Wednesday, November 28, 2012

ERDAS ECW/JP2 SDK v4.3 Available

The latest versions of the ERDAS ECW/JP2 SDK (SDK), version 4.3 are now available on the Intergraph Geospatial Products website. These versions are a maintenance release. While addressing some maintenance issues, there were some slight improvement in memory handling of ECW and JP2 data, as well as some performance improvements.

This release it likely the final release for version 4 of the SDK. Version 5 of the SDK is in the final stages of development, and is already being used in various Intergraph products. Version 5 of  the SDK is shaping up to be a powerful product. There is no official date for release, but should not be far into the new year (2013), is my best guess.

Download ERDAS ECW/JP2 SDK v4.3 at the following web page:  

Wednesday, October 31, 2012

The Release of ERDAS IMAGINE 2013 is Coming Soon

The work on ERDAS IMAGINE 2013 is in the final stages. Indeed, all ERDAS Desktop products (ERDAS IMAGINE, LPS, ERDAS ER Mapper) are preparing for release. Keep an eye out for a press release from Intergraph before the end of this year (2012).

What do I like?

  • The Model Maker transition into Spatial Modeler 
    • Pull architecture for real-time processing to viewer 
    • Faster file creation 
      • Write real-time to 2D Viewer and create a file faster than Model Maker can make a file  
    • Pull architecture based on ERDAS ER Mapper concepts 
  • Python in Spatial Modeler 
    • Use custom designed applications in a Spatial Model 
    • Access other licensed software products in a Spatial Model 
  • GeoMedia vector and raster processing objects used in Spatial Modeler 
  • High performance asynchronous raster data engine 
    • Faster data access in 2D Viewer and Spatial Modeler 
    • Will be expanded in 2014 release 
  • Use of the new v5 ERDAS ECW/JP2 SDK 
    • Used in ERDAS IMAGINE, LPS and ERDAS ER Mapper 
    • Multi-threaded encoding 
    • Introduce ECW v3 file format (ECW v2 has been around since 1999)
      • ECW v3 supports up to unsigned 16-bit data 
      • ECW v3 uses GeoTIFF standard for coordinate reference system definition 
      • Expanded ECW metadata 
    • Faster encoding of ECW 
    • Faster decoding of ECW 
    • Faster ECW v2 and v3 opacity layer calculation 
    • Reduced need for J2I files with JP2 files 
    • More robust support for non-standard JP2 data 
    • More robust support for ill-defined JP2 profiles 
  • Support reading MrSID MG4 Raster  
  • Support reading of MrSID MG4 LiDAR 
  • Point cloud (LiDAR) support 
    • Viewing 
    • Automatic classification 
    • Manual classification 
    • Data preparation 
    • Support read and write of LAS 
  • New LPS user interface 
  • Improved Terrain Prep Tool 
  • Improved MosaicPro (we never stop making MosaicPro better) 
  • New swipe tools (transitions ribbon) 
  • New measurement ribbon for height from shadow and height from building lean and ortho measurements  
  • Real-time spectral indices (using Spatial Modeler technology) 
    • Vegetation 
    • Snow / Ice 
    • Minerals
    • Water 
  • New unsupervised classification tools 
  • Faster unsupervised classification tools 
  • Use of SAM and SCM in supervised classification 
  • New image segmentation algorithm in IMAGINE Objective 
  • Numerous interface improvements to make ERDAS IMAGINE and LPS more intuitive to expert and novice alike 
There is more, but these are what I really like. 

So, keep an eye out for the press releases... from Intergraph. Don't forget, ERDAS is part of Intergraph and   Intergraph is part of Hexagon. The Erdasians have helped our fellow Intergraph employees integrate well within Hexagon. 

Wednesday, September 19, 2012


The ERDAS ECW/JP2 SDK team in Perth Australia has done some work to update the ECW drivers in GDAL. These have been submitted to the GDAL community, and should make their way into a future version of GDAL.

There are multiple changes, but what I want to raise in this post, the performance of GDAL driver was improved in two main areas. While these bottlenecks were not in the ERDAS Desktop and Server product lines, we wanted to ensure our GDAL users had as similar a decoding performance as possible. 
  1.  GDAL 1.9 and earlier versions have a performance bottleneck when a single band is decoded line by line. This is not the preferred method of reading ECW data; nevertheless it is done by some folks. 
  2. GDAL 1.9 and earlier versions did not use the SDK’s implementation of supersampling, and GDAL’s own implementation was slower. Switching to supersampling using the SDK implementation halved time of the execution. 
While working on this small project, we did notice other areas where we can improve ECW support in the GDAL drivers going forward.

The Perth Team in Australia will improve the GDAL drivers more so when we upgrade GDAL to support the v5 of the SDK and the new ECW v3 file format early in the upcoming new year.

As has been policy, developer tools for decoding ECW and JP2 data using the SDK on desktop continue to be free.


Friday, August 3, 2012

Example Python Code from ERDAS IMAGINE's Upcoming Spatial Modeler

The chief Python developer, Fengliang provided me an example Python script that will run a resolution merge (pan sharpen) in the upcoming Spatial Modeler. This example model (below) runs outside of the ERDAS IMAGINE User Interface, indeed ERDAS IMAGINE is not running at all when the Py scipt is launched. ERDAS IMAGINE is installed and licensed, but not running.

The Py script finds all the ERDAS IMAGINE Spatial Modeler functions, and runs the model. This model runs 100% in the background. The ERDAS IMAGINE UI never launches. If the user wants to, they can run the model, start the UI and display the results as a real-time preview in the Viewer. Note, the preview is realtime modeling to screen, like ERDAS ER Mapper Algorithms. When using preview, a new file is never actually created. (But you can create one when you need to!) 

This example Py script, and others will be provided when the software is released.


def Sharpen( model, band, summary, panband ):
      return model.CastToFloat( model.Multiply( model.Divide(band, summary), panband))

# import erdas python module for spatial modeler
from erdas import modeler

# create a process
m = modeler.Model();

# create input operators
pan = m.RasterInput( "E:/Demo/Brovey/le7039035000009250_pan.img", "Float", "Nearest Neighbor");
multispectral = m.RasterInput("E:/Demo/Brovey/le7039035000009250_multi.img", "Float", "Nearest Neighbor");

# add three bands together
r = m.SelectBand(multispectral, '4:4');
g = m.SelectBand(multispectral, '3:3');
b = m.SelectBand(multispectral, '2:2');
sum = m.Add(1, r, g, b)

# calculate R, G, B bands for stacklayer
stack = m.StackLayers( Sharpen( m, r, sum, pan ), Sharpen( m, g, sum, pan ), Sharpen( m, b, sum, pan ) )

# to output operator
m.RasterOutput(stack, "E:/Demo/Brovey/output/brovey-output-python-6.img", "u8", Thematicity=modeler.Thematicity.continuous)

# finalize the process

Thursday, June 28, 2012

Last Few ERDAS Desktop 2012 Development Weeks Deliver Solid Results

During the last few weeks the ERDAS Desktop software development teams have delivered solid results.
  1. We have seen Point Cloud (LiDAR) display performance speed up significantly.
  2. User feedback from the Hexagon 2012 Conference prompted refinement in the LiDAR Ribbon controls.
  3. ERDAS IMAGINE’s Spatial Modeler providing real-time results from over 30 indices as well as pan sharpening.
  4. Python is now running Spatial Modeler without launching ERDAS IMAGINE UI.
  5. Asynchronous multi-threaded raster data delivery engine plugged into Spatial Modeler and into the 2D Viewer.
  6. Began work placing the latest ERDAS ECW/JP2 SDK into ERDAS IMAGINE, LPS, and ERDAS ER Mapper.
  7. MosaicPro improvements improves how customers can handle mosaicking images with clouds. 
There are more, but I can’t talk about all of them. There are a lot of great things happening in Atlanta, Perth, Singapore and Hyderabad as we march towards the 2012 ERDAS Desktop Products release.

Oh yes, and did I tell you that ERDAS IMAGINE functions are being placed into GeoMedia? Yes... Vector coming into ERDAS IMAGINE and raster id going into GeoMedia. 

Finally, I had wanted to do some webinars on the LiDAR tools, but have decided to wait until we complete the changes gathered from user feedback at Hexagon 2012. I expect to do the webinars before too late in July.

Wednesday, June 6, 2012

Intergraph Announces Point Cloud support Coming in ERDAS IMAGINE 2012

The announcement below is the fruit of software development work being done in Atlanta, Georgia USA; Pasadena, California USA; and Singapore as well as the press release written in Atlanta. The Intergraph corporate headquarters are in Huntsville.

Steve du Plessis and I will begin a series webinars on this topic later this month.

The reason we use the term Point Cloud rather than LiDAR, point cloud covers more data sources than LiDAR. LiDAR is but one data source.

The announcement is as follows: 

Intergraph® Highlights Advanced Point Cloud Capabilities at Hexagon 2012
New ERDAS IMAGINE® Features Enable You to Better Understand Your LiDAR Data

HUNTSVILLE, Ala., June 6, 2012 – At the Hexagon 2012 conference, Intergraph® is demonstrating a new technology that enables you to fully leverage the wealth of information in point cloud data. The centerpiece of the new functionality is the augmented viewing capability embedded in ERDAS IMAGINE®, which allows users to simultaneously view LiDAR spatially linked in 2D, 3D and as a user-specified profile that can reveal cross sections of the point cloud.

“Intergraph’s new functionality truly unlocks the potential of LiDAR, taking full advantage of its rich three-dimensional content,” said Steve du Plessis, Global Product Line Executive – Remote Sensing, Intergraph. “You can open up these dense point clouds, see inside them and break them apart to reveal the information they contain to gain a new perspective of the world.”

Within the three new views, ERDAS IMAGINE provides a range of options that make it easy to understand and extract the information contained in the point cloud. Different filtering options allow you to view the points in dozens of permutations. For example, you may only want to see first-return points classified as vegetation. To improve interpretability, flexible colorization settings offer the ability to view by elevation, return, classification, file, intensity or the RGB encoding of each point which provides a lifelike, three-dimensional color view of a point cloud. Additionally, you can view the point cloud draped over your 2D imagery, thereby obtaining a better understanding of the above-ground features portrayed by the points.

The profile view provides unique capabilities for analyzing point clouds. While the standard profile view is defined by a user-specified box drawn over the 2D view of the point cloud, you can also use a polyline vector to obtain a profile view of a longer strip within the cloud. ERDAS IMAGINE can then automatically roam at a user-specified speed along this long profile view. These features are especially useful in the utilities market, where they can be used to inspect power line corridors for encroaching vegetation or other line issues. As problems are visually detected, you can pause, make measurements, record the position and add notes. ERDAS IMAGINE can then export this information to a spreadsheet that can be given to field crews who maintain the corridors.

ERDAS IMAGINE’s new point cloud technology also offers a set of tools to classify, RGB encode, subset and filter your point data. The classification tool will enable you to specify criteria for automated point classification. Users can then leverage the editing tools within the 2D and profile views to refine the results. For example, you can delete, reclassify, bias or set a constant elevation on selected points with the security of multiple-step undo and redo capabilities. In addition to thorough views of the point clouds themselves, ERDAS IMAGINE provides comprehensive views and editing of the associated metadata. You can batch and distribute all these functions for faster processing, which is especially helpful since many users opt to create smaller tiles from large LAS files.

“Intergraph is in a unique position to provide the leading point cloud solution on the market,” continued du Plessis. “As part of Hexagon, our portfolio includes in-house airborne and terrestrial LiDAR hardware technology from our sister division Hexagon Geosystems, powerful GIS analytics from GeoMedia®, the world’s leading remote sensing capabilities in ERDAS IMAGINE and LiDAR data management and serving capabilities in ERDAS APOLLO. In addition, Intergraph’s vertical market approach offers extensive domain expertize in utilities, transportation, government and security. Together, not only will our point cloud technology strengthen any organization’s LiDAR-handling capabilities, but also its user-centric implementation streamlines workflows so users can quickly and easily obtain the most from point clouds and maximize the return on your LiDAR investment.”

To learn more about the Intergraph geospatial product portfolio, please visit

Friday, June 1, 2012

ERDAS IMAGINE 2012 is looking really good.

This release is going to be hot. What’s going to be in it?

The upgraded Spatial Modeler is more than an upgrade. This is more than an update of Model Maker that sits on top of the Spatial Modeler Language. All the Modeler Maker capabilities that Esri copied to create Model Builder are there, plus the things customers have demanded.
  • Faster file to file processing
  • On-the-fly raster processing (lessons learned from ER Mapper) 
  • On-the-fly vector processing (lessons learned from GeoMedia) 
  • Replacement of Model Maker User Interface 
  • Python Scripting 
But this is the beginning. Spatial Modeler is being designed with 64-bit processing as well as sharing tools with GeoMedia in mind. The architecture has been designed in a manner that will allow features and functions to be added in service packs.

The Python Scripting is very powerful. Customers will be able to connect other software applications with Spatial Modeler as part of their data processing work-flows. ERDAS IMAGINE's Spatial Modeler can act as a agnostic geoprocessing hub.

There is more….
  • Point Cloud (LAS, MrSID) visualization in 2D and 3D, profiles, measuring, point editing, RGB encoding… I am working with Steve du Plessis to put together a Webinar on point clouds.
  • 16-bit ECW, and ECW is faster… not just a little faster, but must faster and scalable.
  • Better and faster support for JPEG2000 data. 
  • LPS is fully Ribbonized.
  • New Segmentation algorithm that customers are saying  is the best they have seen to date. 
  • Improved ER Mapper Algorithm capabilities. 
  • GeoMedia Vector and GRID capabilities. The GeoMedia Buffer in Spatial Modeler is real time. You can test your buffer processes in a trial and error method in the viewer and then write to file (if you want to). 
  • Faster edge detection in MosaicPro.
  • New real-time spectral indices tool. 
  • Radar analyst work-flow tools in the ribbon, including ship tracking capabilities. 
There are a lot of moving pieces, and there are more items coming that cannot listed at this time. I have to wait a little longer on some.

Last year at Hexagon 2011 I discussed that the wall between raster and vector was coming down. It is more than that, mush more.... raster, vector, point clouds, imagery, photogrammetry, radargrammetry, and more all in one software.

I am again working on a research love of mine, change detection. We have the algorithms and processing performance we need to make high resolution change detection work for the whole geospatial community, not just the expert. This is expected post 2012. Exciting times are here.

Wednesday, May 9, 2012

Update on ERDAS ECW/JP2 SDK v5.0

As promised earlier.... here is an ERDAS ECW/JP2 SDK update

The ERDAS ECW/JP2 SDK (SDK) and ERDAS APOLLO Essentials Image Web Server (IWS) development teams in Perth, Western Australia are well along the way towards the latest version of the SDK, v5. The motivations behind this release of the SDK are increasing encoding performance, 16-bit ECW support, decreasing file size without lowering quality, expanding beyond Windows OS support (all without hurting decoding performance).

The best way to improve encoding performance is to more fully utilize system resources by expanding the use of multiple threads. This is being done.

In the v4.x SDK versions we see a 25MB/second encoding throughput of 3-band 8-bit image data. That is very fast for wavelet compression and we could have just sat on our hands and said, “We are the champions (my friend).”

But sitting back and wondering at our own accomplishments is not our way. Tests have confirmed SDK v5 is encoding 3-band 8-bit data between a 65 to 85MB/second. When encoding 3-band 16-bit data, encoding speeds close 25 MB/second are being observed. Final tuning will tell where we end up, but so far when using an 8-core CPU, encoding performance seems to be primarily limited by disk i/o.

To add support for 16-bit, we need to change the ECW file format. So, at the same time the Perth team decided to do some other things to the ECW file format. File format changes have provided a decrease in file size when using the previous target compression ratios. In short, smaller files with the same image quality (the preverbal ..yet true.. win-win scenario).

The work is being done to add support for Linux, with a foundation for the SDK to be ported to other OS versions as demand requires.

Also, metadata support in the new ECW v3 file format has been expanded. We have added better projection definition options, imbedded statistics (min, max, mean, median, mode, etc.), and much more.

Also provided when using the SDK to stream in ECW and JP2 encoded image data via ECWP from IWS, faster ECWP streaming. (BTW, in case you didn't know, ECWP performance leaves JPIP performance in the dust.)

There is one thing to note, the ECW v3 file format cannot be read by earlier ECW SDKs. The old v2 ECW file format could not be expanded to allow the file format improvements needed, so we created ECW v3. Of important note, the desktop read-only SDK will continue to be made available at no charge.

What is the ‘un-official’ time-table? Expect to see ERDAS ECW/JP2 SDK v5 in the ERDAS IMAGINE / LPS, ERDAS ER Mapper,  ERDAS APOLLO Essentials (aka IWS), and GDAL after the northern hemisphere’s autumnal equinox, and an upgrade of ERDAS' ECW for ArcGIS Server made available a few months following following the ERDAS Desktop releases.

Keep an eye on the ECW Web Page.

Tuesday, May 1, 2012

ERDAS Desktop 2011 Version 11.0.5 now available

ERDAS Desktop 2011, Version 11.0.5 is now available. This version fixes both internal and customer reported issues and contains enhancements as well. One installer will install ERDAS IMAGINE, ERDAS ER Mapper and LPS. It is available to customer with and without support.

There is no need to install the previous 11.0.1, 11.0.2, 11.0.3 and 11.0.4 versions as 11.0.5 contains all the items in those versions.

ERDAS 2012 is due later this year, so 11.0.5 may be the final 2011 version.

To download, see: ERDAS IMAGINE Product Download Page

Thursday, April 5, 2012

ERDAS Desktop 2011 Version 11.0.5 almost ready

The ERDAS desktop teams are completing the next version of ERDAS IMAGINE, LPS and ERDAS ER Mapper, ERDAS Desktop 2011 Version 11.0.5. This version will be mostly a maintenance release of customer needed items, but there will be some customer requested enhancements as well. This version will likely be the final for ERDAS Desktop 2011, as we are well into the development of ERDAS Desktop 2012.

Before late April 2012, visit the ERDAS IMAGINE, LPS or ERDAS ER Mapper product download page at to get your copy of this new version. Your ERDAS Desktop 2011 licenses will work with v11.0.5.

Thursday, February 16, 2012

ERDAS ER Mapper Development Team

Intergraph has stood up an ERDAS ER Mapper Development Team in India in the city of Hyderabad state of Andhra Pradesh. I visited the development team last week. The Hyderabad development development team is working with the Norcross, Georgia development teams to deliver remote sensing software products to the GIS community. I do believe good things will come from the Hyderabad  ERDAS ER Mapper Development Team in the ERDAS Desktop 2012 release, as well as into the future.

BTW, Intergraph was one of the first software groups to open an office in India, beginning way back in 1987. Now you see Amazon, ADP, Facebook, Google, Microsoft, and more are in Hyderabad. See:

Monday, January 23, 2012

Hexagon merges ERDAS into Intergraph

There seems to be a little confusion in the market-place concerning Intergraph and ERDAS.  At the end of July 2011, Hexagon merged ERDAS into Intergraph. All ERDAS employees, products and services are now Intergraph Employees developing new products based on ERDAS products, as well as supporting ERDAS customers using those products.

Customers of ERDAS products will continue to communicate with ERDAS employees and use The transition should be transparent to customers, as legal and back-office changes have formed the majority of the changes.

You will still find Brad Skelton, Ian Anderson, Jeff Dooley, Mladen Stojic, myself, and many more Erdasians in the Norcross (metro Atlanta) office. You will find Laura Ramage, Michelle and Joe Mostowy, Andy Garratt, Dean McCormick, Mike Lane, and many other Erdasians scattered around the globe. As well, some of the old-time ER Mapper employees are still around, such as Mark Sheridan, Chris Ribble, and more. You will find Ionic employees such as Niko Vanraes, Dimitri Monie, and others still hard at work as well.

In my opinion, the change can be considered somewhat like Coca-Cola purchasing Dr. Pepper... there was no change in the wildly successful Dr. Pepper formula.

Where will this fusion take us, while I cannot say exactly what is happening on my blog, I can say that my earlier post of  "Hexagon 2011; the razing of the wall" and what we started with in ERDAS IMAGINE 2011, V11.0.4 should be strong clues.

Tuesday, January 3, 2012

ERDAS Desktop 2011, Version 11.0.4 Released

Intergraph has released ERDAS Desktop 2011, Version 11.0.4. The new version is available for download from the ERDAS IMAGINE, ERDAS ER Mapper and LPS product download webpages.

This is more than a maintenance release. There are enhancements, as well as Intergraph has added new capabilities to strengthen the ties between the GeoMedia and ERDAS Desktop products. Expect more announcements by Intergraph later this year (2012) of tighter integration between the GeoMedia and ERDAS Desktop product lines.

The ERDAS Desktop 11.0.4 Version Description is located here:

Until more announcements are made, please refer to Field Guide posts: