Welcome to GISWeekly! On Monday, November 29, I will be attending Autodesk University 2010 in Las Vegas, Nevada and will report on that event via
GISCafe Today blog as well as full reports appearing in AECWeekly on December 6 and GISWeekly on December 13. I hope to see many of you there. In the meantime, have a happy and nourishing Thanksgiving Day.
GISWeekly examines select top news each week, picks out worthwhile reading from around the web, and special interest items you might not find elsewhere. This issue will feature Industry News, Top News of the Week, Acquisitions/Agreements/Alliances, Announcements, New Products, Around the Web and Events Calendar.
GISWeekly welcomes letters and feedback from readers, so let us know what you think. Send your comments to me at Email Contact
Susan Smith, Managing Editor
Workflows and Integration Characterize ERDAS 2011 Release
By Susan Smith
Applications for the defense industry are obviously vital to the nation's security, however, many of those applications may find their way into the mainstream geospatial lexicon in years to come. By the same token, technology developed for commercial usage such as the cloud, cell phones and handheld tablets open up possibilities to the defense industry, whose users need to access servers from devices with low compute power -- without the worry of lack of bandwidth.
That need for processing power is part of the new soon-to-be-released ERDAS 2011 family of products, announced at GEOINT in New Orleans. The new 2011 family includes new versions of ERDAS IMAGINE, LPS, ERDAS Extensions for ArcGIS 10, ERDAS APOLLO and other leading desktop and server products. Press materials report that ERDAS Engine is a solution that boosts processing power for ERDAS IMAGINE and LPS, leveraging existing hardware resources for increased production needs or situations requiring faster workflow throughput.
According to ERDAS president Joel Campbell, the company has spent a lot of time improving their Image Analyst workflows. Most of their customers of this type of technology are in defense and other areas of security. Campbell is hopeful that others “in some areas of just traditional image analysis where they need to ingest, manage and organize imagery and are looking for features of some kind or intelligence” will be interested in the product as well. ERDAS has attempted to build tools to streamline that process for them to make it more wizard-like, adding some features to support that workflow, such as one-button push to PowerPoint, even publishing downline as KML so it can be readily viewed in Google Earth.
“While the biggest user community of that type of technology is defense, we also find people in port security and forestry who are looking at particular forest stands, who go through a similar process,” said Campbell.
These enhancement in Image Analyst workflows are a part of the IMAGINE product, and Campbell said they have not raised prices significantly in many years. “The defense guys are getting a real bargain.”
Among ERDAS' clients in Europe are the Dutch Cadastre, a big user of IMAGINE, as well as ERDAS' objective feature extraction capabilities. “They are doing lots of prototyping and they have a little of everything -- different software from different companies and some open source,” said Campbell.
The Port of Rotterdam is another Dutch customer, primarily an APOLLO user that employs the product to manage all the imagery around the port.
LPS 2011 is a workflow-oriented photogrammetry system for production mapping, including full analytical triangulation, the generation of digital terrain models, orthophoto production, mosaicking, and 3D feature extraction. In the LPS 2011 release is distributed processing, which enables users to leverage multi-core CPUs and multiple networked servers to increase production throughput and decrease the amount of time taken in processes that have been time-intensive until now.
Another major enhancement this year, to the LPS 2011 with the eATE add-on is the ability to collect terrain data. The new technology includes new terrain preparation tools as well as some enhancements to the automated terrain extraction tools.”
“This is opening up a new set of customers for us,” claimed Campbell. “In the space of engineering companies, photogrammetric shops, aerial survey companies, and even national mapping entities who are heavily engaged in production of terrain and constantly looking for more efficient ways to collect it. These may not have been customers historically.”
The new enhancement has the speed and accuracy of automatically creating terrain, shortening the time to do large areas into hours instead of days. Coupled with the ArcGIS extensions released in the weeks prior to the ERDAS 2011 release that has been updated to support ArcGIS 10, there are a few new features in the Stereo Analyst extension.
The Image Analyst workflow is inside IMAGINE. What crosses all ERDAS products n the new release is Bing Maps integration.
“We have entered into an agreement with Microsoft to allow us to provide access to Bing Maps to our customers,” explained Campbell. “In some cases that will be a product extension that a customer would acquire from us, in others like in the case of APOLLO access to the Bing extension is included as part of the product offering, but it gives access in a fully integrated way to the full width and breadth of the Bing Maps API. We treated it as any other data source, allowing you to perform functions against it, just like you would if you had a local image.”
Bing Maps integration gives customers a base map to start with, and a reference point if they are trying to align some data, or if they have imagery and want to see and align features. Bing Maps API offers layer control so users can look at just the roads, or point features or the imagery, giving a lot of opportunity to fill out a customers' working environment.
ERDAS has added support for the Microsoft SQL Server to APOLLO, and enhanced ECW support. Also they have launched a new SDK for ECW, and built in clustering support into the APOLLO server. This makes it possible for users to do clustered server configurations, manage some load balancing if you have a large implementation, enhanced cataloguing, increased ability and optimized the ability to deliver in tile formats through the Web Mapping Tile Service (WMTS), and built some new metadata editing tools that work in bulk instead of one record at a time.
Features continue to be added to the APOLLO server to make its management more effective, such as the ability to log in remotely and act as a full administrator, and configure services and the system. The catalogs between IMAGINE and APOLLO are also very tightly integrated in the new release. There is a “shoebox” in IMAGINE and one in APOLLO when retrieved.
“We have also enhanced the ability to be able to do some spatial modeling in IMAGINE and create a spatial model that does a flood zone analysis, and then publish that model as a web processing service in APOLLO,” Campbell said. “So someone can go to your APOLLO server and say, 'well I'd like to use this data and this service to find a flood zone area and it would do that processing on the fly.'
As the APOLLO and IMAGINE products continue to evolve independently they also work hand in hand with tighter integration.
Currently, just APOLLO is available on the cloud. ERDAS is working on how to leverage the cloud for remote processing for technologies such as eATE that might need access to many processors in order create terrain.
“There is this whole class of people that need access to lots of computing power but they only need it every now and then,” said Campbell. “Instead of buying a lot of infrastructure, we can leverage what they need it for in our context by leveraging the cloud.”
Campbell said the defense and military industry is interested in what might be termed a “private cloud” where data is hosted inside their environment but works the same way as the public cloud. The NGA is pursuing this alternative. There are many remote users who may have low compute power on handheld tablets or mobile phones, who want to ask questions on a server with the ability to have the answers delivered back to their devices, which translates to the need for simpler processing.