8966 Rate this article:
No rating

Image Analysis on Mobile, Enterprise, & Desktop

Mark Alonzo

I usually end my blog posts with a question, this week I’m starting out with one.  What does it mean to have image analysis on mobile, enterprise, and the desktop?  I’m going to give you my answer in this posting, but I’d like to hear yours, too.  So stay tuned for the question at the end.

Let’s start with image analysis on the desktop.  To me, this means running an image analysis application, like ENVI, on my local desktop.  It could also mean running the ENVI tools integrated with Esri’s ArcMap application on my laptop, either way it’s still an application accessing data on my local system using local resources such as the file system and CPU.  That’s how image analysis software and hardworking desktop machines have been put to good use for some time.

Essentially the same thing could be said for mobile devices.  ENVI doesn’t run on iPhone or Android, but there are some cool smart phone apps out there that could be used to put special effects on photos.  Conceivably, vendors could make lightweight versions of our favorite image analysis tools that would run on a mobile device so that we can take image analysis with us.  Then, I could take pictures (images) with my smart phone and do some analysis on the spot.  Of course, there are storage space and processing power limitations, but it is conceivable.

Both of these examples are what many would call “disconnected”.  The data and the tools reside and run on the same device or system without any dependency on other systems. A “connected” application is one in which the data and the tools are located on separate systems or devices. An app that requires a connection to a database to run is “connected”. The image below is a representation of the tricky, and somewhat ubiquitous, cloud.  Notice that the cloud is connected to all devices.  I’m going to consider the cloud to be the Enterprise for my purposes here but it could also be represented as one or a bunch of servers.


My original question asked about image analysis on mobile, enterprise, and desktop?  But what happens when we consider all of the platforms together?  If your data lives in the cloud, or on a server, you can pull it down to your desktop or mobile device for image analysis and viewing.  Derived imagery products can be pushed back to the cloud, or enterprise, and made available for viewing with desktop, mobile, or web clients.  I could view imagery hosted in the cloud on my mobile device, make a request from my mobile device to the cloud to run a process on that image, such as a line of sight, then view the results back in my mobile client.

I might do something similar from the desktop.  Let’s say I have a large image I want to process in order to do some further analysis on a smaller subset.  If the original image is hosted in the cloud, I could take advantage of that to run the first process, and then pull the result down to my desktop to continue my analysis.  I might then push the final result back to the cloud where I can access it later from my mobile device or a web client running on a virtual system.

Earlier, I talked about taking a photo with my mobile device for image processing which isn’t that far flung.  If my mobile device can act as a sensor, I can collect data that I then push back to the server so that other users can access it and perform their own processing.

I could go on, but hopefully you’re getting the idea.  I usually talk about enterprise as on-line or client-server environments, and that’s part of it.  However, enterprise can also refer to the interaction of data and applications across desktop, mobile, and online environments.

I’m headed to the 2012 Esri International Users Conference in a few weeks, there I expect to see a wide range of examples as to how we can combine GIS and image analysis on desktop, mobile, and enterprise platforms to solve some of today’s complex problems.  Perhaps I’ll see you there, too.  In the meantime, how do you combine, or want to combine, these tools and platforms?

Please login or register to post comments.