysis on-the-fly from this same device based on observations while in the field. With the image processing being tasks handled on remote servers, extensive computing capacity is no longer required on your local device. This enables not just remote access to image processing, but also the ability for on-demand visualization and display of entire databases full of different images and results.
Having the image processing tasks performed on the same servers, or on servers closer to, where the imagery is stored is also more computationally efficient, since imagery does not need to be first transferred to local computers and results then transferred back to the servers. This is particularly relevant for large data archives, where even simple changes to existing algorithms, or the addition of new algorithms, may necessitate re-processing vast volumes of data.
Although the concept of cloud computing is not new, it has become apparent that the software and hardware landscape has evolved, making cloud computing for geospatial analysis significantly more attractive than ever before.
Attendees of the VISualize conference earlier this year received a sneak-peek at the ENVI Services Engine. The software was also recently on display at the GEOINT conference this past October. However, official release of the software isn’t scheduled until early 2013. For more information: http://www.exelisvis.com/