Author: Guss Wright
Hello. My name is Chief Warrant Officer 3 Augustus Wright. I’ve been working with Harris since August as part of the U.S. Army’s Training with Industry program. Now why on earth would the Army send one of its warrant officers to live what many soldiers would call “The Good Life”? That entails working with and training alongside civilians -- not to mention living as one -- for a year!
The answer is actually quite simple. The Training with Industry program recognizes that it is necessary for selected individuals to actually leave the Army for a while to get hands-on training with the tools needed to do their jobs. Afterward, these industry trainees return with refined and improved skills, as well as knowledge about leading trade practices to move the Army forward. Given my background, I was a good candidate. I’m a 125D Geospatial Engineering Technician and I’ve been serving on active duty for 17 years. Prior to being selected to train with Harris on its geospatial software ENVI, I served as the direct support officer-in-charge for the U.S. Army Europe’s 60th Geospatial Planning Cell.
Over the years, I’ve completed three separate combat deployments to Iraq in support of Operation Iraqi Freedom and Operation New Dawn totaling 38 months. In 2003, as a Topographic Sergeant with the 4th Infantry Division I provided situational awareness and geospatial tactical decision aids to ground force commanders in Iraq and was awarded a Bronze Star Medal. As the Sr. Topographic Sergeant for the 25th Infantry Division in 2005, I used ENVI and ArcGIS to perform predictive analysis on IED trends. ENVI was primarily used to detect disturbed earth and spectral differences in road materials. In 2009, as the GEOINT Officer in Charge for the3rd Infantry Division, I used ENVI and ArcGIS to detect IED making materials and disturbed ground. I also used ENVI and ArcGIS to plan the emplacement of strategic checkpoints that were a part of the strategy to end operations in Iraq. During this tour I was awarded a second Bronze Star Medal.
Since I started with Harris three months ago, I’ve worked side by side with expert software engineers. Not only am I learning a ton, but I’m also in the unique position to provide user feedback from the Army geospatial community about what additions to the software would be helpful. One intuitive add-on that I’m extremely excited about is the dynamic symbolization of the electromagnetic spectrum in the ENVI DataManger, Layer Manager, Metadata and Spectral profile. Another is Supplementary Color Band Symbology that takes into account the wide range of remote sensing and imagery exploitation experience of users within the Defense & Intelligence Community. This new feature, which is applicable in both academia and day-to-day tradecraft, will help more experienced soldiers teach novice soldiers while ensuring product quality and integrity in practical use cases. Keep an eye out for these enhancements in future releases of ENVI.
Supplementary Color Band Symbology
Dynamic symbolization of the electromagnetic spectrum in the ENVI Data Manger, Layer Manager, Metadata and Spectral profile.
I’ll be posting blogs on a regular basis to keep you up to date on what I’m learning and let you know how this new “deployment” as an Industry Trainee with Harris is going.
Categories: ENVI Blog | Imagery Speaks
Author: Hans Augustus
When you’re in the foothills or mountains of Afghanistan or Iraq a statement like “identify targets and threats to get the actionable intelligence you need” is appealing. And while “improve your situational awareness with ENVI”is a marketing claim, when you’re in theater, using the technology saves lives.
Just a few years ago, I was one of the many individuals that were prevented from executing “Out-side the Wire missions”. What that means is because there were identified threats that would have jeopardized my life and the lives of those I was charged to protect, we had to get creative about getting the job at hand done. Our mission was to enable the rebuilding of a country without engaging the enemy. ENVI not only provided a way to ensure mission effectiveness, but helped us successfully navigate through hostile territory with a high level of confidence that we would not be engaged by enemy forces.
Today I support Harris to expand ENVI's capability and bring it to a wider market. As a Contracts professional, my organization ensures the safe deployment of the ENVI product which means following and enforcing government and commercial regulations that allow Harris to provide civil, commercial as well as military use. ENVI software solutions are widely being used to: quickly identify mineral deposits using hyperspectral imagery; develop models to improve predictability of weather patterns; and, improve situational awareness about unknown terrain before a mission even gets started. As these capabilities are incorporated into an organization, they must be vetted to ensure they are legally integrated into existing software and appropriately licensed to various users in accordance with their needs. Contract Managers are at the front line of ensuring success in getting the product to market and getting it into the hands that need it.
It’s been several years since I have been a customer of the robust capabilities offered by ENVI. There are many new threats facing our military members. ENVI continues to rise to meet those challenges by providing an even higher level of confidence today. ENVI will help today’s war-fighters successfully complete their missions and return safely. Better yet, ENVI can thwart disaster by providing the GIS that identifies a high-risk threat that jeopardizes the lives of our military members before they get into harm’s way. As a contract professional ensuring the success of ENVI is more than making a sale, it’s a new mission.
Author: Patrick Collins
Airbus has recently released their WorldDEM™ product, which is a worldwide elevation product that has a better resolution than freely available elevation sources such as SRTM and GMTED. Accurate elevation information is important to conducting accurate analysis, because low resolution data can cause noise to appear especially when comparing datasets from two different time periods.
Below is an Airbus image taken after a landslide that occurred in Malin, India in 2014. I was able to retrieve a WorldDEM dataset over the area taken not too long before this landslide occurred.
I also ordered two of the post-event images in order to extract a passive point cloud from the two datasets, which would allow me to extract extremely detailed elevation information over the area. This is important because it can be difficult to get information over a remote area that has experienced a disaster or inclement weather. Being able to extract one-to-one elevation information from a high resolution imaging satellite such as Pleiades makes it easy to get accurate information in this type of scenario. Here you can see the extracted point cloud.
After extracting the Digital Elevation Model from the point cloud, I then subtracted the point-cloud generated DEM from the WorldDEM product to get an elevation change difference over the area. You can see from the image below that the resolution difference between the two elevation datasets has caused some anomalies to appear between the pixel sizes of the lower resolution dataset.
When we look at a color slice of this data we get the following.
This doesn’t quite give us the accuracy we are looking for with our data, however we can use convolution filtering within ENVI to smooth this dataset into something a little more readable. After running a convolution filter with a thirty-five pixel size we are left with the following image.
This final image shows how our elevation analysis has clearly captured the location of the landslide within this region. As mentioned before, the ability to capture and analyze extremely accurate elevation data from space allows government and other first responders the ability to quickly get important information about an area, even one that is remote or inaccessible.
For more information on WorldDEM and how high resolution elevation information can increase the accuracy of your analysis results, check out the joint Airbus / Harris webinar on the topic from last week located here.
Tags: ENVI, Cloud, ENVI LiDAR, Airbus, Harris, passive, WorldDEM, point, change, detection
Author: Adam O'Connor
One of the most exciting new features in the upcoming ENVI 5.3 Service Pack 1 release is an implementation of the popular Fmask (Function of mask) algorithm that provides automated cloud and cloud shadow detection in multispectral images. The initial focus of the ENVI implementation is on the generation of a cloud mask raster that can be used in subsequent image processing analysis to mask-out all cloud+shadow pixels. Furthermore, the ability to invert the mask in tools such as the Classification Workflow will allow users who are actually interested in analyzing the clouds to mask-out all non-cloud pixels.
The Fmask (3.2) algorithm will be exposed in both a new "Calculate Cloud Mask Using Fmask Algorithm" desktop application tool and associated "ENVICalculateCloudMaskUsingFmaskTask" routine in the programmatic API. Both the GUI tool and API task will create a cloud mask for Landsat 4-5 TM, Landsat 7 ETM+, Landsat 8 OLI/TIRS and NPP VIIRS M-Band datasets (we plan to expand support to include Sentinel-2 in a future release). This tool/task requires the following inputs:
- An image containing multispectral bands calibrated to top-of-atmosphere (TOA) reflectance
- A thermal-band image calibrated to brightness temperatures (in Kelvins)
- A cirrus-band image calibrated to TOA reflectance (applicable to Landsat 8 only)
Here is an example input Landsat 7 ETM+ scene with what I call popcorn clouds that has been calibrated to top-of-atmosphere (TOA) reflectance using the "Radiometric Calibration" tool in ENVI:
Author: Jason Wolfe
Working with the ENVI software engineering group gives me an opportunity to test new and enhanced features from the perspective of an end user. ENVI 5.3, Service Pack 1 will be released soon, and it will include some improvements to the Seamless Mosaic tool (and associated API) to better support Landsat imagery. I tried it myself by creating a mosaic using a coastline mask and by reducing the amount of visible cloud cover. I think the result turned out great. In this article I will share some tips that will help you to create a high-quality image mosaic.
The above image is a true-color mosaic of Sicily comprised of three different Landsat 8 scenes. I chose this particular location because it has overlapping flight paths from different dates. The following figure shows the source images displayed in ENVI:
Landsat 8 scenes are available for download from the U.S. Geological Survey EarthExplorer web site. Here are some tips for finding images that will produce the best visual quality:
The next steps involve a little preparation to get the images ready for a mosaic.
The scenario that I’m describing here is producing a true-color mosaic for visual analysis. If you are creating a mosaic for spectral analysis, you should first calibrate the source images to spectral radiance and correct for atmospheric effects. You do not want to perform any color correction or feathering in the Seamless Mosaic tool since it would compromise the spectral integrity of the data.
Ocean pixels can skew the statistics used for effective color balancing among the scenes. To begin the process of excluding ocean pixels, I downloaded the GSHHG high-resolution global coastline dataset (Wessel and Smith, 1996). I used ENVI to extract the vector record of just the Sicily coastline, then saved that vector record to a shapefile. I used the coastline boundary as a mask for each source image such that ocean pixels were labeled ‘NoData’. You could use this same process to confine your mosaic to an administrative or watershed boundary.
I saved each masked image to a new ENVI-format file that only included the visible bands. This would reduce the processing time in creating the mosaic.
Once I imported the masked scenes into the Seamless Mosaic tool, I experimented with various display options (while previewing the results) to achieve the best visual quality. I knew that I could stack the scenes where they overlap so that a better scene is on top. The following figure shows an example where I reordered Scene 2 to the top of the display stack to hide unwanted cloud cover.
I experimented with some color-matching options and found that computing histogram statistics based on the overlapping areas (versus the entire image) more effectively minimized the tonal and color differences between scenes.
At the boundaries where the scenes meet, I chose to automatically generate seamlines with a feathering distance of 50 pixels on both sides of the seamlines. This eliminates any hard edges between scenes and more effectively blends the images where they meet.
Finally, I saved the mosaic to GeoTIFF format. The process only took a few minutes to complete. To learn more about creating mosaics, please visit the Seamless Mosaic topic in our online Documentation Center.
Wessel, P., and W. H. F. Smith. “A Global Self-consistent, Hierarchical, High-resolution Shoreline Database.” Journal of Geophysical Research 101 (1996): 8741-8743.